Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (YSS, RDW, BKSY, SPIR, LUNR, VOYG, VSAT, ONDS, UMAC) which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Basket Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Space investors essentially are just multiplying large numbers by distant dates.

We would rather start with a procurement schedule:

• The Space Development Agency’s demonstration tranche called for 28 satellites.

• GAO puts Tranche 1 at 165 and Tranche 2 at 264.

• Since December, SDA has separately ordered 108 Tranche 3 Tracking satellites for about $5.25 billion.

This schedule is proof that a meaningful part of the architecture is now being bought in lots. Who actually makes money from this thematic is now more clear.

The old model rewarded years of engineering around one exquisite asset. The new one should reward platform reuse, supplier throughput and, eventually, the services that sit on top of a larger installed constellation.

However, government procurement can turn a sound industrial thesis into a bad quarter without changing the strategic demand at all. We are expressing the shift through an equal weight basket not because the businesses are equally good, but because the eventual profit pool is still unsettled/unclear.

Basket Overview

This section will be pretty brief as they are thematically linked, and can be further mapped to different layers within the thematic:

If the Pentagon is moving toward a larger, more frequently refreshed system, the margin does not have to stay with the company that bolts the actual satellite together.

A supplier can out-earn the prime, and a data business can earn better returns on the constellation than the company that built it. The next few quarters should tell us where the economics are settling.

The Setup

The setup is essentially the new cadence around an old mission. Missile warning, tactical communications and persistent intelligence have been strategic for decades but the willingness to distribute more of those functions across larger constellations and refresh the technology is what is new.

GAO says SDA plans tranche awards roughly every 2 years and expects each tranche to require replacement around 5 years after launch. SDA’s own Tranche 3 release describes targeted technology refreshes at roughly 2 year intervals. This is a very different demand pattern from a bespoke satellite designed once, nursed through a 10+ year program and then replaced by another bespoke satellite.

However, SDA has kept awarding new tranches while some technical elements and architecture-level schedules remain immature. Contractors have absorbed unplanned work. In other words, this is not your typical mass production scenario. It is an attempt to impose industrial cadence on a technically difficult system while the requirements are still moving.

This tension is good for public equities in the space though. Strategic urgency supports demand and capacity investment. Immature procurement and engineering make the path from award to revenue unusually noisy.

It also raises the value of common buses, repeatable payload integration, qualified components, automated testing, ground systems and data products that can be sold again after the satellite is already up. Conversely, it raises the cost of being late: a supplier that misses one tranche may wait for the next cycle while the factory it built sits underutilized.

It’s worth noting the shift toward repeatable hardware is not unique to space. The same procurement logic is already visible lower where the Pentagon is pushing inexpensive autonomous systems and domestic drone-component capacity:

• Unusual Machines (UMAC) is scaling U.S. motor production from tens of thousands toward roughly 100,000 units per month.

• While Ondas (ONDS) is assembling a broader portfolio around autonomous ISR, counter-UAS and precision strike.

• They are not space exposures, but they are useful evidence that Washington increasingly values production capacity and replenishment alongside technical sophistication.

Causal Mechanism

Volume alone is not enough. The thesis only works if repeat production actually lowers engineering intensity, improves utilization and, eventually, produces cash. So far, the financials are mixed:

• York and Redwire are showing signs that manufacturing economics can improve, but both still illustrate how procurement timing and working capital can keep cash conversion weak.

• BlackSky is beginning to show better service-layer economics, while Spire remains earlier in that transition.

• Intuitive Machines and Voyager have substantial demand visibility, but both are still consuming significant cash; Voyager also carries the added burden of Starlab investment.

• Viasat offers a useful picture of what a more mature version of the model can look like: less exciting growth, but clearer funded demand and actual free cash flow.

Awarded vehicles, options, unfunded scope and internal pipelines are often presented alongside firm work even though they carry very different levels of certainty. York, LUNR and Viasat all make that distinction visible in different ways: there is a meaningful gap between what has been identified, what sits on an awarded contract vehicle, what is included in reported backlog and what has actually been funded. The issue is that they should not be underwritten like committed revenue.

The investment case therefore comes down to a fairly mundane sequence: repeat orders, less engineering per unit, better gross margin, manageable working capital, then free cash flow per share. Any company can look good at step 1. The basket becomes interesting when 2 or 3 names start reaching step five without another financing in between.

On current evidence, RDW and BKSY have the cleanest operating proof, albeit for different reasons. RDW is showing industrial margin improvement across a broader defense/space portfolio; BKSY is showing that the data layer can scale at a higher incremental margin.

YSS remains the purest test of the factory thesis, but the 2026 guidance reset is exactly why purity and risk/reward are not the same thing. VOYG stays in the basket because propulsion, energetics and mission infrastructure are directly relevant to the capacity build.

Timeline

The next 12 to 24 months need evidence that the orders already being discussed can be built, billed and collected.

Late 2026 / 2027:

• BKSY expects two Gen-3 satellites in Q3

• YSS has Nemesis targeted for Q4

• Spire expects sequential cash-use improvement.

• In 2027 York task-order conversion, Redwire margin durability, Voyager integration and Viasat's mid-teens DAT growth target.

2028-2029:

• SDA says the 36 accelerated Tranche 3 Tracking satellites should be launch-ready by end-2028

• The 72 standard Tranche 3 Tracking satellites are scheduled for FY2029. By then the market should know whether frequent replenishment is an operating reality or just a planning convention.

Key Risks

• Procurement timing: Continuing resolutions, task-order timing and protests can create the same mismatch elsewhere.

• Engineering and fixed-price risk: SDA is deliberately pushing cadence before every technical element is mature. That can be good for volume and bad for contractor economics if the same “standard” platform keeps absorbing redesign work.

• Backlog quality: The sector rewards big contract numbers. It should reward funded, executable work. Footnotes on all these companies should be carefully examined.

• Capital intensity and dilution: YSS, BKSY, LUNR and others have recently relied on equity or IPO proceeds while building capacity. A successful program can still be a poor stock if shareholders repeatedly finance the working-capital gap.

• Incumbents: Lockheed, Northrop, L3Harris and other traditional primes are already winning proliferated contracts. The architecture can change without the profit pool migrating entirely to newer public companies.

• Correlation: In a procurement slowdown or risk-off tape, the basket will trade like one small/mid-cap defense-space factor long.

Conclusion

The public space trade used to mean giant defense primes or companies selling a very distant future. The current basket is more interesting because the government is already buying the architecture in repeatable generations, and public investors can own several points in the chain that supplies it.

The most important lesson from the latest quarters is that production cadence and shareholder cash flow are moving at different speeds:

• YSS can put 21 satellites on orbit and still cut annual guidance.

• RDW can post record margin and backlog while using free cash.

• BKSY can show service leverage and still sell $150 million of stock.

• LUNR can report $1.8 billion of backlog with parts of that backlog much less funded than the headline suggests.

If repeat awards begin to convert with less engineering intensity, better utilization and improving free cash flow per share, this stops being a collection of space stories and starts looking like an industrial category. But if the companies stay dependent on equity markets to bridge every new tranche, the Pentagon may still get its factory, but shareholders will have paid for it.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (AM US) which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Antero Midstream (AM US, $10.2b market cap) is usually reduced to a simple label: a single-customer Appalachian midstream company tied to Antero Resources.

AM’s dependence on Antero Resources is the central risk in the story, but it also creates an unusual operating advantage:

• Substantially all of AR’s current and future Appalachian acreage is dedicated to AM for gathering and compression, and the two companies coordinate development and infrastructure planning.

• AM is therefore not building speculative pipe and waiting for third-party volumes. It is building against a visible drilling schedule, on acreage controlled by the producer responsible for almost all of its throughput.

In addition, the asset base has changed:

• AM closed the HG Midstream acquisition in February 2026, adding gathering and integrated water infrastructure in the core Marcellus, while divesting non-core Utica assets later that month. The company traded geographic breadth for density around AR’s highest-priority inventory.

• The first full quarter of the new footprint produced record gathering volumes of 4.124 Bcf/d, up 19% year over year, and compression volumes of 4.036 Bcf/d, up 17%.

• The quarter was not a clean operating-leverage print. Adjusted EBITDA rose only 2%, fresh-water delivery declined 16%, high-pressure gathering declined 7%, and acquired water volumes carried a different cost profile.

• However our thesis is that AM can integrate the new footprint, shift activity toward higher-value services, and preserve free cash flow after the dividend while funding a new regional connectivity layer.

East Side Express is a strategic bridge:

• AM has begun construction on its first intrastate regional pipeline, a bi-directional system intended to connect dry-gas supply with multiple long-haul and regional pipelines and local markets. Management presents the project as the infrastructure companion to roughly 700 future dry-gas locations and early well results that are materially better than legacy offsets.

• If the project works, AM moves one step beyond captive gathering: it remains anchored by AR, but gains more routing flexibility and a longer runway for dry-gas development.

The balance sheet is no longer the gating issue it was:

• AM generated $80 million of adjusted free cash flow after dividends in Q2, its twelfth consecutive positive quarter on that measure. In July, it received approximately $371 million from the Veolia litigation and used the proceeds, together with revolver borrowings, to call the $650 million notes due in 2028. Pro forma leverage fell to 2.8x, liquidity exceeded $600 million, and the nearest maturity moved to 2029.

The variant view is straightforward: What looks like customer concentration can also function as infrastructure coordination. AM does not need AR to pursue unconstrained growth. It needs AR to keep drilling the dedicated core, the HG integration to convert into better water and compression economics, and East Side Express to turn dry-gas inventory into regional connectivity. If those pieces align, AM becomes less a static gathering utility and more a toll road built directly into its customer’s development plan.

Company Overview

Antero Midstream operates the infrastructure between the wellhead and the downstream market. Its system includes low-pressure gathering, well-pad and centralized compression, high-pressure gathering, water delivery and handling, and equity interests in processing and fractionation facilities. The company reports two operating segments: Gathering and Processing, and Water Handling.

The physical footprint is substantial but concentrated. AM reports 731 miles of gathering pipeline, 4.8 Bcf/d of compression capacity and 1.6 Bcf/d of joint-venture processing capacity. In Q2 2026, the processing joint venture also ran 40 MBbl/d of fractionation volume, with both processing and fractionation capacity fully utilized.

AM earns per-unit fees on gathering, compression and water services rather than taking direct ownership of the commodity. Most relevant gathering and water fees include annual CPI adjustments. Certain new gathering and compression assets requested by AR can be supported either by minimum-volume commitments or by a cost-of-service mechanism. Long-term acreage dedications cover substantially all of AR’s current and future Appalachian acreage.

This creates a business with limited direct commodity-price exposure but meaningful indirect exposure. Lower natural-gas or NGL prices do not mechanically reduce AM’s fee per unit. They can still influence AR’s drilling schedule, completion timing and production trajectory.

AM’s economic engine is therefore best understood as volume × service intensity × contract structure, not as a direct bet on the spot price of natural gas. The company’s most important asset is the development map it shares with AR.

The Setup

The setup begins with a portfolio rotation:

• AM acquired HG Midstream on February 3, 2026. The assets sit in the core Marcellus in West Virginia and include gathering pipelines and integrated water infrastructure. From closing through June 30, the acquired operation contributed $61.2 million of revenue and $12.4 million of net income, including transaction expense. AM and AR are also modifying their commercial arrangements to incorporate well-pad compression and water services associated with the acquired acreage.

• Twenty days later, AM closed the sale of substantially all of its Utica midstream assets. The disposed footprint included 118 miles of gathering pipelines, 0.7 Bcfe/d of compression capacity, 85 miles of water pipelines and 12 water impoundments. The proceeds were used to reduce debt.

AM exited a separate Ohio footprint and added infrastructure embedded in AR’s core West Virginia development area. The result is a more concentrated system. In midstream, density can improve planning, increase asset reuse, and make incremental projects easier to sequence around a known drilling program. The potential advantage is that every dollar of infrastructure capital is aimed at the customer most likely to use it.

The Q2 operating data show the first-order effect: Gathering volume rose 19% and compression volume rose 17%. The HG acquisition and 80 connected wells drove the increase, partially offset by the Utica sale and natural declines. The mix underneath the headline:

• Well-pad compression, a new reported category following the HG acquisition, averaged 747 MMcf/d.

• Centralized compression declined 5%.

• High-pressure gathering declined 7%, primarily reflecting the Utica divestiture and natural decline.

• Fresh-water delivery declined 16% because of completion timing and location.

• Other water handling increased 131%, largely because water delivered on the HG acreage is currently charged at cost plus 3%, alongside higher blending, trucking and disposal activity.

This is why volume growth and EBITDA growth diverged. Some acquired throughput arrived with additional operating expense. Some high-pressure activity left with the Utica assets. Water activity shifted from the legacy fixed-fee delivery category into cost-plus and cost-of-service categories. Direct operating expense increased 34% year over year, while adjusted EBITDA increased 2%.

AM now has more throughput, but management must prove that integration, commercial restructuring and service layering can improve the earnings content of that throughput.

The sponsor backdrop is supportive. AR produced a record 4.144 Bcfe/d in Q2, up 21% year over year, raised full-year production guidance to 4.15–4.2 Bcfe/d, and guided to higher sequential production in the second half. More production and more connected wells create more opportunities to gather, compress and handle water across the dedicated footprint.

The setup therefore has three legs:

1. A larger core footprint: HG adds immediate gathering and compression throughput plus future water integration.

2. A cleaner capital structure: the Utica sale and Veolia proceeds reduced debt and pushed out the nearest maturity.

3. A new connectivity layer: East Side Express is intended to connect dry-gas supply with multiple regional and long-haul outlets.

Causal Mechanism

1. Customer concentration becomes planning visibility

Almost all of AM’s revenue comes from AR, but the concentration also creates unusual visibility. AR has dedicated substantially all of its current and future Appalachian acreage to AM, while AM’s capital program is built around AR’s drilling schedule.

That alignment enables a just-in-time infrastructure model. AM can see where wells are going, when they will be completed and what gathering, compression and water capacity they will require before deploying capital, rather than building ahead of uncertain third-party demand.

The causal chain is simple:

AR selects the next core development area → AM builds the required infrastructure → wells connect to the dedicated system → fixed fees convert throughput into recurring cash flow → new production data informs the next capital decision.

2. Contract structure converts commodity risk into activity risk

AM’s contracts are commodity-insulated, not commodity-independent.

AM is paid on volumes handled, not the market value of the molecules. Most gathering, compression and legacy water fees carry CPI escalators, while certain new high-pressure projects can receive either 10-year minimum-volume commitments or cost-of-service pricing targeting a 13% return over seven years.

But the protection is incomplete. Low-pressure gathering and water pipelines generally lack minimum commitments.

The fixed-fee model protects the unit rate; it does not guarantee the unit count.

AM is therefore best understood as a derivative of AR’s drilling economics, with contractual dampeners between commodity prices and cash flow.

3. Core density increases the value of service layering

The HG assets add more than gathering volume. They add well-pad compression and a broader water footprint around AR’s core acreage. AM plans to connect the acquired system with its legacy gathering and water network, with management expecting the water contribution to become more visible in 2027.

The opportunity is service layering. The same acreage can generate gathering, compression and water fees across the life of a well, increasing revenue density without requiring a new customer.

Q2 showed both the upside and the complication. Well-pad compression scaled quickly, while fresh-water delivery declined. Other water handling surged, but much of it was lower-quality cost-plus activity burdened by trucking, disposal and blending expense.

The integration thesis is not simply more water volume. It is a better mix of pipeline-based services with stronger margin conversion.

4. East Side Express can turn captive infrastructure into regional infrastructure

East Side Express is AM’s first intrastate regional pipeline and arguably its most important organic project. The system will run east-west, expanding dry-gas access to multiple long-haul and regional pipelines.

The value is better optionality at the basin exit. Gathering assets are only as valuable as their outlets; constrained takeaway or weak local pricing can impair otherwise attractive wells. More routing flexibility increases the number of markets accessible from the same dedicated inventory.

AM ties East Side Express to roughly 700 future dry-gas locations. Its Q2 presentation also showed a 2026 dry-gas pad producing 67% more cumulative gas over its first 90 days than two 2012 offsets.

One pad does not prove a basin-wide productivity shift. But better well performance plus better connectivity is the mechanism that can extend the dry-gas development runway.

The feedback loop is potentially powerful:

Better wells support drilling → drilling fills gathering and compression assets → East Side Express creates additional market access → better market access supports development economics → AM earns fees across a longer inventory runway.

This is not guaranteed. AM has not disclosed in the public materials reviewed a final project capacity, total capital requirement or definitive commissioning date for East Side Express.

5. Balance-sheet capacity reduces timing risk

Infrastructure projects fail as often from financing constraints as from poor geology. AM’s balance-sheet reset changes the probability distribution.

The company received approximately $371 million in damages and interest from Veolia in July 2026. It used those proceeds and revolver borrowings to call the $650 million senior notes due in 2028 at par. On a pro forma basis, leverage fell to 2.8x, liquidity exceeded $600 million, and the nearest maturity moved to 2029.

The strategic benefit is durable: AM no longer needs East Side Express or the HG water integration to pay back immediately in order to avoid a near-term maturity problem. It can sequence the projects around operations rather than around a refinancing deadline.

The company also generated $80 million of adjusted free cash flow after dividends in Q2, its twelfth consecutive positive quarter. Growth is useful only if AM can fund it without giving back the balance-sheet improvement.

Timeline

The key events are no longer hypothetical. AM has closed the transactions, received the litigation proceeds, retired the 2028 notes and begun East Side Express construction. The remaining questions concern integration and operating conversion.

Second half of 2026 — Volume conversion test

Management expects higher gathering and water volumes to drive EBITDA growth in the back half. AR has also guided to higher sequential production. The key test is whether AM’s earnings growth begins to converge toward its volume growth as the acquired system is integrated.

2027 — Water integration becomes measurable

Management expects the water projects associated with the new footprint to help drive high-single-digit EBITDA growth in 2027. The quality of that growth matters more than the headline: pipeline-based service growth and margin conversion would be more important than cost-plus volume alone.

Key Risks

1. AR is the customer, counterparty and operating transmission mechanism

Substantially all of AM’s revenue is derived from AR. Any deterioration in AR’s production, drilling schedule, liquidity, leverage or operating performance can flow directly into AM’s throughput and cash generation. AR also owned 29% of AM at year-end 2025, and the companies share certain officers, directors and service arrangements.

2. Fixed fees do not eliminate commodity exposure

AM has little direct exposure to commodity prices, but AR does not. A prolonged deterioration in natural-gas or NGL economics could cause AR to reduce activity. Because low-pressure gathering and fresh-water pipelines generally lack minimum-volume commitments, lower drilling and completion activity would affect AM even if contracted fees remain unchanged.

3. Q2 volume growth was lower quality than the headline suggests

Gathering and compression volumes rose sharply, but adjusted EBITDA grew only 2%. Direct operating expense increased 34%. Other water handling grew 131%, but much of the increase came through cost-plus HG water, blending, trucking and disposal. AM must demonstrate that the acquired volume can be integrated into a more attractive service mix rather than simply adding revenue and expense in parallel.

4. East Side Express is still an under-specified project

The strategic logic is clear, but key project economics and operating milestones remain undisclosed in the public materials reviewed. Capacity, total capital, final in-service timing and firm commitments will determine whether East Side Express is a high-return extension of the existing system or a longer-dated capital project with uncertain utilization.

5. Processing capacity is full

AM’s processing and fractionation joint venture operated at 100% utilization in Q2. Full utilization supports current earnings but leaves limited spare capacity for additional rich-gas volumes.

6. Water remains completion-dependent

Fresh-water delivery declined 16% in Q2 because of the timing and location of AR’s completions. Water activity can move sharply between quarters, and the mix between fixed-fee delivery, cost-plus services, blending, trucking and disposal changes the margin outcome.

7. The balance-sheet improvement could be re-spent

The Veolia proceeds removed the nearest maturity and reduced leverage, but management now has greater capacity for growth projects, repurchases or additional transactions. A new acquisition cycle or a materially larger East Side Express budget could reverse the risk reduction before the operating benefits are proven.

Conclusion

The bear case sees one customer, one basin and a business whose activity ultimately depends on commodity economics. All three points are true. The bull case is not that these risks disappear. It is that the same concentration produces a coordinated infrastructure model with unusually high visibility into where, when and how the next wells will be developed.

AM entered 2026 with a gathering-and-water footprint built around AR’s legacy acreage. It now has a larger core Marcellus system, well-pad compression, integrated HG water assets and a first regional dry-gas connectivity project. The company has also removed its nearest maturity and preserved free cash flow after dividends through the transition.

The next phase is about conversion:

• Can record gathering and compression volumes generate faster EBITDA growth as the acquired assets normalize?

• Can the water business shift from activity growth to margin growth?

• Can East Side Express turn dedicated dry-gas inventory into a regional connectivity advantage?

• Can AM fund the buildout while preserving its post-dividend cash surplus and lower leverage?

If the answers are yes, the market will have to stop treating AM as a static captive gathering system. The company would still be tied to AR, but the tie would increasingly resemble a long-duration infrastructure schedule rather than a simple customer dependency. AM will remain concentrated. The question is whether that concentration continues to buy it better visibility, denser infrastructure and higher returns on the next dollar of capital.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (ECVT US) which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Ecovyst (ECVT US, $1.4b mkt cap) entered 2026 as a focused sulfur-services company. The Advanced Materials & Catalysts business was sold on December 31, 2025, and $465 million of the proceeds was used to repay term debt. Continuing operations now center on sulfuric acid regeneration, virgin sulfuric acid, industrial treatment services and catalyst activation. The June 30 acquisition of Calabrian added sulfur dioxide and related derivatives.

The core refinery service is a closed loop:

• Sulfuric acid is used in alkylation units to produce alkylate, a high-octane gasoline component. The acid loses strength during the process.

• Ecovyst collects the spent material, transports it to a regeneration plant, restores it and sends fresh acid back to the refinery.

• Chemical production is only one part of the service. Logistics, plant reliability, customer integration and safe handling determine whether the loop runs without interruption.

The contract base is unusually important for an industrial company:

• Approximately 90% of 2025 sales came from contracts with raw-material pass-through provisions.

• Regeneration agreements typically run five to ten years, and roughly 40% of production capacity serves customers under staggered multi-year commitments.

• Ecovyst reports that its top-ten customer relationships average more than 50 years. Those arrangements do not remove volume or operating risk, but they reduce direct exposure to spot pricing.

The financial position changed with the portfolio sale:

• Full-year 2025 sales were $723.5 million, adjusted EBITDA was $172.0 million and adjusted free cash flow was $78.1 million. Year-end net debt leverage was 1.2x.

• In the first quarter of 2026, sales increased 50% and adjusted EBITDA increased 87%. About $33 million of the quarterly sales increase came from sulfur-cost pass-through, so the reported growth rate overstates the underlying change.

• Higher regeneration activity, more than 30% growth in virgin-acid volume and favorable contract pricing still produced a clear improvement in gross profit, EBITDA and cash flow.

Calabrian expands the product range without moving Ecovyst far from its existing operating skill set. The acquired sites in Port Neches, Texas, and Timmins, Ontario, produce sulfur dioxide, sodium bisulfite, sodium thiosulfate and sodium metabisulfite for mining, water treatment, specialty chemicals, food processing and pharmaceuticals. The first scheduled earnings update after closing is August 5. Because the transaction closed on the final day of the quarter, second-quarter results will contain little operating contribution from Calabrian. The more useful information will be the consolidated outlook, integration plan, leverage bridge and current demand commentary.

Figure 1 — Ecovyst entered 2026 with lower leverage and stronger first-quarter operating results.

Company Overview

Ecovyst now operates across four connected activities:

• Sulfuric acid regeneration is the largest source of operating identity. Refineries use sulfuric acid in alkylation, then send the spent acid to Ecovyst for regeneration. Ecovyst manages collection, transport, processing and return delivery. Some customers are connected by captive pipeline; others are served by barge, rail or truck. The service is tied to refinery utilization, alkylate economics and customer turnaround schedules.

• Virgin sulfuric acid supplies mining, municipal and industrial water treatment, chemicals, construction materials and other process industries. The Waggaman assets acquired in May 2025 added Gulf Coast production capacity. Management attributed part of the first-quarter 2026 volume increase to Waggaman and said total virgin-acid volume grew more than 30% from the prior year.

• Treatment services and Chem32 add related process capabilities. Ecovyst treats hazardous and non-hazardous industrial waste and provides ex-situ sulfiding and activation for hydro-processing catalysts used in conventional and sustainable fuels.

• Calabrian broadens the portfolio into sulfur dioxide and derivatives. The overlap is practical: sulfur chemistry, process safety, industrial logistics and many of the same end markets. The acquired product groups serve mining and water treatment while adding food-processing, pharmaceutical and specialty-chemical applications.

Before Calabrian, 2025 reported sales were split between regeneration and treatment services at 49.9%, industrial, mining and automotive uses at 45.3%, and other uses at 4.8%. Industrial, mining and automotive sales rose sharply from 2024, partly because of higher sulfur-cost pass-through and the Waggaman contribution.

Figure 2 — Industrial, mining and automotive uses represented a much larger share of reported sales in 2025.

The Setup

Reported revenue can move faster than the underlying economics. Most contracts allow Ecovyst to pass changes in sulfur, freight and other cost inputs through to customers. When sulfur prices rise, sales and cost of goods sold rise together. Gross profit, adjusted EBITDA and cash flow provide a cleaner view of operating progress.

The 2025 figures show the difference:

• Sales increased from $598.3 million to $723.5 million, a gain of $125.2 million. Ecovyst disclosed that approximately $77 million of the increase came from sulfur-cost pass-through. The remaining $48.2 million reflects the combined effect of volume, contract pricing, Waggaman and other changes.

• Adjusted EBITDA was $172.0 million, compared with $172.7 million in 2024. Higher pricing and virgin-acid volume were offset by lower regeneration volume, customer downtime, maintenance, inflation and transportation costs.

Figure 3 — Sulfur-cost pass-through represented about 62% of the 2025 sales increase.

The first quarter of 2026 was stronger:

• Regeneration-services sales grew at a double-digit rate as refinery utilization remained high, alkylate economics were favorable and customer downtime improved from the prior-year period.

• Virgin-acid volume increased more than 30%. Adjusted EBITDA rose to $39.8 million from $21.3 million, while adjusted free cash flow improved to $4.2 million from negative $13.0 million.

The same revenue adjustment still applies:

• Of the $71.9 million increase in quarterly sales, approximately $33 million came from sulfur-cost pass-through.

• Even after separating that amount, the quarter showed higher volume and favorable pricing. Gross profit increased to $36.4 million from $19.1 million, which confirms that the improvement was not limited to revenue presentation.

Management’s May outlook called for 2026 sales of $890 million to $970 million, adjusted EBITDA of $180 million to $195 million, adjusted free cash flow of $40 million to $55 million and capital spending of $80 million to $90 million. The sales range assumed approximately $155 million of sulfur-cost pass-through. The outlook was issued before Calabrian closed and did not provide a full combined-company bridge. August should clarify the post-close earnings base and capital requirements.

Causal Mechanism

Ecovyst’s economics begin with geography:

• Sulfuric acid is costly and operationally demanding to transport, so the business is regional. Ecovyst’s network is concentrated in the Gulf Coast and California, where approximately 64% of U.S. refining capacity is located.

• The company can deliver by barge, rail, truck and, in certain cases, captive pipeline. A new supplier would need more than production capacity. It would need permits, transport access, sulfur-handling expertise, a record of safe operation and facilities close enough to customer demand.

Regional assets reduce logistics risk

Refinery alkylation is a continuous industrial process. Spent acid must be removed and regenerated acid must return at the required strength and schedule. Delays can affect the customer’s operating plan. Location and logistics therefore carry economic weight even when the underlying molecule is standardized.

The Waggaman assets fit this network:

• They add regional acid capacity in Louisiana rather than introducing a remote product line.

• Ecovyst is also investing in two projects intended to serve a growing virgin-acid customer base.

• But the relevant test is utilization: incremental capacity should translate into volume, contribution margin and cash generation without creating an excessive maintenance burden.

Contract provisions stabilize spreads

Most Ecovyst contracts include minimum-volume protection, quarterly price adjustments or both. Approximately 90% of 2025 sales occurred under contracts with some form of raw-material pass-through. Freight is generally passed directly to customers. Regeneration contracts typically run five to ten years, while virgin-acid contracts generally run one to five years. Roughly 40% of production capacity serves customers with staggered multi-year commitments.

These provisions reduce the effect of raw-material inflation on unit economics, although timing differences and contract disputes can still affect a quarter. They also make the sales line less informative when sulfur prices change quickly.

Customer integration supports retention

• Ecovyst reports that products representing more than 95% of 2025 sales held an estimated number-one or number-two supply position.

• Its top-ten customer relationships average more than 50 years.

Both figures are company statements, but the operating context supports the conclusion that switching is not frictionless. A refinery replacing its regeneration provider must rework logistics, quality controls and operating coordination around a process that cannot tolerate unreliable supply.

The service represents a small portion of a customer’s finished-product cost while remaining necessary to production. That combination can support durable relationships and pricing discipline. It does not eliminate customer bargaining power; the top ten customers represented 61% of 2025 sales, so concentration remains an important counterweight.

Figure 4 — Long contracts, pass-through terms and regional assets support continuity.

Calabrian adds adjacent chemistry

Calabrian gives Ecovyst a second sulfur platform:

• The acquired business produces sulfur dioxide and derivatives at two North American sites.

• The end markets overlap with Ecovyst’s existing mining and water-treatment exposure and add food-processing, pharmaceutical and specialty-chemical uses.

The integration case rests on operating overlap rather than a broad diversification claim:

• Ecovyst already handles sulfur inputs, hazardous materials, continuous-process plants and regional industrial logistics.

• Commercial overlap may also create cross-selling opportunities, but management has not yet provided enough post-close detail to treat those benefits as established.

Figure 5 — Regeneration, virgin acid and sulfur derivatives use a common set of industrial capabilities.

Timeline

The portfolio changed quickly over the last fourteen months:

• Waggaman added Gulf Coast capacity in May 2025.

• The Advanced Materials & Catalysts sale closed at year-end, followed by a $465 million term-loan repayment.

• First-quarter 2026 results showed improved volume and pricing.

• Ecovyst then completed a $100 million term-loan add-on in May and closed Calabrian on June 30.

Figure 6 — Waggaman, the divestiture and Calabrian all feed into the August 5 update.

Five items deserve close attention on August 5:

1. A combined-company outlook. The May guidance was issued before the acquisition closed. Investors need a bridge from stand-alone Ecovyst to the post-close sales, EBITDA, cash-flow and capital-spending ranges.

2. Calabrian integration costs and milestones. The transaction closed at quarter-end, so the income statement will say little about operating contribution. Management commentary should identify near-term costs, plant priorities, customer retention and the timing of any commercial or procurement benefits.

3. Waggaman utilization. First-quarter virgin-acid volume was strong. The next report should show whether the increase continued, which end markets drove it and how the two growth projects affect capacity and capital spending.

4. Regeneration uptime. Lower customer downtime helped first-quarter results. Refinery turnarounds and unplanned outages can move quarterly volume materially, so current schedules matter.

5. Cash conversion and leverage. Capital spending remains elevated, and the Calabrian financing increased debt from the post-divestiture low. EBITDA growth needs to convert into free cash flow after maintenance, growth projects, interest and integration spending.

Key Risks

The main risks can be observed directly in operating data.

• Customer concentration: The top ten customers represented approximately 61% of 2025 sales, and the largest customer represented 12%, or $89 million. Further refinery consolidation could strengthen customer bargaining power.

• Customer downtime and turnarounds: Extended outages reduced regeneration volume in 2025. Similar events can affect sales, plant utilization and transportation costs.

• Pass-through presentation: Higher sulfur prices can create rapid reported sales growth without a comparable increase in gross profit. The 2026 sales outlook includes approximately $155 million of sulfur-cost pass-through.

• Capital intensity: The current 2026 capital-spending range is $80 million to $90 million. Maintenance, environmental and process-safety spending are necessary parts of the business.

• Calabrian integration: The acquisition adds two sites, a Canadian operation, new product groups and purchase-accounting complexity. Customer retention, safe plant operation and integration discipline will determine the result.

• Balance-sheet re-expansion: Ecovyst funded part of Calabrian with a $100 million floating-rate term-loan add-on. Management estimated combined net debt leverage of approximately 2x at closing.

• Regional exposure: Gulf Coast facilities face hurricanes and freezes; California facilities face earthquakes and wildfires. Concentrated assets improve logistics but raise exposure to regional disruptions.

• Refining exposure: Regeneration demand remains linked to refinery utilization and alkylate production. Long-run changes in gasoline demand, refinery capacity or fuel specifications could affect the service base.

Figure 7 — Concentration, capital spending, pass-through and operating continuity are the central monitoring points.

Conclusion

Ecovyst’s quarterly results will continue to reflect refinery outages, maintenance schedules, sulfur prices and industrial demand. Those variables are part of the business rather than temporary noise. The investment case depends on whether the regional network, contract structure and customer relationships produce stable margins and cash generation across those fluctuations.

Three measures provide a practical scorecard:

• First, underlying volume should grow after removing sulfur-cost pass-through.

• Second, adjusted EBITDA should reflect the benefit of higher utilization and contract pricing.

• Third, adjusted free cash flow should remain credible after capital spending and integration costs.

The August 5 report will add a fourth test: management must explain how Calabrian changes the earnings base, capital requirements and leverage path. A useful update will separate stand-alone Ecovyst performance from acquisition effects and give investors concrete milestones for the two acquired sites.

Refineries rely on acid regeneration to keep alkylation units operating. Mining, water-treatment and specialty customers rely on dependable regional supply of sulfur-based chemicals. Ecovyst earns its place in those processes through plants, logistics, contracts and an operating record built over decades. The next stage is to show that the broader sulfur portfolio can improve cash generation without weakening that operating discipline.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (GPN US) which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Management sold Issuer Solutions, bought Worldpay and described the result as a focused merchant-commerce company. Investors heard a more familiar story: another large payments deal, another integration, more debt and another set of adjusted numbers.

Global Payments (GPN US) is interesting precisely because the transaction looks like the opposite of simplification.

Underneath the deal structure is a cleaner product map. Integrated payments, vertical software, point of sale, enterprise acquiring, ecommerce and international distribution now point at the same buyer: the merchant. The company no longer has to split capital and attention between merchant commerce and a separate issuer-processing business.

The important nuance is that this is not an integration-success story (yet). It is an integration-proof story. Global Payments does not need to become a high-growth software company. It needs to show that a much larger merchant platform can preserve growth, remove duplicate cost, keep service stable and turn reported earnings into cash.

In Q1 2026, adjusted net revenue was $2.86 billion, normalized constant-currency growth was 4.5%, adjusted operating margin reached 39.9% and adjusted EPS was $2.96. Management said the quarter finished above internal expectations and kept the full-year framework intact. One quarter does not settle a multi-year integration, but the combined perimeter did not begin with an obvious operating break.

What makes the setup workable is that the next phase is visible.

• The company has disclosed $600 million of annual cost synergies over three years, at least $200 million of revenue synergies over time and a leverage objective of roughly 3.0x within 18-24 months.

• The merchant perimeter is more coherent than the transaction optics suggest.

• Integrated and embedded payments were already the strongest organic growth engine before Worldpay.

• Worldpay adds enterprise, ecommerce and international reach that Global Payments did not have at the same scale.

• The next six quarters should show whether the larger platform is becoming easier or harder to operate.

The market sees a serial acquirer. The product map increasingly looks like a merchant platform. Both views can be true at once. The investment question is which one becomes more visible in the financials?

Figure 1 - The operating test. Q1 establishes the baseline; the synergy and leverage targets define the next phase.

Company Overview

Global Payments sits behind the checkout. It authorizes transactions, settles funds, manages fraud, connects merchants to software, powers point-of-sale systems and supports commerce across physical and digital channels.

That description sounds generic because payments infrastructure is supposed to disappear into the background. Merchants only notice it when authorization rates fall, settlement breaks, software stops talking to the processor or support becomes slow. The product is therefore part utility and part workflow. Processing provides scale; software and embedded distribution make the relationship harder to replace.

The Worldpay transaction materially enlarges that system. At announcement, management described a combined platform serving more than 6 million customers, processing roughly 94 billion transactions and $3.7 trillion of payment volume across more than 175 countries. There is real scale here, the harder question is whether it becomes useful scale.

Before Worldpay, Global Payments reported three merchant service lines:

• Integrated & Embedded generated about $3.41 billion of 2025 revenue and was the clearest organic growth engine.

• Core Payments produced roughly $2.98 billion and was broadly flat.

• POS & Software contributed about $1.32 billion, with comparisons affected by the AdvancedMD and payroll disposals.

This mix tells us the most interesting part of the business is the distribution layer around it. When payments are built into restaurant software, dealer-management systems, healthcare workflows or vertical SaaS, the processor reaches the merchant through a product the merchant already uses to run the business.

Worldpay adds a different kind of reach: larger enterprise merchants, ecommerce, cross-border capability and a broader international footprint. Global Payments contributes more vertical software and embedded distribution. The strategic case rests on those pieces becoming complementary rather than simply larger.

Figure 2 - Pro forma scale at the April 2025 transaction announcement. Revenue and EBITDA include announced run-rate expense synergies.

Figure 3 - Integrated & Embedded was the clearest reported growth engine before Worldpay.

The Setup

The setup is essentially a credibility gap:

• Investors see a company that promised simplification and then announced another large acquisition.

• The product map shows a company that sold the non-merchant business and concentrated the remaining portfolio around commerce.

Global Payments can be strategically more focused and operationally more complicated at the same time. That is what makes the next year important.

The April 2025 announcement landed badly because the burden of proof was already high:

• Payments has spent years consolidating, separating and recombining assets.

• Worldpay itself moved from Vantiv to FIS, then into a private-equity-controlled structure, and now into Global Payments.

• Investors were being asked to believe that another owner could integrate the same large processing estate more effectively.

The transaction closed on January 9, 2026, which is when the operating clock actually started. Q1 was therefore a baseline. The 4.5% normalized constant-currency growth rate and 110 basis points of normalized margin expansion matters because they showed no immediate collapse in the combined perimeter.

The more important signal may be governance:

• In September 2025, Global Payments added two independent directors in cooperation with Elliott Investment Management and created a dedicated Integration Committee to oversee Worldpay and synergy realization.

• That does not guarantee execution, but it moves the integration closer to the board and makes slippage harder to hide behind broad strategic language.

We are not claiming this will be a seamless integration. The base case is that merchant growth and service remain intact while duplicate infrastructure and overhead come out. If that happens, the transaction begins to look like a difficult simplification rather than another layer of complexity.

Figure 4 - The first consolidated proof point. Normalized comparisons include pre-acquisition Worldpay and exclude Issuer Solutions and other divested businesses.

Causal Mechanism

The causal mechanism is a distribution-stack story:

• Merchants rarely wake up wanting a new payment processor. They want a restaurant system that reduces order errors, a dealership platform that reconciles payments, an ecommerce stack that improves authorization or a checkout that works across store, mobile and web. Payments becomes more valuable when it is attached to the operating workflow.

• Global Payments already had a meaningful position in that layer. Its Integrated & Embedded service line grew from roughly $2.97 billion in 2023 to $3.41 billion in 2025, while the other two reported lines were flatter or affected by divestitures. That is the clearest evidence that software-led distribution, rather than commodity processing alone, is the part of the model with the best organic shape.

• Worldpay broadens the top of the funnel. It adds enterprise merchants, ecommerce volume, international acceptance and a large installed base through which Global Payments can sell fraud tools, omnichannel products, vertical software and value-added services. Global Payments brings more embedded and vertical-market distribution into the combined company. The revenue opportunity is therefore less about inventing a new product and more about moving existing products through a wider set of relationships.

• Recent announcements are useful as product-map evidence. Global Payments unveiled an AI-enabled Genius handheld for restaurants, signed an exclusive agreement with CKE Restaurants, expanded its embedded-payments partnership with Lightspeed DMS and had Link2Gov selected as a preferred IRS partner. However, none of those releases disclosed a material revenue contribution. They should be read as evidence that the company still has relevant distribution and product activity during the integration, not as proof that the revenue-synergy target has arrived.

The second part of the mechanism is cost:

• Management expects $600 million of annual cost synergies over three years. That target is more controllable than the $200 million-plus revenue opportunity because duplicate vendors, facilities, infrastructure and support functions can be mapped and removed. But cost savings are only useful if the merchant does not feel them first. The clean outcome is lower cost with stable service. The bad outcome is lower cost created by cutting into implementation, support or product development.

The third part is cash:

• A transaction this large creates purchase-accounting noise, integration expense and working-capital movement. Adjusted earnings will therefore look cleaner than the cash-flow statement for a while. The right test is not to accept or reject adjusted results on principle. It is to watch whether operating cash flow, capital spending, integration outlays and debt reduction begin to line up over time.

The chain is simple: a focused merchant perimeter creates a more coherent product roadmap; the combined distribution base creates more places to sell those products; cost removal funds investment and improves operating conversion; cash generation gives the company room to reduce leverage. The chain breaks if integration harms service, merchant growth slips or adjusted earnings fail to become cash.

Figure 5 - How the integration becomes operating proof. Management targets are reported; the sequence is Ridire Research interpretation.

Timeline

The next year is about proof. The second half of 2026 should show whether the Q1 baseline was durable. Normalized growth should remain close to the mid-single-digit framework, margin progress should continue without deterioration in client commentary, and the company should begin making the cash effects of integration easier to understand.

By 2027, the conversation should move from closing the transaction to operating the platform. Investors should be able to see more than cost removal: cross-sell evidence, product adoption, stable retention, fewer duplicated systems and a clearer path toward the leverage objective.

An integration this large can always produce one clean quarter. The important signal is the sequence. Stable growth, margin progress, cash conversion and lower leverage moving together would be difficult to dismiss. A widening gap between those measures would be equally difficult to explain away.

Figure 6 - Timeline.

Key Risks

The right risk lens is straightforward: the deal may be strategically coherent and still be operationally too hard. The main risk is not that electronic payments stop growing. It is that the benefits remain trapped inside a system that is too complex to deliver them cleanly.

• Service risk: platform migrations, support changes and cost removal can create outages, slower implementation or merchant attrition.

• Growth risk: Worldpay adds scale, but scale does not fix weak product velocity or lost share. The combined company still has to win against Fiserv, Adyen, Stripe, Toast, Block and vertical software providers.

• Synergy-quality risk: the company may hit a cost number by cutting too deeply into product, sales or service. Margin expansion would then flatter the short-term result while weakening the franchise.

• Cash-conversion risk: purchase accounting and integration charges can keep the gap between adjusted earnings and cash open longer than expected.

• Leverage risk: the debt load reduces room for error. A slower integration, weaker merchant volumes or higher funding costs would make the balance-sheet objective harder to reach.

• Complexity risk: the company may end with a cleaner corporate perimeter but an internally fragmented collection of platforms, contracts and regional operating models.

Conclusion

Global Payments is not a clean story. That is precisely why it is interesting.

• The market sees another processor buying another processor. The better interpretation is that Global Payments sold the part of the company serving issuers and concentrated the remaining portfolio around one problem: helping merchants accept payments and run commerce workflows.

• The company does not need to prove everything at once. It needs to prove that the larger platform can preserve mid-single-digit growth, improve margins without damaging service, convert earnings into cash and make the product map more legible. Those are ordinary operating goals. The difficulty is achieving them inside an extraordinary integration.

• Q1 says the new perimeter did not start broken. That is a low bar, but an important one. The next six quarters will show whether Worldpay becomes a wider distribution system for embedded payments and merchant software, or simply another processing estate that Global Payments has to manage.

The payments spin nobody believes in is not that the company found a new market. It is that a business known for complexity may finally have built a perimeter coherent enough to simplify.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that does not currently hold long or short positions in the securities discussed herein, but may in the future, which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary & Special Note

2. Company Overviews

3. The Setup

4. IP Monetization Capacity Score

5. Causal Mechanism

6. Timeline

7. Key Risks

8. Conclusion

1. Executive Summary

_{Special Note: This article, along with the proprietary measures and analytical insights presented throughout, was co-authored by Carson Hawley. Carson brings deep experience in the video game development industry and is currently exploring new opportunities. If you’re interested in collaborating with Carson or discussing the industry with him, you can connect with him on LinkedIn HERE.}

Most software investors are underwriting the same trade in different wrappers: fewer employees, more automation, higher margins. Our view is that this is too crude for the video games subsegment.

Games are not simply software. They are libraries, characters, worlds, songs, battle systems, fan memory, recurring communities, and half-depreciated catalogs that can be resold for decades if the owner keeps the franchise alive.

In that structure, labor is not automatically deadweight. A writer, artist, engineer, localization team, or live-ops group attached to owned IP is closer to a refinery than an overhead department. The value sits in what the company can issue from the library.

This basket expresses that distinction:

• Long the publishers that own the economics of the next unit of content: Nintendo, Capcom, Sega Sammy, Krafton, NetDragon, and Koei Tecmo.

• Short Roblox, where usage is large but shareholder capture is polluted by creator payouts, infrastructure, safety, moderation, and regulatory exposure.

• In brief: long owned-IP publishers where content economics accrue to the franchise owner; short Roblox, where UGC scale carries higher leakage, moderation burden, and governance cost. Full disclosure: Basket inclusion also cross-referenced names above our IP-capture threshold with Ridire Research’s proprietary fundamental risk/reward metrics.

There is a nuance with Sega worth highlighting: Sega Europe did announce layoffs and sold Relic. However, Sega’s IP still requires a creative workforce to convert legacy assets into products, licensing, mobile adaptations, remasters, and transmedia revenue. A company with weak IP uses productivity tooling to shrink but a company with real IP can use it to issue more revenue-bearing content from the same library.

This distinction can be standardized. We developed an internal score used here called IP Monetization Capacity: a measure of how much of the productivity gain actually reaches shareholders. (More on this in Section 4)

For the basket components, Nintendo and Capcom justify the top scores: Nintendo owns the full stack; Capcom has the cleanest franchise execution. Sega ranks as recovery torque, not quality earnings, valuable library. Krafton is live-service leverage through PUBG. NetDragon and Koei Tecmo are narrower throughput cases. Roblox scores poorly because usage leaks into creator payouts, safety, moderation, and infrastructure before it reaches shareholders.

Basket Calculation: (7974 JP Equity * 0.23 + 9697 JP Equity * 0.21 + 6460 JP Equity * 0.17 + 259960 KS Equity * 0.17 + 777 HK Equity * 0.13 + 3635 JP Equity * 0.09) / (RBLX US Equity * 0.40)

2. Company Overviews

The basket is a stack of different IP-duration profiles:

• Nintendo (7974 JP)

  • It owns the hardware interface, first-party software economics, royalty rails, characters, merchandise, and filmed-entertainment extension.

  • FY2026 net sales rose 98.6% to ¥2.31tn, operating profit rose 27.5% to ¥360.1bn, and Switch 2 hardware reached 19.86m units with 48.71m units of Switch 2 software.

  • The debate is how much of Switch 2 becomes high-margin software and recurring IP income rather than a one-cycle hardware pull-forward.

• Capcom (9697 JP)

  • The company reported its ninth consecutive year of record profits, eleventh consecutive year of more than 10% operating-profit growth, and 59.07m consumer game units sold.

  • Capcom’s edge is discipline: allocate capital to franchises with global repeatability, keep the catalog alive digitally, and avoid cheapening the product with obvious automation artifacts.

• Sega Sammy (6460 JP)

  • FY2026 was not clean. Net sales rose, but operating income declined and the company reported a net loss tied partly to Entertainment Contents weakness and impairments around Rovio and Stakelogic.

  • The offset is asset value: Sonic, Persona, Like a Dragon, Football Manager, Rovio/Angry Birds, and the broader catalog. Licensing revenue associated with transmedia rose 31.6% year over year, which is the part of the P&L that matters most for the thesis.

• Krafton (259960 KS)

  • PUBG remains the profit pool. Q1 2026 revenue rose 56.9% year over year to KRW1.371tn, operating profit rose 22.8% to KRW561.6bn, and PUBG franchise revenue exceeded KRW1tn in the quarter.

  • Krafton has the live-service surface area where faster iteration, personalization, and content cadence can matter quickly. The risk is internal disruption and the upside is that a global cash engine funds the transition.

• NetDragon (777 HK)

  • FY2025 gaming and app revenue was RMB3.3bn, or 73.2% of total revenue. Gaming and app core margin rose 3.9 points to 27.4%, gaming R&D expense fell 33.2%, and management disclosed 25% AI-driven workload with a 50% end-2026 target.

  • It is not the purest franchise-quality long but it is possibly the best case study:

    • Alignment is unusually strong: founder-chairman Liu Dejian and Liu Luyuan were deemed to control 40.73% of shares at year-end 2025.

    • NetDragon also uses Tang Yu, an AI-powered virtual executive, as rotating CEO for organizational affairs and strategic execution at its flagship subsidiary

• Koei Tecmo (3635 JP)

  • FY2026 Entertainment segment sales were ¥82.5bn. Console and PC sales increased with 14 new titles, total units rose 84.8% to 14.1m, overseas units rose 88.4%, and the digital ratio reached 61.8%.

  • Headcount increased by 151 to 2,835. Koei is still adding production capacity because the unit economics of controlled IP justify it.

• Roblox (RBLX US) is the short hedge.

  • Q1 2026 revenue rose 39% to $1.4bn, bookings rose 43% to $1.7bn, DAUs rose 35% to 132m, and hours engaged rose 43% to 31bn.

  • The cost of governing their growth is simply too high. Roblox reduced expectations because of safety initiatives, expects sequential DAU pressure, and guided full-year 2026 bookings growth to 8–12%, with a projected net loss of $1.035bn–$1.175bn.

  • This was published over a year ago but many of the key points still hold, if you would like to read it it can be accessed here for free:

3. The Setup

The market has over-learned this generic software lesson: it sees automation and immediately looks for layoffs. That works when the company sells “workflows”. It misses the economics of owned entertainment.

A relevant tangent on selling “workflows”: Alex Karp (CEO of PLTR), of all people, recently changed our perspectives on the whole thing in this interview. If these models are really that powerful, why is everyone still selling tokens? Why not just sell the workflow that makes you a billion dollars?

It is a crude simplification, but if LLMs created that much enterprise value, vendors would charge against the value created rather than simply billing token usage. Raw model access is not the scarce asset. The money should accrue to the person who owns the workflow, the distribution, the proprietary context, and the end-market where the output can actually be monetized. Tokens are the commodity layer.

Furthermore, the market has already produced a kind of reverse-engineered confession:

• Social media is full of users posting the same failure mode: the model was not “good enough,” the agent wandered, the output still required human cleanup and somehow the token bill was spectacular.

• At some point the question stops being technical and starts being economic, consider:

  • Do the labs not see the inefficiency in how these agents behave? Or do they see it perfectly well, but have no incentive to fix it quickly?

• Their pricing model answers the question:

  • They charge for token consumption.

  • They are paid when the machine thinks longer, loops more, retries more, and burns more inference.

  • Their incentives are straightforward: They get paid when you consume more tokens, not necessarily when you reach the right answer faster.

This incentive structure is worth keeping in mind because it extends well beyond foundation models. The economic question is not simply who builds the best model, but who owns the layer where value is ultimately captured. In enterprise software that may be the workflow but in gaming, it is the intellectual property.

A publisher with deep IP is not merely trying to produce the same number of games with fewer people. It is trying to increase the number of credible commercial surfaces from a finite library: sequel, remaster, DLC, mobile adaptation, localization, animated series, film, merchandise, in-game event, live-service season, or catalog relaunch. The marginal product is not code or a workflow, rather it is another monetizable appearance of a familiar world.

The long/short construction therefore is essentially owned issuance versus platform leakage.

4. IP Monetization Capacity Score

The score below is an internal rubric, not a disclosed company metric. It asks one question: when production friction falls, how much of the benefit reaches the equity holder?

The components are deliberately ownership-weighted: depth of owned IP, proof of franchise monetization, ability of labor to convert catalog into product, integration of productivity tooling, transmedia optionality, and the absence of safety or regulatory leakage.

• Nintendo scores highest because it owns the stack.

• Capcom scores nearly as well because execution discipline has already turned catalog depth into repeatable earnings.

• Sega ranks below them because the P&L is messier, but above most of the group on latent IP optionality.

• Krafton has live-service torque.

• NetDragon has the cleanest disclosed productivity evidence.

• Koei has niche IP and visible production investment.

• Roblox ranks materially lower because the platform captures scale only after paying creators and policing the system.

The nuance in interpreting the score is that the crucial test is not whether the company can automate work. The test is whether the company owns something worth producing more of.

A bad software business can automate and become a smaller bad software business.

A good IP business can automate and issue more product without surrendering the economics.

5. Causal Mechanism

Library ownership → production capacity → lower iteration friction → more commercial surfaces → earnings durability → multiple support

• Nintendo and Capcom are the cleanest versions. They already have global franchises and operating discipline. Production leverage is incremental, not existential.

• Sega: The market has good reasons to distrust the P&L: impairments, underperformance, and integration noise. But the IP base is too valuable to price only off recent execution. If Sonic, Persona, Like a Dragon, Football Manager, and Rovio begin showing up as a broader licensing and transmedia engine, the market can stop underwriting Sega as a troubled content house and begin underwriting it as an under-monetized library.

• Krafton: PUBG gives the company a high-frequency surface where better tooling should show up in cadence, events, personalization, and operating leverage faster than in a traditional boxed-title model.

• NetDragon: The company is already giving investors the kind of disclosure the theme wants: measurable workload automation, R&D compression, and margin improvement. The risk is legacy game maturity; the attraction is visibility into the mechanism.

• Koei Tecmo is the overlooked version. Smaller franchise base, lower global heat, but credible unit growth, digital mix, overseas acceleration, and headcount investment. It belongs in the basket because not every IP throughput story needs to be a large-cap.

• Roblox is the short because the mechanism runs in reverse. Content volume rises, but so do the liabilities attached to running a youth-heavy UGC platform. More engagement does not automatically mean cleaner equity economics. The company can keep growing and still deserve a lower multiple if investors reclassify it from “frictionless platform” to “regulated social infrastructure with creator leakage.”

6. Timeline

The trade has three windows: productivity proof, franchise conversion, and Roblox guidance pressure.

• 2026 is the proof-of-mechanism year. NetDragon should be judged against the 50% workload target, gaming R&D efficiency, and whether margin improvement offsets legacy revenue maturity. Krafton should be judged on whether the internal transition improves live-service cadence without damaging culture or product quality.

• FY2027 is the Sega test. Sega needs cleaner Entertainment Contents performance, continued licensing/transmedia growth, and evidence that Rovio is not another capital-allocation scar. It needs enough evidence that the library is worth more than the cleanup discount.

• Nintendo’s next test is attach and mix. FY2027 guidance calls for ¥2.05tn of net sales, ¥370bn of operating profit, 16.5m Switch 2 hardware units, and 60m Switch 2 software units. Hardware sell-through opens the door; software attach, first-party mix, and IP-related income decide the margin profile.

• Capcom’s test is cadence. PRAGMATA, Onimusha, and the broader slate should be monitored less as isolated launches and more as evidence that Capcom can launch new or revived IP without weakening the catalog model. The draft notes PRAGMATA’s rapid sell-through as a validation point for this cadence argument.

• Roblox is the near-term hedge catalyst. The next few quarters should show whether safety changes are a temporary DAU disturbance or a structural cost of doing business. If bookings stay near the reduced 8–12% guide and safety costs remain elevated, the short continues to have oxygen. If DAUs stabilize quickly and older cohorts monetize better, the short becomes dangerous.

7. Key Risks

The biggest risk is mistaking production leverage for franchise quality.

• Sega can fail the trade by remaining a cleanup story. The IP is valuable, but investors will not pay for latent library value indefinitely if execution keeps producing impairments and missed titles.

• Capcom and Nintendo can fail through brand dilution or cycle mispricing. Capcom’s discipline is part of the asset; if production tooling leaks visibly into the final product and consumers read it as cheap content, the quality multiple should compress. Nintendo’s risk is more classical: Switch 2 can trade like a hardware cycle if pricing, component cost, tariffs, or attach disappoint.

  • Side note on memory: the console tax. Nintendo’s hardware risk is the collision between a fixed-price consumer device and an AI-driven memory market. Consoles need DRAM and NAND; AI infrastructure is now competing for the same supply chain with a much higher willingness to pay.

    

    • That leaves console makers with an ugly choice: absorb the cost, raise hardware prices, or protect margin and risk slower adoption. Switch 2 is not the thesis by itself.

    • It is the installed-base ramp that feeds the higher-margin part of the model: first-party software, attach, catalog, DLC, subscriptions, licensing, and broader IP monetization. If memory inflation turns the hardware layer into a margin fight, the market can temporarily stop valuing Nintendo as an IP compounder and start treating it like a hardware name again.

    • Same basic issue for Sony and Microsoft: AI may lower the cost of content production, but it can raise the cost of the box that content depends on.

• Krafton can fail through culture. An aggressive internal transformation that reads as restructuring rather than product improvement can damage morale and slow output. Live-service companies do not get infinite chances to disrupt themselves.

• NetDragon can fail if cost compression arrives after the revenue base has already faded. Productivity helps only if the company has enough demand to monetize.

• Roblox can break the short if safety investment normalizes quickly and the company monetizes older users at a higher rate. The short is not premised on cultural irrelevance. It is premised on a worse shareholder capture rate than the market is willing to price.

8. Conclusion

• The basket expression is long owned IP where production leverage accrues to the owner; short UGC scale where production volume increases governance burden before it reaches shareholders.

• Engagement is not the same thing as owned economics.

• A company cutting labor because workflows are replaceable is not the same as a company retaining creative labor because the labor sits next to monetizable IP. Cost story vs issuance story.

The basket expression remains:

• Long 7974 JP, 9697 JP, 6460 JP, 259960 KS, 777 HK, 3635 JP. Short RBLX US.

• Calculation: (7974 JP Equity * 0.23 + 9697 JP Equity * 0.21 + 6460 JP Equity * 0.17 + 259960 KS Equity * 0.17 + 777 HK Equity * 0.13 + 3635 JP Equity * 0.09) / (RBLX US Equity * 0.40)

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (Shoucheng Holdings, 697 HK), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Shoucheng Holdings (697 HK) is becoming a listed proxy for China’s robotics commercialization stack. It is a listed platform with indirect exposure to one of China’s most visible robotics champions (Unitree), plus operating infrastructure that can help robotics companies move from prototype to sales, service, deployment, and field data.

The center of the thesis is Unitree:

• Unitree has cleared the Shanghai Stock Exchange listing committee review, and its prospectus identifies the Beijing Robot Industry Development Investment Fund as a pre-IPO shareholder.

• Shoucheng says it participates in managing that fund’s portfolio and has concentrated investments across more than 20 robotics and embodied-intelligence companies. The market-facing asset is therefore not a parking operator with a robotics footnote.

The setup became sharper after Unitree’s IPO process advanced.

• Xinhua reported that Unitree’s IPO application passed SSE listing review on June 1, 2026, with 2025 revenue of roughly RMB1.7 billion and expected 1H 2026 revenue above RMB1.0 billion.

• The IPO plans to raise funds for robot model R&D, robot body R&D, new product development, and a manufacturing base.

The investment question is now operational: can Shoucheng turn robotics exposure into visible economics through fund marks, Taozhu retail/service channels, deployment sites, maintenance, charging, dispatch, and asset recycling?

Company Overview

Shoucheng describes itself as a service provider for China’s intelligent infrastructure assets. The company’s two core engines are industrial funds and asset operation, but the strategic direction is increasingly organized around a robotics ecosystem: capital, scenarios, supply-chain integration, commercialization, and asset-management exits.

The robotics exposure is broader than Unitree:

• Shoucheng’s annual report lists a portfolio spanning Yushu Technology / Unitree, Noetix Robotics, Galbot, Deep Robotics, Booster Robotics, Xinghaitu, TowardPi Medical, Rossum Robot, X-magtech, Wisson, Volant, Differential Robotics, Matrix Technology, MOTOREVO Robotics, ROBOTERA, MarveLab, and others.

• The listed fields include embodied intelligence, humanoid robots, medical robots, industrial robots, low-altitude economy, core robotics components, and consumer robots.

Shoucheng’s own disclosure says the Beijing Robot Industry Development Investment Fund portfolio, which the company participates in managing, increased by approximately 4x. That is not the same as cash realization, but it gives the stock a direct reason to trade with the private robotics cycle.

The Setup

In simplest terms the setup is Unitree visibility plus China robotics scale.

• China is already the largest industrial robot deployment market globally. The International Federation of Robotics reported that China accounted for 54% of global industrial robot deployments in 2024, with roughly 295,000 annual installations and more than 2 million industrial robots in operation. Chinese manufacturers also reached a record domestic market share of 57%.

• Unitree gives the Shoucheng thesis a visible flagship. Unitree’s prospectus shows humanoid robot revenue moved from a negligible share in 2023 to the largest reported product category in the first nine months of 2025. For 2025 1–9M, humanoid robots were 51.53% of product revenue, quadruped robots were 42.25%, and components were 5.76%.

• The next layer is channel and service density. Unitree’s prospectus shows sales remain heavily offline: for 2025 1–9M, offline sales represented 86.50% of revenue, while online sales represented 13.50%. Domestic sales were 60.80%, with overseas at 39.20%.

That is where Shoucheng’s physical assets enter the story. Robotics commercialization still needs demo floors, service points, repair capacity, training environments, charging locations, field data, and enterprise customer access.

Causal Mechanism

1. Unitree makes the robotics portfolio legible

Unitree is the recognizability layer. The fund stake gives public-market investors a concrete way to think about Shoucheng’s robotics exposure. The Beijing Robot Industry Development Investment Fund appears in Unitree’s shareholder table with 3.8262% pre-IPO ownership, and Shoucheng says it participates in managing the same fund’s portfolio.

2. Taozhu gives the portfolio an offline channel

Shoucheng’s Taozhu New Manufacturing Bureau is the commercial testbed. The company says Taozhu integrates robot experience, sales, maintenance, and service. It has opened Beijing stores including Shougang Park Rongshi Plaza, Beijing Capital Airport Terminal 2, Beijing Capital Airport Terminal 3, and Wangfujing APM. Shoucheng also says nearly 100 robotics companies have signed as authorized agents, covering industrial robots, service robots, and intelligent solutions.

The Taozhu KPI is not really store count. The KPI is whether those stores become a lower-friction path from robotics manufacturer to customer: product trial, purchase, installation, maintenance, and repeat service.

3. Parking sites become robot service locations

Shoucheng’s parking disclosures have started to move toward service revenue. The company says static traffic assets are evolving from parking fees toward multi-format commercial revenue. Disclosed operating KPIs include innovative business revenue contribution of 20%, revenue yield per space up 17%, turnover up 7.1%, and vacancy down 8%.

The company also lays out the robot-field services explicitly: dispatch fees, robot charging / swapping / custody fees, maintenance and OTA services, commerce and advertising, and intelligent business docking fees. Management describes a shift toward charging, operation, maintenance, and dispatching, using parking lots as distributed nodes for unmanned-vehicle closed-loop operations.

4. Industrial parks provide real deployment data

Shoucheng says it manages more than 100 parking facilities and millions of square meters of industrial parks across Beijing-Tianjin-Hebei, Chengdu-Chongqing, East China, and the Greater Bay Area. The company states these assets can provide robotics companies with real operational data. It also disclosed an automatic charging robot pop-up store with Wisson at Chengdu ICD Mall, tied to a “parking + charging” model.

In other words Shoucheng is not only a financial investor in robotics companies. It can offer commercial scenes, physical traffic, operating data, service density, and customer touchpoints.

5. REITs and securitization create the exit route

The REIT layer is supporting infrastructure. Shoucheng has disclosed infrastructure funds and REIT-related strategies across technology parks, data centers, consumer infrastructure, clean energy, and urban development assets. The company’s model is to invest, operate, improve, securitize or exit, and retain fee-bearing economics where possible. The robotics platform becomes more attractive if deployment sites can be improved, monetized, and recycled rather than sitting permanently on balance sheet.

Timeline

The Unitree review already cleared the first public gate. The Shanghai Stock Exchange listing committee announced that Unitree met issuance, listing, and information-disclosure requirements, while also asking follow-up questions on commercialization, competition, downstream demand, technology differences, future R&D, expenses, and performance volatility.

Key Risks

1. Unitree exposure is indirect

Shoucheng has not disclosed a simple direct stake in Unitree at the parent-company level. The exposure runs through fund structures. The Beijing Robot Industry Development Investment Fund is listed as a Unitree shareholder, and Shoucheng says it participates in managing that fund’s portfolio. The economics may be diluted by fund terms, lock-ups, carry structures, minority ownership, and distribution timing.

2. IPO approval does not equal cash realization

SSE listing committee approval is a major milestone, but it is not the same as final registration, pricing, listing, or post-lock-up liquidity. The public-market value assigned to Unitree will drive perception, but Shoucheng’s realized economics depend on the fund’s ownership, holding period, exit path, and accounting treatment.

3. Robotics adoption can stay demonstration-heavy

Unitree’s growth is impressive, and the humanoid mix shift is visible. The harder underwriting question is recurring commercial deployment. The SSE committee’s questions on application fields, commercialization progress, competition, downstream demand, and future performance volatility point directly at the unresolved issues.

4. Taozhu needs store economics

Taozhu has strong locations and nearly 100 authorized robotics partners, but public disclosures still need harder commercial KPIs: GMV, conversion, gross margin, service revenue, repair volume, and repeat customer activity. Store count alone is not enough.

5. Parking-site monetization may take longer than expected

Shoucheng has disclosed the right service categories: charging, dispatch, maintenance, OTA, advertising, and business docking. The proof will be segment revenue. Until those services appear in numbers, the parking-infrastructure leg remains a forward operating claim.

6. Fund marks are not distributions

The reported fourfold increase in the Beijing robotics fund portfolio is strategically important. It still needs to convert into realized gains, recurring management economics, or cash distributions before it can support a durable re-rating.

Conclusion

Shoucheng’s most interesting asset is its position inside China’s robotics funding and commercialization network. The Unitree look-through stake makes the story visible. Taozhu, parking facilities, industrial parks, charging sites, maintenance channels, and REIT exits explain how that visibility can become operating economics.

The stock works if three things happen over the next 12–24 months:

• Unitree becomes a liquid valuation anchor,

• Taozhu shows real sales and service activity,

• and Shoucheng’s physical sites begin reporting robotics-linked revenue.

Without those proofs, 697 HK remains a listed infrastructure/fund platform with a robotics narrative. With them, it becomes one of the cleaner Hong Kong-listed rails into China’s robotics commercialization cycle.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (CNX), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. Casual Mechanism

4. Key Risks

5. Conclusion

Executive Summary

This is a shorter piece briefly highlighting a smaller natural gas asset we believe remains overlooked by the market: CNX Resources Corporation is a disciplined Appalachian natural gas asset with long-duration reserves, strong free cash flow generation, and second-order tailwinds from a structurally improving U.S. gas demand backdrop.

The market still tends to value Appalachian gas producers as cyclical commodity equities trapped in oversupplied domestic markets. That assumption is becoming less durable.

LNG exports are steadily globalizing U.S. gas markets, while AI-related power demand, industrial reshoring, and grid reliability needs are increasing the strategic importance of dispatchable natural gas generation.

CNX benefits from that shift indirectly rather than explicitly. The company generated roughly $646mm of FCF in 2025 and reported 9.7 Tcfe of proved reserves, giving it meaningful reserve duration without requiring aggressive production growth.

The thesis is less about near-term gas torque and more about the possibility that long-duration U.S. gas reserves deserve a higher quality multiple if domestic and global gas demand remain structurally firmer than the market expects.

Company Overview

CNX Resources Corporation is an Appalachian natural gas producer focused on development, production, and midstream operations across the Marcellus and Utica regions. The company’s appeal is mainly capital discipline. CNX has consistently prioritized:

• Free cash flow generation

• Share repurchases

• Long reserve life

• Controlled capex intensity

That matters because Appalachian gas historically suffered from a “trapped commodity” narrative: abundant supply, weak regional pricing, and limited export pathways. The setup is slowly improving.

The company reported 9.7 Tcfe of proved reserves as of year-end 2025, providing significant inventory duration relative to current production levels.

Causal Mechanism

The thesis does not require CNX to become directly tied to AI infrastructure spending like many other natgas plays out there right now. Instead, the mechanism is:

U.S. LNG exports continue to rise as global buyers seek non-Russian (and non-blockaded?) energy supply in addition to long-term energy security for AI buildouts.

At the same time, natural gas remains one of the few scalable and dispatchable power sources capable of supporting incremental grid load growth. AI-related data-center demand is part of that backdrop, but only one component of it. The important distinction is that CNX is not really a “theme stock.” It is a fundamentally solid gas asset that may benefit from second-order thematic tailwinds:

• LNG export growth

• Grid reliability demand

• Industrial reshoring

• AI-related electricity consumption

Key Risks

The largest risk is simple: CNX is still a commodity-linked equity. Natural gas prices remain volatile and heavily influenced by:

• Weather

• Storage levels

• Production growth

• LNG outage risk

• Pipeline constraints

CNX also actively hedges production, which can reduce upside participation during stronger gas environments:

There is also persistent regulatory and infrastructure risk in Appalachia, particularly around pipeline permitting and regional basis dynamics. Finally, if LNG demand growth or power-load growth disappoints, the market may continue treating Appalachian gas reserves as structurally low-multiple assets.

Conclusion

CNX is best viewed as a disciplined, cash-generative Appalachian gas producer with long-duration reserves and improving macro context. The investment case does not directly rely on speculative AI enthusiasm or aggressive production growth. Instead, it rests on a more incremental but credible idea:

U.S. natural gas may remain strategically important for longer than the market expects, and long-life, free-cash-flow-oriented gas assets could be revalued accordingly.

CNX offers exposure to that possibility through reserve depth, capital discipline, and shareholder-oriented cash deployment rather than growth narratives.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds short positions in the securities discussed herein (RPD, TENB, QLYS, FROG), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overviews

3. Casual Mechanism

4. Key Risks

5. Conclusion

Executive Summary

The market is likely getting the direction of cybersecurity spending right, but the distribution wrong. AI should increase security budgets. Agents, non-human identities, AI-generated code, data leakage, prompt injection, runtime model abuse, and automated workflows all expand the enterprise attack surface.

But the more important question is not whether cybersecurity spending grows. It is who captures the incremental dollar.

Our view is that the next leg of security spending increasingly accrues to broad platforms that can unify identity, data, cloud, code, AI runtime behavior, and policy enforcement into a single control plane. That structurally disadvantages narrower point-solution vendors whose products solve important, but increasingly subscale, pieces of the problem.

This creates a short opportunity in an equal-weight basket of:

• TENB — Tenable

• RPD — Rapid7

• QLYS — Qualys

• FROG — JFrog

These are not “zero” businesses. They have real products, real customers, and real recurring revenue. The short thesis is more subtle:

They can remain useful while becoming less strategic.

That is where software multiple compression happens. The product still matters, but the market stops valuing it as a control point. The AI security era does not reward another dashboard, another scanner, another compliance workflow, or another isolated DevSecOps tool. It rewards the platform that can answer a higher-order question:

What is the real business risk of this identity, workload, code change, data flow, or AI-agent action, and should it be allowed?

That answer requires integrated context. Integrated context favors platforms. Platforms compress point solutions.

Company Overview

TENB — Tenable

Tenable is primarily a vulnerability and exposure management company. Its core historical value proposition is visibility: helping enterprises understand where vulnerabilities, misconfigurations, and exposures exist across their environment.

That was a strong category in a world where visibility itself was scarce. (Worth noting here for any bulls reading that TENB is in our view the best company relatively speaking in this basket)

But the category is moving from vulnerability identification toward automated, contextual risk prioritization and enforcement. The buyer wants an integrated risk graph that understands cloud posture, identity privileges, runtime exposure, sensitive data, asset criticality, and remediation priority.

A scanner says:

“This asset has a vulnerability.”

A modern control plane says:

“This externally reachable workload contains sensitive data, has excessive permissions, is tied to a privileged identity, and should be remediated before lower-severity findings.”

Tenable’s 2025 filings already show the nuance: revenue still grew 11% to $999.4 million, but new enterprise platform customer additions were 1,667 in 2025, down from 1,689 in 2024 and 1,788 in 2023. Customers with $100K+ annual contract value did grow to 2,161, but the new-customer trend is not screaming greenfield acceleration. The business remains real, but the question becomes whether Tenable is the future control plane or simply one signal source feeding someone else’s platform.

RPD — Rapid7

Rapid7 sits in an increasingly uncomfortable middle ground. It has useful products across vulnerability management, detection and response, security analytics, and SecOps workflows. But in the current market, being “somewhat broad” is not the same as owning the platform.

The future security architecture is less about aggregating alerts and more about enforcing policy across live systems. That requires identity context, cloud context, endpoint telemetry, data sensitivity, AI runtime behavior, and automated remediation.

The company’s own numbers make this the cleanest expression of the basket. Rapid7 exited 2025 with $860 million of revenue, up only 2% year over year, and $840 million of ARR, flat year over year. Its initial 2026 guide called for full-year revenue of $835–843 million, implying a 2–3% decline.

Then in Q1 2026, ARR declined 0.6% year over year to $832 million, and full-year revenue guidance remained roughly $836–842 million, again implying a 2–3% decline.

That does not mean Rapid7 disappears. It means the company may already be showing what point-solution compression looks like when the market demands broader platform ownership: stable relevance, weaker growth, and a harder narrative.

QLYS — Qualys

Qualys is the classic durable-but-less-strategic risk. The company has a strong legacy franchise in vulnerability management, compliance, and cloud-based security assessment. Its installed base is real. Its renewal base is likely more durable than the most aggressive bears assume.

But durability is not the same as incremental relevance.

Qualys generated $669.1 million of 2025 revenue, up 10% year over year, with an extremely strong 47% adjusted EBITDA margin and operating cash flow equal to 46% of revenue.

That is not a broken business. It is almost the opposite.

The issue is that Qualys increasingly looks like a very profitable mature security utility at the exact moment the market is trying to identify the next strategic AI security control plane. The company has over 10,000 customers worldwide, including a majority of the Forbes Global 100, and no customer accounted for more than 10% of revenue in 2025.

That scale is a real asset. But it also frames the short case correctly: the debate is not survival. The debate is whether the market should pay a platform multiple for a company whose growth profile is increasingly mature.

The real risk is reclassification. Qualys may move from “strategic security platform” to “mature scanning and compliance utility.” In software, that distinction matters enormously for valuation.

FROG — JFrog

JFrog is different from the other three because it sits closer to the software supply chain and DevSecOps layer. That gives it a real bull case: AI-generated code increases code volume, dependency usage, package risk, and release velocity. In theory, that should increase demand for software supply chain security.

The numbers reflect that stronger fundamental setup. JFrog grew fiscal 2025 revenue 24% to $531.8 million. Cloud revenue grew 42% in Q4 2025 and represented 48% of total revenue. Net dollar retention was 119%, customers with more than $1 million of ARR rose to 74 from 52, and customers with more than $100K of ARR rose to 1,168 from 1,018.

So FROG is the least “obvious” short in the basket on near-term fundamentals.

But that is exactly why it belongs in the basket. The issue is not whether software supply chain security matters. It does. The issue is whether JFrog owns the control plane or becomes attached infrastructure inside someone else’s developer workflow.

If AI-assisted development becomes increasingly centered around GitHub, GitLab, hyperscaler CI/CD, cloud-native developer platforms, and AI coding environments, standalone DevSecOps vendors may lose platform independence.

JFrog may grow. The question is whether the market eventually decides the strategic control point migrated elsewhere.

Causal Mechanism

1. Best-of-breed becomes best-integrated

The prior cycle rewarded category depth. The next cycle rewards cross-domain context.

Agentic security requires a system that can understand:

• who or what is acting,

• what data is being touched,

• what privileges are involved,

• what code or workload is affected,

• what the runtime behavior looks like,

• and whether the action should be blocked, allowed, or remediated.

A narrow tool can provide signal. But the platform that correlates the signal and enforces the decision captures the economics. That is the core risk for this basket. TENB, RPD, QLYS, and FROG may provide useful inputs. But the market increasingly pays for the control plane, not the input.

2. “Good enough and already installed” becomes lethal

Enterprise buyers do not always choose the best product. They often choose the product that is already deployed, already approved, already integrated, already budgeted, and good enough. That favors companies like Microsoft, Palo Alto, CrowdStrike, Zscaler, Cloudflare, Google/Wiz, and other broad platforms.

A bundled product does not need to be perfect. It only needs to be good enough to make the standalone vendor justify:

• another vendor review,

• another contract,

• another agent,

• another dashboard,

• another integration,

• and another renewal.

That is a much harder sale than “we are the best product in the category.”

3. AI compresses the labor value of narrow workflows

A lot of legacy security software monetized analyst labor. It helped teams triage alerts, prioritize vulnerabilities, create tickets, document compliance, or manage workflows. AI can automate more of that labor layer.

As AI summarizes vulnerabilities, generates remediation steps, writes patches, classifies alerts, maps dependencies, and automates workflows, the value shifts away from tools that merely identify or organize problems.

The value shifts toward platforms that can close the loop:

detect → prioritize → decide → enforce → remediate.

Point tools that stop at detection or workflow management risk losing pricing power.

4. Incremental budget moves to newer control layers

The fastest-growing areas of security are likely to be:

• AI runtime protection,

• non-human identity security,

• data security posture management,

• cloud-native exposure management,

• software supply chain enforcement,

• and agent policy orchestration.

The issue is that these categories increasingly converge, for example:

• AI runtime security needs data context.

• Data security needs identity context.

• Identity governance needs behavioral context.

• Cloud posture needs runtime context.

• DevSecOps needs production feedback.

The overlap becomes the product. That overlap is difficult for narrow vendors to own without broad telemetry, distribution, and enforcement rights.

5. Multiple compression can happen before fundamentals break

This is the most important point. The basket does not require immediate revenue collapse. These companies can still grow. They can still renew customers. They can still print respectable margins. But if investors decide the terminal value of the category is lower, the multiple can compress first.

That is classic software derating:

• revenue remains real,

• product remains useful,

• customers remain sticky,

• but strategic scarcity declines.

Key Risks

1. The basket companies successfully broaden

TENB, RPD, QLYS, or FROG could expand into broader platforms themselves. They have customers, brands, data, and engineering talent. If they credibly move from narrow workflows into integrated control planes, the short case weakens.

2. Best-of-breed remains resilient

Security buyers often distrust bundles. In high-stakes environments, “good enough” may not be acceptable. If specialist depth continues to matter more than integration, point solutions can persist longer than expected.

3. M&A risk

Any of these names could become acquisition targets for larger platforms seeking customers, telemetry, developer workflows, or category credibility. That creates obvious short-squeeze risk.

4. AI creates more fragmentation, not less

Cybersecurity history often favors more tools, not fewer. If AI creates many new subcategories rather than consolidated architectures, point vendors may find new growth vectors.

5. Timing risk

Enterprise security displacement is slow. Installed bases are sticky. Compliance workflows do not disappear overnight. The thesis may be right directionally but take years to express.

Conclusion

This is an equal-weight short basket of TENB, RPD, QLYS, and FROG. The thesis is not that these are bad companies. It is not that their products stop working. It is not that cybersecurity spending declines. Rather that AI changes the architecture of security. The market is moving from fragmented point tools toward integrated control planes that combine identity, data, cloud, code, runtime behavior, and automated enforcement.

That shift favors broad platforms and compresses standalone vendors.

The most dangerous version of software disruption is not when a product becomes less scarce. That is the risk here. TENB, RPD, QLYS, and FROG may remain useful. But in the AI security era, usefulness is not enough. The premium accrues to whoever owns the control plane.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (Daicel Corporation, 4202 JP), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Daicel (4202 JP) is a cheap Japanese chemicals platform where the market is fixated on cyclical earnings, while the product portfolio quietly carries real exposure to two structurally well-funded demand pools: health-science tooling and AI-linked semiconductor / server content.

Daicel is interesting precisely because it does not look like an AI or healthcare stock on first pass. It screens as a mature chemicals name: low multiple, respectable dividend yield, low price-to-book, and earnings pressure concentrated in legacy segments. That surface-level read is directionally correct, but incomplete.

The portfolio underneath includes chiral columns and separation services used by pharmaceutical customers; health ingredients and a needle-free delivery device on the medical side; and photoresist polymers, electronic solvents, cycloaliphatic epoxies, and LCP content tied to semiconductor manufacturing and AI server hardware on the technology side.

The important nuance is that this is not a “pure-play theme stock” hiding in Japan rather it is a misclassified platform. Engineering Plastics, Materials, and Safety still anchor the P&L; which is why the market continues to underwrite the company on cyclical chemicals logic. Yet that same legacy base funds the optionality. If investors begin to value Daicel as a portfolio with a higher-quality mix, rather than as a homogenous commodity-chemical exposure, a re-rating becomes possible without requiring heroic group-level growth.

What makes the timing workable is that the supporting evidence is no longer entirely conceptual:

• In 9M FY2026, Medical / Healthcare sales rose 10.7% and operating profit rose ~75%.

• Smart segment swung from a loss to a profit.

• Company disclosures have become more explicit about AI-linked semiconductor demand and LCP use in GPU sockets and cooling fans for AI servers.

The market is still paying for a tired cyclical but the product map suggests something more interesting.

Figure 1 — Selected Valuation Metrics: Daicel still trades like a cyclical chemicals stock, not like a strategically relevant supplier to health and AI infrastructure.

Company Overview

Daicel Corporation is a diversified Japanese chemicals company with five reported businesses: Medical/Healthcare, Smart, Safety, Materials, and Engineering Plastics. The company traces its roots to cellulose chemistry and celluloid, but the portfolio now spans acetyl chain products, airbag inflators, electronic materials, functional films, engineering plastics, chiral columns, functional foods, and medical devices.

• Engineering Plastics is the largest segment by revenue and the core of the AI-infrastructure angle. Daicel fully owns Polyplastics and is integrating its engineering plastics business into Daicel effective April 2026. Polyplastics brings POM, PBT, PPS, LCP, and COC products plus global technical solution centers. The company disclosed that LCP capacity started operation in Taiwan in February 2025, and that Engineering Plastics demand includes PPS/PBT for AI servers and LCP demand in AI-server and electronic-device applications.

• The Smart segment is the semiconductor-materials vector. Daicel discloses advanced-technology products including polymers for photoresists, solvents for electronic materials and high-performance films. The company identifies increased demand in AI semiconductors and cutting-edge logic/memory as relevant to these products, while also discussing front-end and back-end semiconductor process opportunities.

• Medical/Healthcare is smaller in reported revenue, but strategically useful. It includes chiral columns/stationary phases, separation and purification services, novel drug delivery devices, cosmetic ingredients, and functional food ingredients including Equol, Urolithin, Lactobionic acid and Astrohop. The segment is not large enough today to drive the whole equity story, but it gives Daicel a credible healthcare picks-and-axes lane.

Figure 2 — Sales Mix, simplified. The legacy base is still large which is why the market continues to classify Daicel as a cyclical chemicals company.

The Setup

The setup is a classification error:

• The market sees deteriorating cyclical earnings, China exposure, capex intensity, and raw-material volatility.

• The product map shows a company with genuine exposure to the tools and materials that sit underneath two of the more durable end-markets in global industry.

Those two views are not mutually exclusive. The stock is interesting because both are true at once.

The bad news is easy to see: In 9M FY2026, group sales declined 1.8% year over year and operating profit fell 25.0%. Materials operating profit fell 45.2%, and Engineering Plastics fell 25.6%. When the profit pools that dominate the group are moving in that direction, it is unsurprising that investors refuse to award a premium multiple.

The more interesting news is hiding one layer below that. Medical / Healthcare posted 10.7% sales growth and 75.1% operating-profit growth. Safety also improved sharply. Smart swung from an operating loss to a profit. In other words, the parts of the portfolio most relevant to the “health and AI picks-and-shovels” framing are not merely conceptually attractive; they are beginning to improve in the financials as well.

Figure 3 — Operating profit trend, simplified.

The base case is not that Daicel becomes an AI pure-play. The base case is that the market stops penalizing the whole company as a structurally ex-growth chemicals collection once the quality mix becomes more visible.

Casual Mechanism

The causal mechanism is a materials-stack story. AI demand creates incremental capex in data centers, accelerators, memory, advanced packaging and high-speed electronics. That demand does not only accrue to GPU designers; it creates a pull-through for high-purity chemicals, photoresist polymers, precision connectors and heat-resistant engineering plastics. Daicel sits lower in the stack, where the unit economics are less glamorous but often more durable because qualification cycles, purity requirements and customer-specific formulations create switching frictions.

Figure 5 — Mechanism map linking AI and health demand to Daicel categories.

• The health case rests first on chiral separation. Daicel describes its optical-isomer separation business as having a world-leading position, with operations across Japan, the U.S., France, China, and India. This is better than typical healthcare equity, it is a tool-maker business. Demand is tied to pharmaceutical analysis, separation, purification, and synthesis, and management is explicitly broadening target applications from small molecules toward peptides and oligonucleotides. That is exactly the kind of exposure one wants in a picks-and-shovels framing: less dependent on a single drug outcome, more tied to the activity layer.

• The second part of the health case is smaller today, but strategically useful. Daicel has launched ASTROHOP, a health-food ingredient aimed at aging-related muscle concerns, and is also pursuing a needle-free subcutaneous drug-delivery device. Neither line needs to become a huge standalone business for the equity to work.

• The AI case is more credible than a quick screen would suggest. Through the Smart business, Daicel sells high-performance polymers for photoresists and electronic solvents used in semiconductor manufacturing. The company also highlights strong positioning in high-boiling-point solvents used as resist solvents and thinners, and explicitly references sales growth opportunities tied to AI demand. In a market where “AI exposure” is often stretched beyond recognition, this is relatively clean: these are enabling materials.

• The second layer of the AI case runs through engineering plastics. Polyplastics LCP has a leading position globally, and company materials specifically reference rising use in GPU sockets and cooling fans for AI servers. This matters because it turns the AI thesis from a pure semiconductor-fabrication story into a broader content-per-box story.

Figure 6 — Theme Exposure Matrix

Timeline

The next year is about proof. Investors simply need evidence that the financially relevant pieces of the portfolio are becoming less small, less volatile, and more legible. The May 12, 2026 result is therefore the first real forcing function. After that, the quality of management commentary around HPP integration, Smart growth, LCP demand, and medical / healthcare margins will matter more than broad thematic language.

Figure 7 — Catalyst Path

Key Risks

The right risk lens is straightforward: the thematic product lines may be real, but they may remain too small to matter at the pace the market wants. That is the main risk. The other risks are mostly variants of that same issue.

• Scale risk: Smart and Medical / Healthcare remain small relative to Engineering Plastics and Materials. Product relevance alone does not guarantee group-level rerating.

• Cyclical-earnings risk: Materials and Engineering Plastics still dominate the P&L. If those pools remain weak, investors may continue to anchor on trough optics.

• Execution risk: EUV polymers, AI-server LCP content, and medical-device programs require customer qualification and commercialization discipline. Delays would push out the thesis timeline.

• China / macro risk: Daicel itself flags uncertainty around China, raw-material costs, and competitive pricing pressure. Those factors can overwhelm segment-level progress for long stretches.

• Capex and asset-productivity risk: This is not a pristine capital-light story. The market will need evidence that rising asset intensity is being converted into better returns.

• Value-trap risk: A low P/E and P/B are only useful if the market begins to believe the quality of the portfolio is improving. Otherwise the stock can stay statistically cheap for years.

Conclusion

Daicel is a credible picks-and-axes supplier to AI and health, but the cleanest version of the thesis is not that it is a pure AI or healthcare stock. The clean version is that the market is pricing a cyclical Japanese chemical company while the product mix increasingly includes higher-value, qualification-heavy materials for AI semiconductors, AI servers, pharmaceutical workflow, functional foods and medical materials.

The stock offers a value-backed way to own that transition: low trailing multiples, a meaningful dividend yield, buyback/capital-return support, and a May 2026 guidance catalyst. The near-term hurdle is credibility. Daicel needs to show that the March 2026 COC impairment is a reset, that Materials can stabilize, and that Engineering Plastics/Smart can convert AI and semiconductor demand into margin rather than merely revenue.

The bottom line is Daicel looks like a low-expectation cyclical chemicals name. The better interpretation is a misclassified picks-and-shovels supplier to health and AI, funded by legacy cash flow and priced as if none of the optionality matters.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (Shionogi & Co., 4507 JP), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. Company Overview

3. The Setup

4. Causal Mechanism

5. Timeline

6. Key Risks

7. Conclusion

Executive Summary

Shionogi (4507 JP) used to be easy to understand: a conservative Japanese pharma company with a large HIV-linked royalty stream, high margins, and a clean balance sheet. That version of the story is now stale.

The company is using its cash-flow base to do three things at once:

• Deepen exposure to ViiV Healthcare.

• Buy a U.S. rare-disease commercial business through RADICAVA.

• Turn Fetroja / cefiderocol into a more strategic antimicrobial-resistance asset through U.S. government procurement.

The result is a re-underwriting of what Shionogi’s earnings base should be worth after the balance sheet is put to work. The issue is whether management has just spent the old balance-sheet story on assets that extend the duration of the earnings base. The next serious checkpoint is the FY2026 guide, especially the bridge from ViiV, RADICAVA, financing cost, amortization, and post-deal leverage.

Company Overview

Shionogi is a Japanese pharmaceutical company with a historical strength in infectious disease, small-molecule drug discovery, HIV, influenza, and antimicrobial resistance. The company also has exposure to CNS / pain, sleep, rare disease, and domestic Japanese prescription drugs. The current reported business still looks unusually profitable:

Figure 1 — FY2025 guide progress

At the Q3 FY2025 stage, Shionogi had already reached 72.1% of full-year revenue guidance, 80.4% of full-year operating profit guidance, and 84.2% of full-year profit-attributable guidance.

Figure 2 — Revenue mix

Prescription drugs contributed ¥86.7bn in Apr–Dec FY2025, overseas subsidiaries / exports contributed ¥48.9bn, and royalty income contributed ¥201.3bn.

Within the royalty line, the HIV franchise was ¥193.4bn. In plain terms: Shionogi is still economically anchored by HIV royalties more than by its domestic prescription-drug business.

A notable warning worth addressing: the valuation looks reasonable on the surface (~12–13x forward), but it is not obviously cheap. More importantly, the “net cash” screen may overstate balance-sheet comfort because Shionogi is using bridge financing to fund the ViiV and RADICAVA transactions.

However, that does not kill the bull case. The setup is no longer “cheap net-cash royalty compounder.” It is whether Shionogi can use its balance sheet to turn royalty income into a broader specialty pharma platform with more U.S. exposure, more owned products, and more control over future cash flows.

The Setup

The market knew Shionogi as a low-drama royalty compounder. Management is now forcing investors to decide whether it should be valued as something broader: a royalty-backed specialty pharma acquirer with more U.S. exposure and more control over its future cash-flow streams.

Figure 3 — YoY revenue growth bridge

There are three visible legs to the setup:

• First, Shionogi increased its economic interest in ViiV Healthcare to 21.7%, while GSK retained 78.3%. ViiV issued new shares to Shionogi for $2.125bn, and Pfizer exited its holding.

• Second, Shionogi completed the acquisition of global rights to RADICAVA for $2.5bn. The company says the transaction should be immediately accretive in FY2026 and add roughly $700mn in annual global sales. RADICAVA is approved for ALS, and RADICAVA ORS / IV RADICAVA have been used to treat more than 22,000 people with ALS in the U.S. to date.

• Third, Shionogi received a BARDA Project BioShield contract for Fetroja / cefiderocol. The contract is initially funded at $119mn, with multiyear options up to $482mn, and includes U.S. manufacturing, procurement, and development work for infections caused by high-priority biothreat pathogens.

Figure 4 — Disclosed 2026 strategic transaction / contract amounts

Is this disciplined reinvestment or balance-sheet dilution dressed up as strategy? Below is the pro-forma FY2026–FY2028 net debt / EBITDA and EPS accretion after ViiV, RADICAVA, JT / Torii, interest expense, amortization, and dividend policy.

Revenue steps up to ~¥720bn by FY2028 on RADICAVA, Torii, and ViiV, with operating profit ~¥278bn and EBITDA ~¥323bn; EPS reaches ~¥301, still lagging EBITDA as interest and amortization come through after the ~$4.5bn of deals. The shape here points to disciplined reinvestment: earnings scale and if leverage trends down and cash flow follows into FY2028, the step-up holds.

Causal Mechanism

The mechanism is simple enough to write on one line:

HIV cash flow funds control-like economics and U.S. commercial capacity; if those acquired assets produce stable earnings, Shionogi’s multiple should attach to a longer-duration base, rather than a fading royalty stream.

• The ViiV leg is the cleanest. GSK reported 2025 HIV sales of £7.687bn, up 11% CER, driven by Dovato, Cabenuva, and Apretude. Long-acting medicines contributed more than 75% of total HIV growth in 2025, with Cabenuva contributing 55%. Cabenuva sales reached £1.402bn, up 42%, and Apretude sales were £439mn, up 62%. So Shionogi’s HIV economics are not just legacy royalty tailwind. The product mix is shifting toward long-acting treatment and prevention. It’s likely the HIV line can remain durable for longer than a “mature royalty” model implies.

• The RADICAVA leg is different. It is not just another product line. It gives Shionogi an operating footprint in U.S. rare disease. The company’s December acquisition announcement explicitly framed the deal as a way to create a U.S. rare-disease commercial platform that can support future launches, including Fragile X syndrome, Jordan’s Syndrome, and Pompe disease programs.

• The AMR leg is smaller but strategically useful. Antibiotics are usually difficult businesses because stewardship caps volume and reimbursement is uneven. Government procurement changes the shape of the economics. BARDA is not a guarantee of a blockbuster. It does, however, make Fetroja more than a hospital-sales story; it becomes part of a national preparedness budget.

• The JT / Torii transaction rounds out the domestic base. Shionogi agreed in 2025 to succeed Japan Tobacco’s pharmaceutical business, acquire Akros, and make Torii a wholly owned subsidiary. JT’s pharmaceutical business had fiscal 2024 net sales of ¥44.942bn. The strategic logic is not glamorous, but it is coherent: more small-molecule discovery capacity, more domestic breadth, and more non-infectious-disease exposure.

The whole story depends on one discipline test: acquired earnings must be good enough to offset the loss of balance-sheet purity.

Timeline

At this point, everything is in motion, but the next guide and actual execution is the real event. May 12 is the next report. If the FY2026 bridge is clean, the story re-rates. If not, it stalls.

Key Risks

1. The balance sheet has changed

Shionogi disclosed ¥300bn of bridge borrowing for the ViiV investment and said total bridge loan financing reached ¥660bn when combined with the RADICAVA-related bridge loan. The company plans to refinance into mid- to long-term financing by the repayment deadline, and said FY2027 impact was still being examined. The stock’s old appeal included excess cash and low financial complexity. The new appeal requires confidence in management’s capital allocation.

2. RADICAVA must be more than acquired revenue

RADICAVA adds scale and U.S. rare-disease infrastructure, but ALS is a hard market. The acquisition price is $2.5bn, and Shionogi may pay additional royalties on future sales. The deal can work, but it needs durable sales and follow-on use of the commercial infrastructure. Otherwise, the market will treat it as bought growth rather than strategic extension.

3. Xocova / respiratory revenue is volatile

The acute respiratory virus treatment line fell from ¥43.3bn to ¥27.3bn in Apr–Dec FY2025, down 37.0%. That includes Xocova, Xofluza, and Rapiacta. This is the clean bear argument against over-capitalizing Shionogi’s infectious-disease optionality: outbreak-linked revenue is not the same as a durable annuity.

4. AMR is strategically important but commercially awkward

The BARDA contract is positive, but antibiotics remain structurally unusual assets. Stewardship limits volume by design. Government procurement helps, but it also means timing, funding, and manufacturing obligations matter. Fetroja also carries a label history that includes an observed increase in all-cause mortality in certain critically ill patients in a study versus best available therapy; Shionogi’s own U.S. safety language flags this issue.

5. The next guide can disappoint even if the strategy is right

FY2026 guidance has to absorb multiple moving parts: ViiV accounting, RADICAVA sales, financing cost, amortization, JT / Torii consolidation, U.S. SG&A, and dividend policy. A messy bridge could delay the re-rating even if the long-term logic is sound.

Conclusion

The old Shionogi thesis was balance-sheet quality plus HIV royalties.

The new Shionogi thesis is capital allocation.

That makes the stock more interesting and less forgiving. The company is no longer just collecting from ViiV and paying dividends. It is increasing its ViiV exposure, buying U.S. rare-disease infrastructure, consolidating domestic assets, and pushing antimicrobial resistance into a procurement-backed model.

The strongest version of the thesis is straightforward: ViiV’s long-acting HIV cycle extends the royalty base, RADICAVA adds near-term sales and U.S. commercial capability, and BARDA turns Fetroja into a strategic asset with government demand. In that version, the market moves away from viewing Shionogi as a static royalty story and begins valuing it as a higher-duration specialty pharma cash-flow compounder.

The weakest version is just as clear: management spent a pristine balance sheet at the wrong time, Xocova fades, RADICAVA under-earns, and the pro-forma guide forces investors to model more debt, more amortization, and less simplicity.

The May 12 guide will be very important because it will show which version investors have to underwrite.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (LXP Industrial Trust, LXP), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Why This Story Is Different

2. Start With What Already Works

3. The Real Story Is the Development Spread

4. Phoenix Is the Bridge From Optionality to Reported Earnings

5. Why This Deserves More Credit Than a Typical Land Narrative

6. The Balance Sheet Makes the Optionality Actionable

7. What Has to Happen for the Stock to Work

8. What Would Break It

9. Bottom Line

Why This Story Is Different

LXP is not a direct AI stock. But it may be an indirect AI and MAGA/protectionist winner: its Phoenix land bank sits in the path of semiconductor reshoring, advanced manufacturing migration, and a broader policy push to localize strategic supply chains.

Most REITs are not transformation stories rather they are cash-flow vehicles. Investors buy current income, modest rent growth, balance-sheet discipline, and some NAV accretion over time. LXP is more interesting because it may no longer fit neatly inside that box.

The company today looks like a credible, stabilized industrial landlord. But inside that stabilized portfolio sits something more valuable than “extra land.” It sits on a development pipeline that could convert dormant carrying value into future NOI, NAV creation, and a better growth profile than a standard industrial REIT screen implies.

There is also a timely macro layer here that makes the story more relevant than a normal land-bank pitch. LXP is not an AI stock in the conventional sense, but it is arguably an indirect beneficiary of the same capex cycle. Management has pointed to advanced manufacturing tailwinds tied to TSMC and semiconductor supply-chain migration into Phoenix.

That makes LXP a second-order AI infrastructure and domestic industrial-policy beneficiary because it owns land in a market that may increasingly absorb the warehousing, logistics, and supplier footprint created by semiconductor reshoring and protectionist supply-chain realignment. This is not really a “land story.” It is a land-to-NOI conversion story with a potentially favorable macro backdrop.

Start With What Already Works

• As of year-end 2025, the company owned 108 properties totaling 52.7 million square feet. Roughly 92.6% of square footage was Class A. The portfolio was 97.1% leased with a weighted-average lease term of 4.8 years. It also controlled 514 acres of developable land. The stock does not require a heroic view on development to be investable. The base business is already a modern industrial platform with solid occupancy and a reasonably clean portfolio.

• The internal earnings bridge also looks better than a simple REIT screen may suggest. Management’s fourth-quarter presentation notes that leases expiring through 2030 carry an estimated current mark-to-market of roughly 16%, enough to increase initial annual cash rent by about $29 million, or roughly $0.48 per share, equal to about 15% of FFO. The same presentation says LXP has 1.5 million square feet available for lease, which could add another roughly $16 million of initial annual cash rent, or about $0.28 per share. Layer in average annual rental escalations of 2.8%, with 2025 signed leases carrying average annual escalations of 3.2%, and the existing portfolio already has credible internal growth. Even before the development pipeline contributes meaningfully, LXP has a clear path to higher rents, better occupancy, and incremental earnings.

The Real Story Is the Development Spread

The market usually values REITs by capitalizing current or near-term NOI. That works well for stabilized portfolios. It works poorly when part of the asset base is held in land that generates little current income but can be developed into higher-value industrial product. In those cases, book value can be a misleading anchor. The relevant economic question is not what the land is worth sitting still. It is what the company can earn by building on it.

LXP’s filings show land held for development at roughly $82.97 million on the consolidated balance sheet, with another $11.47 million in non-consolidated land projects. The consolidated land table shows 315 acres in Phoenix carried at about $75.4 million. But the heart of the thesis is whether LXP can convert that land into stabilized industrial product at returns above what the market will ultimately pay for those assets.

If LXP can consistently develop industrial assets at 7%-plus stabilized cash yields and the market values comparable stabilized product at materially tighter cap rates, the value creation is mechanical. The land bank becomes a pipeline of future NAV accretion rather than a passive balance-sheet line item.

Phoenix Is the Bridge From Optionality to Reported Earnings

Phoenix is the clearest expression of that mechanism.

In its 2025 annual letter, management says the company’s 514-acre land bank includes 315 acres in Phoenix. It has already broken ground on a 1.2 million square foot speculative development project there with projected completion in the first half of 2027, an estimated budget of $120 million, and a stabilized cash yield expected in a range of 7.0% to 7.5%. More importantly, management says the remaining 240 acres at the site can support up to four million square feet of industrial development.

This is a multi-year inventory of future NOI.

Phoenix matters for another reason: it is where the macro overlay meets the financial model. Management notes that construction costs are about $20 per square foot below peak levels and points to advanced manufacturing tailwinds tied to TSMC and semiconductor supply-chain migration into Phoenix. That makes the project more than a generic Sun Belt warehouse build. It potentially places LXP inside a broader reshoring and industrial-policy current that has both AI and protectionist characteristics. AI does not only require chips and data centers. It also requires domestic manufacturing ecosystems, supplier networks, and logistics infrastructure. Phoenix is becoming one of those ecosystems.

The key point is not that every macro tailwind must fire perfectly. It is that LXP appears to control land in a market where large-box industrial demand may remain strategically supported, and where the economics of building today look better than they did at peak cost.

As long as land remains land, upside stays abstract. Once the first building is delivered, leased, and stabilized, the market has to stop treating the site as dormant optionality and start treating it as an earnings pipeline. Optionality only rerates when it becomes legible in reported results.

Why This Deserves More Credit Than a Typical Land Narrative

Many companies can point to undeveloped acreage. Far fewer can show an operating history of converting land into value. That is why LXP’s execution record matters so much here.

Management says it has constructed 15 facilities since 2019 at a 7.1% weighted-average stabilized yield on first-generation leases and generated $91 million of sales proceeds in excess of cost basis. It also says that in 2025 it sold the Indianapolis and Ocala development properties for $175 million to a user buyer at an implied 5% capitalization rate and a 20% premium to initial cost basis:

That is a highly important proof point. It suggests the upside is not merely “Phoenix land is probably worth more than book.” The stronger claim is that LXP has already shown it can build at one yield and monetize at another. That spread is what turns development from a story into a capital-allocation engine. In other words, the bull case is essentially repeated spread capture.

The Balance Sheet Makes the Optionality Actionable

Optionality in REITs is useless if the balance sheet cannot support it. LXP looks more interesting now because that constraint appears to be easing.

At year-end 2025, the company reported net debt to Adjusted EBITDA of 4.9x after selling assets, repaying debt, and reducing exposure to non-target markets. In the annual letter, management explicitly says that with its key 2025 objectives achieved, 2026 capital allocation priorities will focus on development opportunities, primarily within the existing land bank, while still leaving room for opportunistic buybacks.

That statement is more important than it sounds. It marks a shift from portfolio repair and simplification toward deliberate capital deployment. The thesis strengthens materially once investors believe LXP has both the land and the financial flexibility to start converting it. The distinction here matters:

• A land bank on a stressed balance sheet is deferred optionality.

• A land bank on a repaired balance sheet is potential growth inventory.

What Has to Happen for the Stock to Work

The mechanism is straightforward:

• First, the existing portfolio continues to compound through rent mark-to-market, escalators, and lease-up. That supports earnings growth even without any heroic development assumptions.

• Second, the cleaner balance sheet allows capital to move toward development inside the land bank, especially Phoenix.

• Third, development starts become delivered buildings, delivered buildings become leased buildings, and leased buildings become stabilized NOI. At that point, the market no longer has to value those acres as dormant land.

• Fourth, as the contribution becomes visible, LXP can begin to look less like a plain industrial landlord and more like an industrial REIT with an internal development engine. That is the point where the valuation framework can shift.

That chain is the entire thesis:A sequence of conversion from land to buildings to NOI to NAV.

What Would Break It

The biggest risk is that the spread never gets realized:

• If industrial demand softens, speculative leasing stalls, construction costs re-accelerate, or capital becomes less available, the land bank can remain underutilized for longer than bulls expect. In that outcome, the acreage still exists, but the timing of value realization stretches materially.

• There is also clear concentration in the upside case. Phoenix is the crown jewel. If Phoenix performs, the thesis becomes more obvious. If Phoenix disappoints, the stock may still work as a decent industrial REIT, but the argument for material upside weakens.

Finally, rerating is not automatic. A REIT can create real value and still trade as if it were only a current-income vehicle for longer than investors expect. The market often requires repeated proof before it changes the box it puts a company in.

Bottom Line

LXP is a stabilized industrial REIT whose current earnings base may understate the value of a real internal development pipeline.

• The existing portfolio already supports ownership through rent growth, lease-up, and a healthier balance sheet.

• The Phoenix land bank is what adds asymmetry. And that asymmetry is not happening in a vacuum: Phoenix is also exposed to semiconductor reshoring, advanced manufacturing buildout, and a more protectionist domestic supply-chain regime, which makes LXP an indirect beneficiary of both the AI capex cycle and the political push to localize strategic production.

• If management keeps converting low-carry-value land into 7%-plus stabilized industrial product, and if the market continues to value finished assets at materially tighter cap rates, then LXP has a credible path to create more NAV and more future NOI than a standard REIT multiple implies today.

In short LXP is a landlord with a visible path to turn land into earnings in a market increasingly shaped by AI infrastructure demand and industrial reshoring.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (AerSale Corp, ASLE), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

• Executive Summary

• Company Overview

• The Setup

• Causal Mechanism

• Timeline

• Key Risks

• Conclusion

Executive Summary

AerSale (ASLE) is in the middle of a business-quality transition: from a volatile “asset trader” (whole aircraft/engine sales) toward a more recurring, higher-margin aviation aftermarket platform where TechOps (MRO + proprietary engineered solutions) becomes the primary earnings driver. The inflection is already visible in reported results: in Q3’25, the company delivered no aircraft or engine sales yet still expanded profitability, with management explicitly citing TechOps margin expansion from 13.6% (Q3’24) to 25.3% (Q3’25) as the segment mix and work scope shifted toward higher-margin opportunities.

• The near-to-medium term forcing function is AerSafe, a proprietary, FAA-STC-backed engineered solution tied to an FAA compliance deadline in Q4 2026. As aircraft rotate through maintenance windows, AerSafe deliveries are positioned to remain elevated into 2026.

• Simultaneously, AerSale is coming off a multi-year facilities buildout: management states construction is complete on expansion projects and guided (on the Q3’25 call) to ~$25M of revenue and ~$4–5M of EBITDA contribution in 2026 from the three expansion projects as they ramp into production.

What must be proven for the thesis to compound:

• TechOps margins remain structurally higher even when whole-asset sales are muted .

• Engineered solutions scale without pulling forward all demand into a single “deadline quarter”.

• The company converts its substantial inventory base into cash flow as the model becomes more capital-light.

Company Overview

AerSale is an integrated aviation aftermarket provider organized into two reportable segments:

• Asset Management Solutions: acquires flight equipment and monetizes it through leasing, trading, or disassembly into used serviceable material (USM).

• TechOps: provides MRO (aircraft/component maintenance, landing gear, thrust reversers, aerostructures, modifications/cargo conversions, storage) and sells internally developed engineered solutions permitted by FAA Supplemental Type Certificates (STCs) (proprietary, non-OEM solutions to airworthiness directives/technical challenges).

In Q3’25, AerSale reported total revenue of $71.2M, split between $39.2M Asset Management Solutions and $32.0M TechOps. The quarter also highlighted the company’s deliberate pivot away from pure trading volatility: no aircraft or engines were sold in Q3’25, versus five engines in the prior-year quarter.

The Setup

• “Quality of revenue” is improving even as headline revenue remains volatile.
  While reported revenue can fluctuate with the timing of aircraft and engine transactions, the company’s recurring businesses, USM parts, leasing, MRO services, and proprietary engineered solutions, are becoming a larger share of activity, supporting a more stable earnings profile over time.

• TechOps is the margin inflection point, already happening.
  Management attributed the step-change in TechOps profitability to refocusing MRO facilities on higher-margin work and to engineered solutions volume/mix. Operationally, Roswell has been repositioned toward teardown/decommissioning work (higher-margin scope), while other shops (aerostructures/landing gear) and AerSafe helped offset softer services at certain facilities.

• Capacity expansion is shifting from “build” to “harvest.”
  Management stated construction is complete on expansion projects and the company is transitioning into operations. Separately, AerSale opened a new ~90,000 sq ft aerostructures MRO facility in Hialeah Gardens, FL, described as ~3× larger than the prior location, with upgraded equipment supporting larger widebody structures (B777 nacelles/thrust reversers, GE90/Trent platforms).

• Capital allocation has already cleared the private equity overhang.
  In March 2025, AerSale repurchased ~6.4 million shares at $7.00 per share (~$45M) from its former sponsor, Leonard Green & Partners, and immediately retired them. The transaction effectively completed the sponsor’s exit and removed a recurring technical headwind that often weighs on asset-heavy businesses with private equity ownership. While not an operational catalyst, the removal of a large financial sponsor reduces potential share supply and clarifies the company’s transition into a fully public shareholder base.

Causal Mechanism

The thesis is a stacked mechanism where multiple moving parts reinforce each other:

Mechanism A: Regulatory-driven engineered solutions + maintenance cycles → accelerating high-margin TechOps mix.

• AerSale stated it saw a strong increase in AerSafe deliveries YoY and expects volume to remain elevated through 2026 as the company approaches an FAA airworthiness directive compliance deadline (satisfied by AerSafe installation). At quarter-end, management disclosed that 2025 AerSafe deliveries + backlog totaled >$22M. Management also stated AerSafe should support results through the regulatory compliance deadline in Q4 2026.

• Because engineered solutions are proprietary products sold under FAA STCs, the incremental margin profile is structurally better than labor-heavy commodity MRO (and requires meaningfully less incremental capex once certification and tooling are in place).

Mechanism B: “Build is done” facilities expansion → capacity + capability unlock in 2026.

• In Q&A, management quantified the expansion projects: ~$25M of revenue and ~$4–5M of EBITDA in 2026, stepping toward a higher run-rate at full capacity.

• This contribution is powerful because it is incremental to the pre-expansion base and intended to be more recurring than whole-asset sales, reducing consolidated earnings volatility.

Mechanism C: Leasing + USM monetization → smoother revenue and better operating leverage.

• The company is explicitly shifting its deployment strategy toward assets on lease to create more stable quarter-to-quarter performance (leasing additional converted freighters/engines).

• The financial statements show this operationally: AerSale disclosed reclassification of inventory to equipment held for lease (a signal that inventory is being pushed into recurring lease-earning assets rather than one-time sales).

Mechanism D: Capex taper (post-buildout) + working-capital discipline → potential inflection in free cash flow.

• Cash-flow disclosures are directionally consistent with capex moderating as major buildouts complete. Management has also framed working-capital optimization as part of building a “more resilient business model,” but the proof will be in reported operating cash flow as inventory turns into USM sales, lease revenue, and TechOps throughput.

Timeline

Near-term: FY2025 results (March 5, 2026).
AerSale announced it will report Q4 and full-year 2025 results on March 5, 2026 (after market close), with a conference call the same day. This is the first opportunity for investors to pressure-test whether Q3’s margin profile persists and whether 2026 ramp commentary holds.

2026: “Harvest year” for expansions + AerSafe pre-deadline acceleration.
Key 2026 checkpoints:

• H1 2026: early evidence of expansion project ramp, utilization, and whether TechOps margins remain elevated. (Management’s quantified expectation for 2026 contribution is a direct benchmark.)

• H2 2026: engineered solutions should remain strong as operators work through compliance timelines; management has explicitly anchored AerSafe support through Q4 2026.

Post-deadline (2027 setup): durability question.
After the compliance deadline, the debate becomes: is AerSafe predominantly pull-forward (a “one-time compliance wave”) or does it settle into a multi-year annuity-like stream (replacement/maintenance + broader adoption)? The company’s longer-term narrative also depends on whether AerAware converts from “engineering asset” to “commercial product” via a launch order.

Key Risks

• Used-serviceable material supply loosens faster than demand.
  If OEM delivery rates and retirements normalize faster than expected, USM pricing and demand intensity can soften, pressuring the Asset Management segment and lowering returns on acquired feedstock. This matters because AerSale’s balance sheet is heavily inventory-weighted.

• AerSafe “demand cliff” after the Q4’26 compliance deadline.
  Management is clear that AerSafe is tied to an FAA compliance window through Q4’26. If demand collapses post-deadline and AerSale does not scale additional engineered solutions, TechOps margins could revert toward historical (lower) levels and the “recurring, high-margin TechOps” narrative weakens.

• Whole-asset sales remain structurally lumpy (earnings optics risk).
  AerSale repeatedly emphasizes that whole-asset sales drive quarter-to-quarter volatility; Q3’25 is a reminder that reported revenue can swing materially depending on whether aircraft/engine sales occur. Even if the long-term shift toward leasing/TechOps works, near-term results can look noisy.

• Working capital / inventory monetization risk.
  The company disclosed $371.1M of inventory and cash used in operations year-to-date of $34.3M, reflecting inventory investment. If inventory turns slow (or pricing deteriorates), leverage rises and cash conversion can disappoint.

Conclusion

AerSale’s thesis is best framed as a margin + mix + cash-conversion transition. The company has already demonstrated that profitability can improve even in a quarter with zero whole-asset sales, driven by higher TechOps margins and a pivot toward leasing/USM monetization. The next 12–18 months are about validating durability: AerSafe demand into the Q4’26 compliance deadline, the 2026 ramp of expansion projects (with disclosed revenue/EBITDA expectations), and tangible improvement in operating cash flow as inventory is monetized.

A secondary dimension of the story is AerSale’s position as a sub-scale but integrated platform in a consolidating aviation aftermarket industry. The company combines aircraft and engine feedstock sourcing, USM parts distribution, MRO capabilities, and proprietary engineered solutions, capabilities that are increasingly being assembled into larger platforms by strategic buyers. While not required for the thesis, the company’s integrated capabilities and regulatory certifications create potential strategic value if consolidation in the aerospace aftermarket continues and could potentially be an acquisition target.

Affiliation Disclosure & Disclaimer:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein (Kioxia Holdings, 285A JP), which could influence the views expressed. This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

• Executive Summary – A Different NAND Cycle
  Mix shift, improved earnings quality, and balance sheet repair within a tight supply backdrop.

• Company Overview – Enterprise-Weighted Model
  Revenue composition, manufacturing control, and the Gen8 transition.

• The Setup – Operating Leverage and Cash Conversion
  Price and volume strength, inventory discipline, and revised JV economics.

• Causal Mechanism – Earnings Path Upgrade
  Enterprise mix expansion, supply discipline, node transition, JV compensation, and cash compounding.

• Timeline
  JV extension, Gen8 production crossover, enterprise SSD qualification, and CY2026 supply outlook.

• Key Risks
  Cycle normalization, AI demand variability, consumer weakness, execution, accounting transition, and FX.

• Conclusion – Structural De-Risking
  Less about peak margins, more about sustained earnings quality and capital structure improvement.

• Special Note:
  KIOXIA perpetual contract expected to launch.

Executive Summary

Kioxia (285A JP) is operating from a position of strength in a genuinely tight NAND backdrop. Pricing is firm, bit shipments are expanding, and the operating leverage is translating directly into cash. Q3 FY2025 revenue reached a record ¥543.6bn, non-GAAP operating income was ¥144.7bn (27% margin), and free cash flow was +¥85.7bn, the eighth consecutive quarter of positive FCF. Net debt-to-equity has been reduced to ~80%, materially improving the balance sheet profile. But the more important point isn’t really that “NAND is strong.” It’s more what Kioxia is doing with the strength and how it could exit this cycle.

Two things stand out:

1. Enterprise SSD is now the center of gravity. In SSD & storage, management described the mix as roughly ~60% data center/enterprise and ~40% PCs/other, with data center/enterprise volume and revenue both at record highs (AI server demand plus higher unit prices).

   

2. The SanDisk JV economics have been rewritten in Kioxia’s favor. Kioxia will receive $1.165bn in cash consideration over 2026–2029 (about $290m/year on average) and recognizes ~$131m/year of additional revenue from Feb 2026–Dec 2034 and management explicitly said this revenue contributes directly to operating profit. Historically, ~¥200bn/year of SanDisk-related revenue was effectively booked “at cost.” Simply put this changes the quality of earnings.

   

Guidance is aggressive and importantly specific. For Q4, revenue is guided to ¥845–935bn and non‑GAAP operating profit to ¥440–530bn, with unit pricing expected to increase significantly across applications, total shipments down, and data center/enterprise SSD mix up. Full-year FY2025 guidance calls for record revenue (¥2.1798–2.2698tn) and record non‑GAAP operating income (¥717–807bn). The trade is simple to state: if the company is right that CY2026 stays tight, then Kioxia is easily upgrading mix, upgrading earnings quality, and de‑risking the balance sheet at the same time.

Company Overview

Kioxia is a NAND manufacturer with three practical revenue buckets: Smart devices, SSD & storage, and Other (which includes B2C products like SD cards and revenue from sales to SanDisk recorded via the manufacturing JVs). Management noted all three categories delivered record revenue in Q3.

In Q3 FY2025:

• Smart devices (smartphones/tablets): ¥186.3bn, 34% of total revenue, up 18.4% QoQ, helped by the Gen8 BiCS FLASH transition that “began in earnest” in Q2 and continued through Q3.

• SSD & storage: ¥300.4bn, 55% of total revenue, up 22.8% QoQ. Within this, management pegged the split at roughly 60% data center/enterprise / 40% PCs & other.

That 60/40 call matters because it’s management telling you what’s driving the margin profile: the incremental bit demand is moving toward enterprise sockets (qualification, endurance, performance-per-watt, controller/firmware tuning), not just consumer clearing prices.

On operations, they’re making a point of control: Kioxia says it controls 100% of wafer manufacturing at the Yokkaichi and Kitakami plants and independently manages procurement, automation optimization, cycle-time reduction, lead-time reduction, and yield management. That’s the “why” behind the confidence in investment efficiency and manufacturing competitiveness.

The Setup

• In Q3 both price and volume moved higher. Unit pricing firmed again sequentially, but just as importantly, gigabyte shipments expanded, led by data center and enterprise demand. In memory, simultaneous price and volume strength is usually a signal of genuine supply tightness rather than short-term channel behavior.

• Free cash flow remained solidly positive, leverage continued to decline, and the balance sheet is materially stronger than it was a year ago. Inventory trends tell a similar story. Days declined meaningfully, consistent with demand running ahead of supply and disciplined production execution.

• Layered on top of the operating momentum is the revision to the SanDisk joint venture. What had historically been revenue recognized largely at cost now incorporates explicit consideration and margin-accretive recognition over time.

• Guidance maintains the same posture: firm pricing, improving enterprise mix, and further profitability expansion.

Causal Mechanism

• The first mechanical shift is mix. AI workloads are pulling NAND demand toward enterprise and data center sockets, where qualification cycles are longer, density requirements are higher, and product selection is less price-elastic. Management’s view is straightforward: data center and enterprise demand for AI applications is driving NAND growth, and near-term demand is expected to exceed supply into CY2026.

• Importantly, the demand base is not singular. Enterprise growth is supported by three overlapping vectors: ongoing conventional server replacement, sustained AI inference deployment, and supply constraints in nearline HDD that encourage substitution into high-capacity QLC SSDs. Even if AI sentiment cools tactically, HDD availability and density economics can continue to shift bits toward enterprise SSD adoption. The demand story is therefore broader than a single AI capex cycle.

• Demand alone, however, does not create earnings durability. Supply discipline does. The company expects market bit growth to remain in the mid-teens this year and upper-teens next year, while acknowledging ongoing capacity constraints. The implication is that the industry is not responding with aggressive capacity additions that would immediately compress pricing. Tightness converts mix into margin.

• Overlay the product cycle. The transition to 8th-generation BiCS FLASH is progressing, with Gen8 expected to become the mainstay production node by fiscal year-end, while investment in subsequent generations continues. That sequencing is critical. Cost reductions and competitive positioning are advancing during the upcycle, not deferred to a downturn. It provides margin defense when pricing inevitably moderates.

• Then there is the structural revision to the SanDisk joint venture. The shift toward explicit consideration and margin-accretive recognition through 2034 reframes part of what had historically been throughput economics. Management positions this as reflecting the strategic and technological value embedded in Kioxia’s manufacturing and development capabilities.

• The practical result is improved cash visibility and higher-quality earnings. Higher-quality earnings compound through the balance sheet. Net D/E has already compressed meaningfully over the past several quarters, driven by sustained free cash flow and retained profitability. In prior cycles, leverage amplified volatility when conditions softened. Today, improving capital structure reduces the reflexive downside that has historically characterized memory equities.

The mechanism, therefore, is layered and put simply what changes the earnings path is:

Enterprise mix expansion → tight supply backdrop → node transition supporting cost structure → improved JV economics → sustained cash generation → balance sheet repair.

Timeline

• Jan 30, 2026: Yokkaichi JV extended through end‑2034; $1.165bn consideration over 2026–2029; revenue recognition Feb 2026–Dec 2034 (~$131m/year).

• End of CY2025: shipment of 122TB and 245TB SSD models for customer qualification (high‑capacity 2Tb QLC on Gen8 BiCS FLASH).

• End‑March 2026: Gen8 expected to surpass Gen5 in production ratio and become the mainstay product.

• Q4 FY2025: IFRS requires lump-sum recording of ¥12bn in property taxes for the full year (recorded in Q4); management says profit growth is expected to “significantly” exceed that headwind.

• CY2026 (company view): NAND supply/demand expected to be very tight in the near term, driven by data center demand.

Key Risks

The NAND cycle risk doesn’t go away.
This is still a commodity industry at the core. The real risk is supply responding faster than expected (or demand pulling forward so hard that you get a digestion pocket). Management’s own strength in Q4 guidance implicitly raises the question: is this a peak quarter? Obviously not our view but still a reasonable question.

AI inference demand can wobble.
Kioxia is leaning on sustained AI inference and AI server demand as a driver for record data center/enterprise results. If hyperscalers pause deployments, or storage attach rates disappoint, enterprise SSD mix can normalize faster than investors expect.

Consumer sets are a real watch item.
Management flagged that in smartphones and PCs, there’s a shift toward higher-end AI models, but total sets may decline due to rising costs. If that turns into outright unit weakness, it can still pressure the broader NAND pricing umbrella.

Execution on Gen8 ramp.
The company expects Gen8 to become the mainstay product by end‑March 2026. Any yield, utilization, or transition inefficiency shows up quickly in gross margin, especially if ASP momentum decelerates.

JV optics / accounting noise.
Moving from “revenue at cost” toward explicit compensation improves economics, but it can create reporting quirks and confusion for generalist investors. The risk is the market misreads the income statement during the transition period.

FX and financing sensitivity.
Q3 benefited from FX tailwinds (JPY 153/USD vs 147 in Q2), and refinancing reduced interest cost. Both can reverse.

Conclusion

The easy take is “NAND upcycle.” The better take is that Kioxia is using the upcycle to change its shape.

• Enterprise SSD is becoming the profit engine (and management is talking in enterprise terms: qualification, high-capacity QLC, data center sockets).

• The SanDisk JV revision adds a contractual layer of cash consideration and margin-accretive revenue recognition that should make earnings less purely spot-driven over time.

• Cash generation is not a slide-deck claim anymore, eight straight quarters of positive FCF and a net D/E ratio down to ~80% is a tangible de-risking path.

What needs to stay true for the thesis to work is also clear: pricing can’t roll over immediately, enterprise mix needs to hold, and Gen8 needs to ramp cleanly into mainstay production by fiscal year-end. If those three hold while the JV cash begins flowing (2026–2029), Kioxia exits this phase with a materially improved balance sheet and a higher-quality earnings base, exactly the combination that changes how cyclicals get valued.

Special Note:

Trade.xyz secured the KIOXIA listing via the HIP-3 auction, committing 515.64 HYPE (~$24k USD), which is burned upon settlement. At current pricing, this is a non-trivial expenditure for a single ticker, particularly with the perpetual not yet live.

Affiliation Disclosure:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein, which could influence the views expressed.

Disclaimer:

This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

1. Executive Summary

2. View

3. Why Look at This Now

4. What Long Rates Reflect

5. Productivity and Uncertainty

6. Why Short-Term Inflation Noise Matters Less

7. Why Duration Is the Correct Expression

8. Basket Expression

9. Optional Convexity

10. Why This Framework Exists

11. When This Would Change

12. Closing

Executive Summary

Markets appear to be pricing long-run inflation uncertainty and term premium too pessimistically at a time when productivity pressures, first visible in software and increasingly across services, are beginning to stabilize costs. Long-dated Treasuries are the cleanest way to express a repricing of that uncertainty, benefiting from term-premium compression even without deflation or near-term inflation relief.

View

The market may be overpricing long-run inflation and term premium in an environment where productivity is gradually improving. This note is not about near-term inflation prints or calling a macro inflection point. It’s about how long-dated assets are priced relative to plausible long-run outcomes, and whether that pricing still reflects the direction of underlying cost pressures.

Why look at this now

Inflation debates have become largely binary: either inflation is structurally higher forever, or it is about to collapse. Markets do not require that level of certainty to reprice. They move when the range of plausible outcomes narrows. What has changed recently is not inflation data, but where productivity pressure is becoming visible.

Markets have begun to recognize that AI is already pressuring software and digital services pricing. What initially looked like a margin expansion story is increasingly showing up as:

• higher output per employee,

• slower headcount growth,

• rising competition within services categories,

• and, in some cases, explicit pricing pressure.

That recognition matters because software and services have been among the most inflation-resilient parts of the economy over the past decade. Long-duration assets are especially sensitive to shifts in long-run uncertainty like this. Their pricing depends less on the next year and more on how investors assess cost stability over the next decade or two.

The question is straightforward: is the market still demanding too much compensation for long-run inflation and duration risk given these developments?

Chart 1: Long-run inflation expectations vs realized inflation (US 5y5y breakeven, 10y breakeven, and CPI YoY)

The chart below shows market-implied long-run inflation expectations alongside realized CPI. Despite large swings in headline inflation over the past two years, forward breakeven rates have remained contained, oscillating within a relatively narrow range. This matters because long-duration assets are priced off expectations about the next decade, not the last twelve months. The stability of long-run breakevens suggests that, while near-term inflation noise has been elevated, the market has not materially revised its view of long-run inflation. To the extent term premium has risen alongside nominal yields, it reflects uncertainty pricing rather than a clear shift in long-run inflation beliefs.

What long rates reflect

A long-dated Treasury yield embeds three components:

• expected future short-term rates

• expected long-run inflation

• term premium, which compensates investors for uncertainty over time

Most macro commentary focuses on the first two. This allocation is focused on the third. Term premium tends to rise when inflation outcomes feel unstable and cost structures feel fragile. It tends to compress when investors gain confidence that the long-run environment is more predictable, even if growth is uneven.

Chart 2: U.S. 10-Year Treasury Term Premium (Adrian-Crump-Moench model)

With long-run inflation expectations largely stable, elevated long-end yields increasingly reflect a higher term premium, compensation for uncertainty rather than a repricing of the inflation path.

Chart 3: U.S. 10-Year Real Yield (TIPS)

With real yields already elevated, changes in long-end pricing increasingly reflect shifts in term premium rather than expectations for stronger real growth.

The remaining question, then, is what could plausibly reduce this uncertainty over time?

Productivity and uncertainty

Productivity matters here not because it guarantees lower prices, but because it affects the stability of the system.

Incremental productivity improvements can:

• reduce marginal cost pressure,

• weaken pricing power over time,

• limit the persistence of wage-price feedback,

• narrow the range of long-run inflation outcomes.

Importantly, this process does not stop at software.

If digital services are the first place where productivity gains are visible, physical services are the logical next step as automation, robotics, and humanoid labor systems mature. Even partial substitution in labor-intensive services can materially change long-run cost dynamics without eliminating jobs outright.

None of this requires deflation. It does not require immediate or universal adoption. It requires only that the long-run cost structure of services becomes modestly easier than currently assumed. If that occurs, long-run inflation expectations and term premium can drift lower without any dramatic macro event. That is sufficient for long duration to reprice.

Chart 5: Software and IT Services Operating Margins (S&P 500 Software vs IT Services)

Operating margins in software and IT services have essentially stopped expanding despite continued demand. This is consistent with productivity gains translating into competitive and pricing pressure rather than sustained margin capture. Given software’s historical inflation resilience, early margin pressure here increases the likelihood that productivity effects eventually extend into broader services.

Chart 6: Services inflation: official vs real-time

The chart compares official measures of services inflation with a real-time price index. While the two series differ in construction and volatility, both suggest that services inflation has stopped accelerating. This matters because services inflation has been the most persistent component of CPI. Even modest stabilization here reduces the risk that elevated inflation persists indefinitely, supporting a narrower range of long-run inflation outcomes.

Chart 7: Unit labor costs

Despite elevated wage growth, unit labor costs have moderated as productivity improves, lowering the probability that services inflation remains structurally high.

Chart 8: Labor productivity

Measured productivity growth narrows the range of plausible long-run inflation outcomes by lowering marginal costs and weakening wage–price persistence. One of the largest productivity shocks of the past decade coincided with the shift to remote work, a reminder that removing hours of low-value activity can matter more for output than most incremental technological upgrades. For long-dated assets, the level of inflation matters less than the stability of the cost structure that produces it, which is why productivity primarily acts through term-premium compression well before it appears in inflation data.

Why short-term inflation noise matters less

Near-term inflation is influenced by tariffs, supply chains, fiscal dynamics, and labor constraints, effects that are primarily reflected in short-dated rates. Long-duration assets are priced on longer horizons and respond to changes in confidence about long-run stability rather than transitory volatility. The thesis only fails if policy or structural forces permanently re-anchor long-run inflation expectations or sustain elevated term premium.

Chart 4: Treasury total return by maturity (Bloomberg US Treasury Total Return Indices)

Longer-duration Treasuries exhibit disproportionate total return sensitivity to changes in long-end pricing, making them the most direct way to express a view on term-premium compression.

Basket Expression

AI Productivity Duration Basket allocation:

• TLT — 62.50%

• EDV — 18.75%

• ZROZ — 18.75%

TLT provides liquid exposure to long-duration Treasuries. EDV and ZROZ increase effective duration and sensitivity to changes in term premium. This allocation is intended to be held as a structural position rather than traded tactically.

Optional convexity

In addition to the structural duration sleeve, an optional options overlay could be used to introduce convexity. This overlay is focused primarily on long-dated calls on TLT, with a smaller exposure to long-dated EDV calls, and ZROZ options included only where liquidity allows and sized so a full loss would be immaterial. The purpose is to benefit from a faster-than-expected repricing of long-run inflation or term premium, while keeping downside strictly limited to the premium paid.

Why this framework exists

Long-dated rates currently embed a high degree of caution about the future. That caution may ultimately prove justified. It may also prove excessive if productivity gains continue to show up in services pricing and cost structures.

The asymmetry is the point: a modest reduction in long-run inflation uncertainty can have a meaningful impact on duration pricing, while the downside of a limited allocation remains manageable.

When this would change

This sleeve would be reduced or removed if:

• long-run inflation expectations rise persistently,

• term premium increases structurally due to sustained fiscal pressure,

• productivity improvements fail to extend beyond narrow or isolated use cases.

Absent those developments, long duration remains a reasonable hedge against the possibility that long-run inflation risk is priced too aggressively.

Closing

This is a measured macro bet based on a simple observation: The future may be somewhat more stable, and less cost-intensive, than the market currently assumes. That alone can be enough for duration to work.

One final point, this framework is explicitly about long-run inflation uncertainty and term premium, not about near-term price levels or cyclical reflation. For that reason, several common “inflation hedges” are deliberately excluded.

• Inflation-linked bonds and commodities primarily benefit from rising realized inflation, which is not the view being expressed.

• Financials tend to benefit from elevated term premium and steeper curves, making them a poor fit for a compression thesis.

• Traditional value exposures are similarly sensitive to inflation persistence and higher discount rates.

• However, these could all still work for reasons not mentioned in the scope of this article.

Our framework is designed to benefit from uncertainty repricing rather than inflation acceleration.

Affiliation Disclosure:

The managing principal of Ridire Research is affiliated with a private investment fund that holds long positions in the securities discussed herein, which could influence the views expressed.

Disclaimer:

This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

• Executive Summary

• Company Overview

• The Setup

• Causal Mechanism

• Timeline

• Key Risks

• Conclusion

• Special Mentions (Not Covered In Detail)

Executive Summary

Anthropic’s recent move into legal workflow tooling is a useful stress test for every “AI will eat SaaS” thesis. The headline isn’t that Claude got better at answering questions rather it’s that Anthropic is publishing workflow plugins that can execute real clerical/legal ops tasks, document review, NDA triage, compliance tracking, pushing the company up-stack from model supplier toward workflow owner.

That matters because it clarifies the shape of disruption:

• The most vulnerable incumbents are those whose moats are primarily UI, seats, and workflow convenience, especially where outputs are “good enough,” iteration is cheap, and switching costs are low.

• The least vulnerable incumbents are those embedded in deterministic, audited, regulated, or counterparty-mandated workflows, where an error cascades into lawsuits, regulator attention, or operational shutdown, and where “system-of-record” status matters more than model quality.

This piece is a nuanced response to the deterministic vs probabilistic framing that is becoming popular. The framing is directionally right, but incomplete. The real dividing line is not “AI can’t do deterministic tasks.” It’s that deterministic workflows have external accountability (auditability, provenance, liability, certification, and integration into other systems). In those contexts, AI becomes a feature and an accelerant rather than a wholesale replacement.

That’s the lens behind Don’t Get Fired Software, illustrated through two case studies here and supplemented by broader, well-covered examples at the end.

• Trend Micro (4704 JP): a cybersecurity control plane. The moat is embedded deployment, operational integration, and career risk, the security stack you don’t casually rip out when the downside of a bad transition is a breach and a board interrogation.

• ReposiTrak (TRAK): a compliance/traceability system-of-record in food supply chains. The moat is not “better UX.” It is mandatory network effects, retailers and wholesalers can force suppliers to comply through a particular rail, making participation a cost of doing business.

The implication is practical: AI lowers the cost of building software and compresses margins in “probabilistic productivity” layers, but it does not eliminate the willingness to pay for guardrails where the alternative is reputational and legal catastrophe.

Company Overview

Trend Micro (4704 JP): Cybersecurity as a Control Plane (and a Career Hedge)

Trend Micro is an incumbent cybersecurity vendor that has deliberately repositioned from “point products” into a platform control plane, unifying detection/response, exposure management, and multi-surface security operations. Its investor communications explicitly emphasize platform consolidation and platform-led growth, in other words: selling an operating layer rather than a single tool.

Why it qualifies as “don’t get fired” software:

• Security is not a normal ROI purchase. It is risk transfer. The buyer (CISO/CIO) is protecting the firm and themselves from tail outcomes. That changes switching behavior: transitions are slow, painful, and politically loaded.

• “Deterministic vs probabilistic” is a false binary in security. Threat detection is probabilistic by nature, but the workflow is deterministic: response playbooks, auditability, control coverage, and incident reporting expectations. It’s the workflow that buyers cannot risk breaking.

• AI is additive, not substitutive. AI can reduce analyst load and speed triage, but it also expands the attack surface and increases governance expectations (the organization must prove it is managing cyber risk, not merely buying tools).

Trend Micro also benefits from being a credible consolidator in a market where CISOs increasingly want fewer vendors and a tighter operational loop (alerts → response → reporting). This is exactly the category of software that tends to survive pricing model shifts (seats → usage/outcomes): you can change the meter, but you can’t opt out of the underlying need.

ReposiTrak (TRAK): Compliance and Traceability as a System-of-Record Network

ReposiTrak sits in a different, and arguably cleaner, “don’t get fired” lane: regulated compliance + counterparty mandates in food supply chains. It operates a traceability/compliance platform that functions as a system of record across retailers, wholesalers, and suppliers, where documentation, data standards, and audit readiness are the product.

Why it qualifies as “don’t get fired” software:

• It is compliance infrastructure, rather than discretionary workflow. The job is to produce clean, auditable traceability and compliance data that stands up under scrutiny.

• The moat is forced participation + network effects. Retailers are increasingly requiring traceability across broader sets of foods and earlier timelines than the government baseline meaning suppliers adopt because access to distribution depends on it. ReposiTrak’s own SEC filing explicitly describes retailer requirements like expanded traceability scope (“all foods”), additional electronic data elements, and accelerated deadlines.

• AI assistants don’t replace the rail. An LLM can draft an email or summarize a document; it cannot by itself create a trusted, standardized interchange between thousands of counterparties with data validation, audit trails, and operational integration.

ReposiTrak also leans into a practical reality: suppliers need interoperability (EDI compatibility, standardized capture, and translation of formats across trading partners). That’s “plumbing” work and it tends to be sticky.

The Setup

1) Anthropic’s legal plugin

The market reaction to Anthropic’s legal plugin is less important than the product direction. Anthropic is shipping workflow plugins (including legal) that handle tasks like contract review and compliance tracking, tools that historically lived inside vertical software suites or services-heavy platforms.

This is the broader pattern: foundation model providers are moving from model access to workflow ownership, because workflow ownership captures:

• distribution (where the user lives),

• data context (what the model can pull from),

• and economic rent (the part of value that is billable).

The “legal tech fray” is simply where the first public repricing happened.

2) Deterministic vs probabilistic is useful, but the better lens is external accountability

The deterministic/probabilistic split is a good first approximation:

• Probabilistic outputs (marketing, sales enablement, support, content, planning) allow iteration and tolerate “good enough.”

• Deterministic workflows (payroll, clinical, grid, compliance) punish errors.

But the nuance is this: many workflows are hybrids, probabilistic assistance inside deterministic guardrails. Security is the obvious example. Legal is another: a model can triage NDAs, but the organization still needs provenance, audit trails, and defensibility.

So the practical question for disruption isn’t “Can AI do the task?” It’s:

• Can AI own the system-of-record?

• Can it meet auditability and provenance requirements?

• Can it offer reliability and liability posture that enterprises will accept?

• Can it integrate into existing counterparty and regulator expectations?

Where those answers are “no,” disruption becomes feature-ization instead.

3) Cost of building software is falling, but distribution and trust are not

AI-assisted coding and agentic workflows are compressing the marginal cost of producing software and automating white-collar tasks. That pressure naturally flows into pricing. Seat-based pricing is already under strain, and consumption/outcome-based models are spreading, especially where AI automates away “seats.”

However, “software gets cheaper” is not the same as “customers stop paying.” In risk control planes, what customers are paying for is assurance: uptime, auditability, incident defensibility, and reduced tail risk. Those are not free simply because code is cheaper.

4) The regulatory floor moved, but counterparty enforcement is the ceiling

This is the most underappreciated nuance in ReposiTrak’s story.

The FDA’s Food Traceability Rule originally pointed toward a nearer-term compliance date, but FDA has proposed a compliance-date extension and Congress directed FDA not to enforce the rule prior to the extended date.

If you stop there, you might conclude the urgency is gone. But suppliers don’t work for the FDA; they work for their customers. Retailers and wholesalers can, and do, set requirements beyond the FDA baseline, including broader coverage (“all foods”) and earlier timelines. This is why ReposiTrak is “don’t get fired” software: the enforcement mechanism is commercial, not merely governmental.

5) Cybersecurity governance is now formal disclosure territory

In cybersecurity, the regulatory vector is not “install product X.” It’s that public companies must treat cyber risk as governance risk. The SEC’s rules require disclosure of material cybersecurity incidents (Form 8‑K Item 1.05) and disclosure around cybersecurity risk management and oversight. That drives demand for platforms that make risk observable, operational, and reportable. Again, control planes.

Causal Mechanism

The mechanism is the same in both names, but it expresses differently.

Shared mechanism: “Don’t get fired” budgets behave differently

• In discretionary SaaS, the buyer asks: “What’s the ROI?”

• In risk control planes, the buyer asks: “What’s the downside if this fails, and who gets blamed?”

That second question creates three structural advantages:

1. Budget priority: security/compliance survives cuts because the downside is existential.

2. Vendor stickiness: switching introduces new risk; incumbents become the “safe choice.”

3. Outcome defensibility: even if pricing shifts away from seats, vendors that can demonstrate risk reduction or audit readiness retain pricing power.

Trend Micro mechanism: threat velocity + governance pressure → platform consolidation

• Threat velocity is rising, and the operational burden on security teams is not linear. That pushes enterprises toward platform consolidation, fewer vendors, integrated telemetry, consistent policy and response.

• Governance expectations are rising (disclosure, board oversight), so security becomes more than detection; it becomes auditability and reporting.

• The “AI era” increases rather than reduces the need for security control planes, because AI expands software footprints, data flows, and attack surfaces.

• In fact Claude Code has become a cyber weapon itself:

Anthropic-style disruption here is constrained: an LLM can assist triage, but it does not replace endpoint telemetry, cloud workload coverage, policy enforcement, or response integration. In practice, AI becomes an embedded layer inside the incumbent platform, meaning the vendor that already owns the control plane is well positioned to capture the uplift. Trend Micro’s communications emphasize platform momentum and integration (the direction that matters for survivability under AI pressure).

ReposiTrak mechanism: counterparty mandates + regulatory complexity → forced network adoption

• The traceability/compliance problem is structurally a network problem. It only works when multiple parties exchange standardized data.

• Retailers are increasingly setting requirements that exceed the regulatory baseline (broader scope and earlier deadlines), creating forced adoption dynamics.

• ReposiTrak positions itself as the interchange layer: capture once, normalize/validate, deliver in the format each trading partner requires, in other words the “plumbing” that makes compliance operational.

Anthropic-style disruption is limited here too: an LLM can help interpret requirements or draft responses, but it does not replace the compliance system-of-record, the audit trail, or the counterparty rail. If anything, AI may accelerate adoption by making onboarding and exception handling cheaper, again benefiting the incumbent rail.

Timeline

1) Near-term: AI moves up-stack, markets reprice “workflow incumbents”

Anthropic’s workflow push is already prompting investors to ask which SaaS categories are “workflow wrappers” versus “systems-of-record with teeth.”

This repricing is often indiscriminate at first. That creates opportunities: “don’t get fired” names can get hit with the same multiple compression narrative even when their fundamentals are structurally protected. The investor job is to separate workflow convenience from operational control planes.

2) Cybersecurity: disclosure regime + consolidation cycle

SEC cyber disclosure and governance requirements increase the premium on platforms that support incident readiness, documentation, and defensibility. Concurrently, the market continues to push security consolidation. This favors vendors that can credibly offer a unified control plane and integrate AI capabilities without breaking reliability expectations.

3) Food traceability: the enforcement date may shift, but retailer enforcement persists

The FDA timeline has shifted in terms of enforcement posture. But retailer and wholesaler requirements are already pushing suppliers to operationalize traceability sooner, and often across broader product sets than the FDA’s list.

This creates a two-speed timeline:

• Commercial enforcement (retailers/wholesalers) drives adoption now.

• Regulatory enforcement extends the long tail and increases the “late majority” cohort later.

4) Pricing model shifts: seats → consumption/outcomes, and who wins

As AI automates labor, seats become a weaker billing unit. Consumption-based and outcome-based pricing becomes more common, and vendors must show value in concrete terms.

“Don’t get fired” vendors can adapt because they sell outcomes customers already care about:

• risk reduction,

• compliance readiness,

• auditability,

• faster incident containment,

• fewer recalls/exceptions.

Key Risks

Theme-level risks

1. Pricing pressure is real even for mission-critical software.
   AI lowers switching costs at the margin by making integration, customization, and migration easier. It also shifts procurement toward outcomes. “Never cut” doesn’t mean “never negotiate.”

2. AI-native workflows can commoditize layers around the system-of-record.
   Even if the core rail is sticky, some value can migrate to orchestration layers. The winner is often whoever controls distribution and the interface where decisions are made.

Trend Micro risks

• Platform competition intensifies. Large vendors can bundle security into broader suites and compress stand-alone pricing. Even with strong product positioning, the market can treat incumbents as “legacy” if they fail to anchor the platform narrative.

• AI changes the threat environment faster than vendor roadmaps. If attackers outpace defenses materially, buyer behavior can shift and procurement can rotate to perceived leaders.

• Perception risk. Cyber is a narrative sector. The market can punish “steady” names and reward “story” names even when fundamentals diverge.

(Note these are not unique to Trend Micro; they are endemic to cybersecurity as a market structure.)

ReposiTrak risks

• Regulatory slack can slow urgency at the margins. Even if retailer mandates drive adoption, some suppliers will procrastinate if enforcement is perceived as distant.

• Standards fragmentation. If the industry fractures into multiple incompatible traceability rails, network effects weaken. The counterpoint is that fragmentation is operationally painful, which is why consolidation is the natural endpoint.

• Execution and service risk. The “system-of-record” claim carries expectations: uptime, data integrity, onboarding throughput, and support quality. If the network is perceived as unreliable, it invites alternatives.

Conclusion

Anthropic’s legal plugin is a reminder that the application layer is no longer sacred. When a model provider can ship a workflow that does meaningful work review, triage, track compliance, incumbents built on interface convenience and seat economics should expect pressure.

But the right conclusion is not “AI eats all SaaS.” The right conclusion is:

• Probabilistic productivity layers (where iteration is cheap, switching is easy, and “good enough” works) will be commoditized fastest.

• Deterministic, externally accountable control planes (where auditability, provenance, integration, and liability dominate) will be disrupted differently: AI will be absorbed as a feature, and the system-of-record will remain sticky.

Trend Micro and ReposiTrak sit in that second category. They are don’t get fired software: the security stack you don’t casually replace, and the compliance rail you can’t opt out of when counterparties set the rules. Their moats are not code. Their moats are embeddedness, accountability, and the cost of being wrong. That’s the nuance response to the Anthropic moment: AI is absolutely compressing the marginal cost of software and reshaping pricing models, but it is not erasing the premium enterprises pay for guardrails when the downside is existential.

Special Mentions (Not Covered in Detail)

A few other companies clearly fit the “Don’t Get Fired Software” framework and reinforce the core nuance of this theme. We’re not covering them in depth here for a simple reason: they are already widely covered by the market or have been covered previously by us.

• Microsoft (MSFT)
  Microsoft’s security stack (Defender, Entra, Purview) has effectively become a default enterprise security control plane through bundling, distribution, and governance integration. This is well understood and heavily trafficked territory. AI enhances Microsoft’s position, but does not change the underlying control-plane logic.

• Palo Alto Networks (PANW)
  Palo Alto is the most obvious public-market example of the thesis: premium pricing justified by auditability, platform consolidation, and board-level defensibility. The market already treats PANW as a “safe” security incumbent.

• Digital Arts (2326 JP)
  Digital Arts represents the same dynamic in the Japanese market: deterministic security infrastructure embedded in regulated enterprises, education systems, and public institutions. Its durability stems from trust, certification, and institutional deployment depth. This name has been discussed previously here

Why they’re excluded here:
This piece is not meant to catalogue every beneficiary of the theme. The purpose is to clarify the mechanism:

AI disrupts probabilistic workflows first. It struggles to displace control planes where errors create legal, regulatory, or career-ending consequences.

MSFT, PANW, and Digital Arts already sit squarely in that category, and are priced, discussed, and understood as such. The incremental insight lies in identifying less obvious expressions of the same structural truth (TRAK, TrendMicro), not re-litigating the best-known examples.

Disclaimer:

This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

• Executive Summary

• Company Overview

• The Setup

• Casual Mechanism

• Timeline

• Key Risks

• Conclusion

Executive Summary

LGI Homes (LGIH) is a business driven more by monthly affordability than headline pricing.

As financing conditions ease, entry-level buyers can return quickly, homes sell faster, and existing inventory begins to move which is enough to improve earnings even before profits per home recover. What it does not do immediately is lift margins. Keeping payments affordable often requires incentives and working through higher-cost inventory, which can hold margins down in the short term.

The opportunity is recognizing that the early phase of improvement shows up first in volume and operating leverage, with margin recovery coming later.

Company Overview

LGI Homes is a U.S. homebuilder focused on attainable, entry-level housing. The business is built around a standardized product, fast inventory turnover, and a sales process designed to convert renters into first-time buyers. What matters for this thesis comes down to how the company creates demand and how it manages risk when demand softens. Two structural features matter for this thesis:

• Affordability is the product: LGI does not compete on design or customization, it competes on the monthly payment. Management explicitly frames the sales process around getting buyers to a payment they can qualify for, which makes demand highly sensitive to financing conditions and borrowing costs.

• Wholesale acts as a pressure valve: Alongside retail sales, LGI sells a portion of homes in bulk to institutional buyers, primarily single-family rental operators. This channel helps keep construction moving and inventory from backing up when retail demand slows, but it typically comes with lower margins than traditional home sales.

It is worth noting that LGI also operates mortgage and insurance businesses through joint ventures. These help improve buyer conversion and enable financing incentives.

The Setup: Demand Improves First, Margins Lag

Easier financing tends to bring buyers back quickly, especially at the entry level. As borrowing conditions improve, more households qualify, traffic picks up, and homes begin to sell at a faster pace. That early demand response is real and usually shows up before anything else.

Where things get less intuitive is on profitability. For LGI Homes, supporting affordability has historically meant relying on incentives, adjusting sales mix, and working through higher-cost inventory accumulated during tougher periods. Those actions help move homes and stabilize the business, but they can keep margins under pressure even as demand improves.

That tension is the core of the setup: falling rates can clearly help volumes and sales momentum, while profitability per home takes longer to recover. The early benefit shows up in activity and cost absorption, not immediately in cleaner margins.

Causal Mechanism

• Easier payments bring buyers back first: LGI Homes sells to buyers who sit right on the edge of affordability. For this group, a small improvement in financing can be the difference between qualifying for a mortgage or remaining renters. When that line is crossed, demand responds quickly. This shows up first in sales activity, more orders, a growing backlog, and faster inventory turnover, well before it shows up in profitability.

• Why margins don’t snap back right away: Even as demand improves, profits per home can lag. Keeping homes affordable often requires continued incentives, preferred financing programs, and selective discounting. At the same time, the company may lean more on wholesale sales to keep construction moving when retail demand is still rebuilding. Finally, there is a timing issue: homes being sold today were often built when land, labor, or financing costs were higher. All of this can keep margins under pressure even as the business itself is clearly improving.

• Where earnings actually improve first: The early earnings lift comes from operating leverage, not pricing. As more homes are sold, fixed costs are spread across a larger base, improving overall profitability even if margins per home remain constrained. This cost absorption is typically the first clear financial signal that conditions are turning.

In the early stages of easier financing, the stock can work because volumes recover and costs are better absorbed, even while margins remain capped, earnings can improve simply by selling more homes through the same cost structure. In simplest terms margins are about making more per house; earnings are about selling more houses and right now the machine just needs to run faster, not get better.

Timeline

• Near term: The next earnings update is the first real checkpoint. What matters is not precision guidance, but tone and direction, whether management signals improving demand without leaning harder on incentives, and whether wholesale remains a tactical tool rather than a growing crutch.

• Over the next few quarters: The story becomes clearer as operating data accumulates. A sustained pickup in backlog and sales pace without an obvious step-up in incentives would confirm that demand is returning organically. At the same time, a gradual reduction in wholesale activity would signal that retail demand is strong enough to stand on its own. Improving cost absorption as volumes rise would be the clearest evidence that earnings power is rebuilding, even if margins per home remain constrained.

• On the macro side: This setup works best if financing conditions continue to ease and stay supportive. Stable or improving affordability, paired with steady buyer activity, would reinforce the idea that volume recovery can carry results forward before margins fully normalize.

Key Risks

• Rates don’t fall (or re-accelerate): The buyer cohort is payment-constrained; higher-for-longer would extend the incentive war.

• Incentive arms race intensifies: LGIH ran marketing-heavy promotions (a year-end savings event with price reductions and preferred-lender rate options), which can be necessary to drive pace but pressures margin.

• Wholesale becomes structural: If wholesale remains mid-teens as a % of closings, it may signal retail fragility and keep margins capped.

Conclusion

This is a case where improvement shows up before it looks pretty. LGI Homes does not need higher margins to start working as a stock it just needs buyers to come back. As financing eases, demand can recover quickly at the entry level, inventories can move, and earnings can improve through volume and cost absorption alone.

Margin recovery matters, but it is not the first signal, and waiting for it risks missing the earlier inflection. The stock works if affordability improves enough to sustain pace without escalating incentives, allowing operating leverage to do the heavy lifting. In that sense, this is less a bet on pricing power and more a bet on timing, recognizing that the business stabilizes and earnings recover before margins fully normalize.

Disclaimer:

This publication is for educational and informational purposes only. Any performance referenced is illustrative and tracked on a per-article basis, not as part of a model portfolio or investment program. Nothing herein constitutes investment advice, a recommendation, or a solicitation. → Ridire Research Substack Disclaimer

Table of Contents

• Executive Summary

  • Key Thesis Points

  • Theme Basket Overview & Expression

• Nuclear Medicine Overview

  • Casual Mechanism & Tipping Point

• Competitive Advantages & Structural Edges

• Ecosystem Landscape

• Valuation & Implied Expectations

• Capital Cycle, Policy & Capital Allocation

• Key Risks

Executive Summary

Nuclear medicine, the use of radioactive isotopes for imaging and treating disease, is emerging from niche status to become a frontline modality in cancer care. Our thesis is that recent breakthroughs in radiopharmaceutical therapies and supply-chain infrastructure have brought this field to a tipping point. We identify a long basket of select radiopharmaceutical developers and “picks-and-shovels” suppliers to capture this theme’s multi-faceted upside while mitigating single-name risk.

Figure 1: The global nuclear medicine market is projected to grow significantly, supporting the investment thesis that the field is at a tipping point. This growth is driven by the introduction of new therapies and expanding applications.

Key Thesis Points:

• Demonstrated Efficacy: Targeted radiotherapies have shown compelling clinical results in hard-to-treat cancers (Novartis’s radioligand drugs prolong survival in neuroendocrine tumors and prostate cancer). This validates nuclear medicine as a powerful therapeutic approach.

• Technology & Infrastructure Advances: Improvements in isotope production, targeting molecules, and safety protocols have lowered historical barriers. Modern “theranostic” platforms pair imaging and therapy, ensuring patients get the right isotopic treatment at the right time. Meanwhile, specialized suppliers are scaling up isotope manufacturing and distribution networks.

• Big Pharma Endorsement: Large pharmaceutical companies are making major investments in radiopharma. Recent acquisitions (Bristol Myers buying RayzeBio, AstraZeneca acquiring Fusion Pharma, Lilly acquiring POINT Biopharma) signal a long-term commitment to build radiotherapeutics franchises, bringing deep capital and expertise into the field.

  

  Figure 2: A wave of multi-billion dollar acquisitions by major pharmaceutical companies helps validate the long-term potential of the radiopharmaceutical sector.

• Broad Optionality: Radiopharmaceuticals represent a platform, not a single drug. The same core technologies (radioisotopes + targeting ligands) can be applied across multiple cancer types and even other diseases. This gives the sector numerous “shots on goal”, success in one indication can open the door to many others, delivering outsized growth for the leaders.

• Structural Tailwinds: We see a favorable capital cycle: years of underinvestment in nuclear med are reversing as demand rises. Regulatory agencies and governments are supporting domestic isotope supply and fast-tracking novel radiopharmas. This backdrop creates a supportive environment for the theme to play out over the next 3–5+ years.

Theme Basket Overview:

• We express this theme with a basket of equities spanning the value chain – from drug development to isotope supply. This includes established leaders with commercial products (to provide stability and cash flow), late-stage innovators nearing the market, emerging platform players (for high-upside optionality), and infrastructure providers that profit from broader industry growth.

• The basket balances current revenue generators (Novartis, Telix) with pipeline-dependent biotechs (Actinium, Clarity, Alpha Tau). It also adds lower-volatility “picks and shovels” names (BWX Technologies, Eckert & Ziegler) that benefit from rising isotope demand regardless of which drug wins. This mix is designed to capture the theme’s upside while reducing single-name binary risk.

• By including both growth-oriented biotech and cash-flowing industrials, the basket is positioned to weather different market conditions. In a bull case where multiple radiotherapies succeed, the pure-play biotechs could soar. In a scenario where progress is slower, the infrastructure names and diversified pharma exposure help preserve value.

Thematic Basket Expression

We express the nuclear medicine theme through a diversified long-only basket spanning Revenue Anchors, Infrastructure (Picks & Shovels), and Pipeline Optionality. The construction is designed to capture upside across the full value chain while reducing single-asset binary risk.

Revenue Anchors — 40%

Novartis (NVS) and Telix (TLX AU / TLPPF) anchor the basket with commercial scale, proven execution, and near-term revenue visibility. These positions provide durability and ensure exposure to radiopharma adoption even if pipeline outcomes vary.

• NVS — 20%

• TLX AU / TLPPF — 20%

Infrastructure (Picks & Shovels) — 30%

BWX Technologies (BWXT) and Eckert & Ziegler (EUZ GY) represent the isotope supply and manufacturing backbone of nuclear medicine. These positions benefit from rising sector activity regardless of which specific therapies ultimately dominate.

• BWXT — 15%

• EUZ GY — 15%

Pipeline Optionality — 30%

Actinium (ATNM), Clarity (CLRPF), Perspective (CATX), and Alpha Tau (DRTS) provide asymmetric upside through differentiated radiotherapy approaches. These are intentionally sized smaller to reflect clinical and regulatory risk, with the expectation that only one or two successes are required to materially lift basket returns.

• ATNM — 10%

• CLRPF — 8%

• CATX — 6%

• DRTS — 6%

Nuclear Medicine Overview

Nuclear Medicine is an interdisciplinary field at the intersection of nuclear physics and medicine, encompassing the use of radioactive isotopes for diagnosis and therapy. In practice, this ranges from imaging procedures, for example, PET and SPECT scans where radioactive tracers illuminate biological processes, to radiopharmaceutical therapies that deliver lethal radiation doses directly to disease sites. For decades, nuclear medicine was largely confined to diagnostics (think of cardiac stress tests or FDG PET scans in oncology) and a few niche treatments (like radioactive iodine for thyroid disease). Today, however, a new generation of radiopharmaceutical “theranostics”, pairings of diagnostic scans and targeted radioisotope treatments, is poised to move the field from the fringes of oncology into the mainstream.

Nuclear medicine occupies a unique niche in healthcare, sitting between traditional pharmaceuticals and the world of nuclear technology. It requires a tight integration of science and infrastructure: advanced chemistry to link radioactive atoms with targeting molecules (like antibodies or peptides), as well as specialized facilities to produce, transport, and safely administer these compounds. Historically, this complexity kept radiopharmaceuticals as a backwater in pharma – a bit too “nuclear” for biotech investors, and a bit too biotech for the nuclear industry. That is now changing. The convergence of better molecular targeting (from biotech) and improved isotope production techniques (from nuclear engineering) has created the conditions for a renaissance in nuclear medicine.

Figure 4: Theranostics allows clinicians to "see what they treat" by pairing a diagnostic agent with a therapeutic agent.

Two landmark therapies illustrate this turning point:

• Lutathera (Lutetium-177 dotatate) was approved in 2018 for certain neuroendocrine tumors, becoming one of the first modern radioligand therapies to show a survival benefit in solid tumors.

Figure 5: The mechanism of PRRT, the technology behind Lutathera, which targets neuroendocrine tumors.

• Then in 2022, Pluvicto (Lutetium-177 PSMA-617) was approved for advanced prostate cancer, after dramatically reducing tumor progression in trials.

Figure 6: The mechanism of action for PSMA-targeted radioligand therapy, where the drug binds to and is internalized by prostate cancer cells, delivering a lethal dose of radiation.

• These drugs proved that by combining a radioactive payload with a targeting vehicle (such as a peptide that seeks tumor cells), one can attack cancers that are otherwise incurable, with manageable safety profiles. The success of Lutathera and Pluvicto provided a clinic-ready template for others to follow, sparking intense R&D activity and investment across the sector.

Causal Mechanism & Tipping Point

Why Now? The core mechanism driving this theme is straightforward: targeted radiotherapy works. By delivering ionizing radiation from within the body, right at the cancer cell, radiopharmaceuticals can destroy tumors that evade surgery, external radiation, or conventional drugs. The concept isn’t new, earlier attempts date back to radio-immunotherapy drugs in the early 2000s, but the tipping point came from improvements in precision and proof. Unlike older agents, today’s radiopharmas use more selective targeting ligands (for example, PSMA in prostate cancer is a highly specific marker) and often pair therapy with diagnostic scans to confirm that the drug will hone in on a patient’s tumors. This greatly increases the chances of success and builds physician confidence.

Figure 7: The radiopharmaceuticals market is forecasted to experience steady growth, driven by the clinical success of new therapies.

Several converging factors created the tipping point we’re now witnessing:

• Clinical validation: The strong trial results and real-world outcomes from Lutathera and Pluvicto in the late 2010s and early 2020s gave physicians and investors tangible proof that radiopharmaceuticals can extend and improve lives. After decades of being viewed as experimental, nuclear medicine suddenly had credible champions and published data in top journals showing meaningful survival benefits. This changed the narrative from skepticism (“too toxic, too complicated”) to optimism (“a new pillar of cancer therapy”).

• Supply chain maturation: A therapy is only as good as its ability to reach patients. Historically, one of nuclear medicine’s Achilles’ heels was the fragile supply chain for isotopes. Many medical isotopes are produced in only a handful of aging nuclear reactors worldwide, leading to chronic shortages (for instance, the well-known molybdenum-99 supply crises affecting diagnostic scans). However, the past few years have seen major initiatives to modernize and expand isotope production: new reactor projects and particle accelerators dedicated to medical isotopes, government funding (in the US, EU, and others) to localize supply, and companies like BWXT and Eckert & Ziegler ramping up commercial production lines. These efforts mean the infrastructure is finally catching up to the science, reducing the risk that a breakthrough drug languishes due to lack of material.

• Better targeting & safety: Early radio-drugs sometimes faltered because they either hit unwanted tissues or caused collateral damage. Advances in biochemistry (e.g. more stable chelators that securely hold radioactive metals, and more specific targeting agents) have improved the therapeutic index of radiopharmas. For example, modern radioligands can target antigens almost exclusive to tumors (like PSMA or SSTR2) and clear quickly from the bloodstream, limiting radiation to just the cancer. Moreover, supportive care protocols (such as amino acid infusions to protect kidneys during Lutetium treatments) have evolved to mitigate side effects. The result is a tipping point where the benefits now clearly outweigh the risks in many indications, which was not the case 20 years ago.

Figure 10: The development of different types of radioisotopes, such as alpha and beta emitters, provides a range of therapeutic options with distinct properties for targeting various cancers.

• External pressures in oncology: It’s worth noting the broader context that oncology is hungry for new modalities. Many cancers (prostate, lung, brain metastases, etc.) still have poor outcomes in advanced stages. As immunotherapies and targeted pills plateau in some settings, doctors are looking for the next leap. Radiopharmaceuticals offer a fundamentally different mode of attack, physical cell destruction, that can complement or overcome resistance to other treatments. This has created an openness among oncologists and regulators to try novel approaches, paving the way for radiotherapeutics to get a fair shot where previously they might have been dismissed as too unorthodox.

• Chain Reaction of Adoption: Success breeds success. As initial patients responded well and word spread, more clinical trials launched, drawing in more investment. We’re now seeing a classic feedback loop: positive data → more funding and partnerships → more trials → broader proof-of-concept → increased big pharma interest → even more funding.

Competitive Advantages / Structural Edges

Nuclear medicine exhibits unusually strong structural moats driven by technical complexity, regulatory friction, and supply-chain constraints. The same factors that historically limited adoption now favor incumbents that have already built the required capabilities.

• High barriers to entry: Radiopharma is capital-intensive and operationally complex. Production requires radiation-safe facilities, licensed handling of isotopes, and regulatory approvals that are slow to replicate. On the supply side, companies like BWXT and Eckert & Ziegler have spent decades building nuclear and isotope expertise that newcomers cannot quickly duplicate. On the therapy side, platforms such as Telix and Actinium have accumulated proprietary know-how, chelators, targeting ligands, and manufacturing processes, that compound over time.

A hot cell facility at the Saskatchewan Centre for Cyclotron Sciences, showing the specialized radiation-safe infrastructure required for radiopharmaceutical production.

• Integrated infrastructure and first-mover advantage: Early leaders were forced to integrate vertically to succeed, and that integration is now a competitive edge. Scaled manufacturing, logistics, and hospital relationships allow faster and more reliable delivery than new entrants starting from zero. First-mover adoption also creates operational stickiness: once hospitals integrate a radiopharma workflow, they are more likely to expand usage with the same suppliers.

• Theranostic Model – Patient Selection and Loyalty: Radiopharma companies benefit from a theranostic approach: they often develop a diagnostic scan alongside the therapeutic. This is a competitive advantage because it creates a built-in funnel for their therapy. For example, if a Telix or Clarity Pharmaceuticals provides an imaging agent that identifies patients with a certain tumor target, those patients (and their doctors) are naturally led to consider the corresponding therapy from the same company. It’s a seamless pipeline from diagnosis to treatment that competitors without a paired diagnostic can’t easily replicate. This model also potentially speeds adoption – by demonstrating with a scan that a patient’s cancer is lighting up (i.e. has the target), you give physicians confidence that treating with the radiopharmaceutical will be worthwhile. As a result, companies following this theranostic playbook can gain physician mindshare and patient loyalty that others might struggle to chip away.

A detailed workflow diagram showing the theranostic approach from pre-therapy imaging through post-therapy imaging and treatment.

• Concentrated supply and sticky relationships: Key isotopes are produced by a small number of qualified facilities, creating de-facto oligopolies. Switching suppliers is non-trivial due to quality control, regulatory re-validation, and logistics. Once long-term supply contracts are in place, they tend to be durable, supporting pricing power and stable demand for infrastructure providers.

• Convexity through M&A interest: One could argue another “advantage” for investors in this space is the heightened probability of takeover at premium valuations. Big Pharma’s growing interest means each successful smaller player has multiple suitors. This isn’t a traditional competitive advantage in product terms, but it is a structural feature of the industry now: if a small cap radiopharma shows positive phase 2 data, the odds are higher than in many biotech sub-fields that it gets snapped up (given the current arms race for radiotherapy pipelines). For our theme, this dynamic provides upside optionality, even if a company might not succeed as a standalone long-term, investors could benefit from a buyout at a rewarding price.

In summary, nuclear medicine rewards companies that solve the hard problems: isotope access, manufacturing, logistics, and workflow integration. These structural edges favor entrenched platforms and infrastructure providers, aligning with a basket approach that owns both the therapies and the system that makes them scalable.

Ecosystem Landscape

The nuclear medicine ecosystem has matured into a vertically interdependent value chain, spanning large pharmaceutical platforms, specialist radiopharma developers, and a constrained isotope supply base. Our basket is constructed to capture value at each layer.

Large Pharma Anchor

• Novartis (NVS). Novartis remains the category leader and reference point for radioligand therapy. It was the first large pharmaceutical company to build an integrated radiopharma franchise, combining drug assets with global manufacturing and distribution. Its continued investment and commercialization establish radiopharma as a durable therapeutic category rather than an experimental niche. Novartis serves as both a stabilizing anchor and a benchmark for what scaled success looks like in the sector.

Specialist Radiopharma Biotechs:

• Telix Pharmaceuticals (TLX.AX) – A global leader among pure-plays, Telix already commercialized an imaging agent (Illuccix for prostate cancer PET scans) and is advancing a broad pipeline of therapeutic candidates (targeting renal cancer, brain cancer, etc.). Telix’s success with Illuccix (achieved significant sales within its first year of launch) underscores its strong execution ability. It’s carving out a role as a full-spectrum radiopharma company (diagnostics + therapeutics), somewhat akin to a mini-AAA/Novartis but independent. We consider Telix a bellwether for investor sentiment in this sector – and a potential mid-term acquisition target for a larger pharma given its head start.

• Actinium Pharmaceuticals (ATNM) – A US-based biotech focusing on alpha and beta emitting therapies for hematologic cancers. Its lead program, Iomab-B (using radioiodine to condition bone marrow prior to transplant in leukemia patients), has completed Phase 3 with positive results. Actinium also has collaborations to use Actinium-225 labeled antibodies in combination with other drugs for leukemia. If approved, Iomab-B would open a new niche (radiation conditioning for bone marrow transplant) with little competition, effectively creating a new market category. Actinium’s expertise in antibody radioconjugates gives it a platform it can potentially extend to other blood cancers or even solid tumors. We include it for its late-stage status and unique focus – though we acknowledge regulatory uncertainty around the approval path is a current overhang.

• Clarity Pharmaceuticals (CLRPF) – An emerging platform out of Australia, Clarity is championing copper-based radiopharmaceuticals (Copper-64 for imaging, Copper-67 for therapy). The idea is that copper isotopes have ideal properties for theranostics (Cu-64’s half-life allows centralized manufacture and distribution to imaging sites, and Cu-67 can be produced without a reactor, theoretically easing supply issues). Clarity has multiple drugs in early clinical development, including targets in prostate cancer and neuroblastoma. We include Clarity as a representative of the “next wave” of platform biotechs, it offers exposure to potentially more scalable technology (if copper can be produced via cyclotrons, that’s a game-changer for supply) and has attracted partnerships and government support. It’s still in trials, so it’s a higher-risk, higher-reward piece of the basket.

• Perspective Therapeutics (CATX) – Formed via the merger of Isoray and Viewpoint Molecular Targeting, Perspective combines a legacy brachytherapy business (Cesium-131 implants for treating prostate cancer, which Isoray had sold for years) with new targeted alpha therapies from Viewpoint. Their pipeline includes lead-212 conjugated molecules (Lead-212 is an alpha emitter) for cancers such as melanoma and neuroendocrine tumors. Perspective brings an interesting angle: they have experience in localized radiation (brachytherapy) and are now translating that into systemic targeted therapy. With isotopes like lead-212, they can create tumor-seeking radiopharmaceuticals. This company is early-stage, but it’s a play on alpha-particle therapy, an area that could leapfrog beta emitters in potency if challenges are solved. We include CATX for diversification into alpha therapeutics and to capture upside from any progress in their novel pipeline.

• Alpha Tau Medical (DRTS) – A bit different from others, Alpha Tau’s approach (Alpha DaRT) straddles the line between device and drug. They insert tiny seeds imbued with Radium-224 directly into tumors, which then emit alpha particles that kill the tumor from within over a few weeks. It’s a localized therapy (unlike the systemic injections of other radiopharmas) and is being tested in indications like skin cancers, breast tumors, and glioblastoma. Alpha Tau’s technology showed promising early results in a pilot skin cancer study. If it proves out in larger trials, it could become a new modality for solid tumors that can be accessed by a needle or minor surgery. We added Alpha Tau to the basket to reflect the diversity of approaches in nuclear medicine – it’s a more targeted “procedure-based” therapy that could complement systemic radioligands. It also exemplifies how the nuclear medicine toolbox is expanding (not just injections, but implants too). The company has risk (still in mid-stage trials, and adoption would require training surgeons/radiation oncologists on a new technique), but its upside is high as a potential disrupter of certain surgical or external radiation procedures.

There are other specialist radiopharma companies outside our basket for various reasons, for instance, Lantheus (LNTH) which is a commercial leader in diagnostics (they market Pylarify, an F-18 PET agent for prostate cancer). We chose not to include Lantheus mainly because its radiotherapeutic exposure is indirect (though it has licensed rights to some of POINT’s therapeutic candidates, Lantheus is still primarily an imaging and contrast agent company, and its valuation already reflects the huge success of Pylarify). Similarly, smaller biotechs exist but are either not easily investable for generalists or too early-stage to fit this memo. Our basket focuses on a representative set that covers both the current leaders and the promising newcomers in a balanced way.

Infrastructure & Isotope Supply

• BWX Technologies (BWXT) – A U.S. company with a long history in nuclear reactors (for submarines, power, etc.), BWXT has a growing medical segment. They are working on producing Molybdenum-99 (the source of Technetium-99m, used in millions of scans annually) using innovative reactor technology that doesn’t rely on weapons-grade uranium, a significant advancement in securing domestic supply. They also have initiatives for manufacturing Actinium-225 at scale in partnership with national labs, and are involved in producing other isotopes like Yttrium-90 and Strontium-89. BWXT’s advantage is it has the engineering muscle and capital to build complex facilities, something few pure biotechs can do. It also generates steady revenue from its core defense contracts, giving it financial stability. In our theme, BWXT represents the “macro” infrastructure bet, if nuclear medicine demand skyrockets, BWXT will likely see large, long-term contracts and government support to produce isotopes, boosting its medical segment earnings.

BWXT Medical's Tc-99m generator facility.

• Eckert & Ziegler (EZJ.DE) – A German company that has quietly become a global leader in radioisotope products and services. E&Z provides everything from radioactive source materials (Germanium-68 generators for Gallium-68 production, Cobalt sources for industrial and medical use, etc.) to contract development and manufacturing (CDMO) services specifically for radiopharmaceuticals. For example, if a biotech needs to build a production line for a Lu-177 drug, they might partner with E&Z to design and supply the hot cells and handling systems, or even have E&Z produce it for them. E&Z also produces small quantities of high-purity isotopes for trials. Essentially, it’s the one-stop shop for any company entering nuclear medicine that doesn’t have in-house manufacturing – a picks-and-shovels pure-play. We include E&Z in the basket because its profits rise with the entire sector: as more radiopharma trials commence and products commercialize, E&Z sees increased orders, service contracts, and sales of generators and equipment. It’s more Europe-focused (which is where a lot of isotope R&D happens), nicely complementing BWXT’s North American emphasis.

Eckert & Ziegler's radionuclide production equipment.

Healthcare Providers and Distributors: Another part of the ecosystem (not in our basket, but worth noting) are the nuclear pharmacies and hospital networks. In the US, companies like Cardinal Health and GE HealthCare’s pharmacy network are crucial for last-mile delivery – they take isotopes from producers and compound patient-ready doses daily. While we haven’t included these as positions (Cardinal is a broad healthcare company, GEHC is large imaging modality player), their role indicates that distribution channels are in place. It means scaling a new drug might be easier now because these networks can be tapped rather than built from scratch. Additionally, large hospital systems are investing in theranostic centers (for instance, some academic centers have opened dedicated radioligand therapy clinics). This increasing provider-side investment is a positive sign for the ecosystem’s maturity.

Competition and Landscape Summary:

• Novartis competes mainly with itself (it set the bar).

• Telix competes with Lantheus in imaging and will compete with big pharma radioligand programs in therapy (Novartis, Lilly’s POINT programs) but has a broad pipeline to differentiate.

• Actinium is fairly unique in bone marrow conditioning; its other programs (like CD33 targeting in leukemia) compete conceptually with CAR-T or antibody therapies for leukemia, not with another radiopharma per se.

• Clarity will inevitably compete with whoever is in the same indication (if their copper prostate therapy advances, it would compete vs Novartis’s Pluvicto and Lilly/POINT’s candidate). But Clarity’s bet is that a different isotope and improved targeting can carve a niche or even surpass incumbents.

• Perspective and Alpha Tau are in early stages where the “competition” is more status quo treatments (standard brachytherapy or surgery) rather than another similar company.

• BWXT’s competition is limited (mostly government-run reactors or a couple of smaller firms like Nordion in Canada or NorthStar in the US).

• Eckert & Ziegler similarly is one of few in its space; other players include ITM (a private German company focusing on Lutetium supply and therapies) and maybe Cardinal’s radiopharmacy division in the US. But E&Z has international reach and deep expertise that give it a solid position.

One can see that collaboration is as prevalent as competition, many of these companies will likely partner (a biotech partnering with BWXT for supply, or a big pharma licensing a smaller company’s compound). Our basket essentially picks those we think have strong individual prospects and also strategic value to others in the ecosystem.

Valuation & Implied Expectations

Traditional valuation frameworks are poorly suited to nuclear medicine. Instead of focusing on multiples, the relevant question is whether market expectations for the modality and its infrastructure are too low relative to the likely scale of adoption.

• Large Pharma (NVS): Radioligand therapy is still treated as a non-core option within a diversified pharma model. Expectations appear conservative, shaped more by manufacturing constraints than demand. Upside comes from indication expansion and scale, which would reframe radiopharma as a durable growth engine rather than a niche line item.

• Pure-Play Leaders (TLX): The market prices in at least one successful therapeutic transition beyond diagnostics, but does not fully price the platform nature of theranostics. Success in a single additional program could materially re-rate expectations; success across multiple programs would reframe the company entirely.

• Late-Stage / Platform Biotech (ATNM): Market expectations are low, reflecting regulatory uncertainty and narrow initial indications. This creates asymmetry: approval or partnership would force a rapid re-underwriting, while failure is partially mitigated by diversified basket exposure.

• Early Platforms (CATX, DRTS): Valuations imply a low probability of success. The market assumes most early programs fail. The basket thesis does not require broad success, one meaningful clinical win is sufficient to justify inclusion.

• Infrastructure (BWXT, EUZ GY): These names are valued as steady industrial or nuclear businesses, with medical isotopes treated as incremental. If radiopharma activity accelerates, medical exposure could drive earnings surprises and multiple expansion not currently reflected.

Capital Cycle, Policy & Capital Allocation

• Early speculative funding enabled proof-of-concept and late-stage trials. The current phase is dominated by strategic capital (large pharma M&A, partnerships, and long-term manufacturing investment), which is stickier and less cyclical.

• Capital is being allocated toward manufacturing capacity, isotope access, and platform build-out, not just clinical experimentation. This signals confidence in durability rather than short-term option value.

• Governments view medical isotopes as strategic infrastructure. Funding, regulatory alignment, and reimbursement frameworks have reduced non-scientific friction to adoption. This lowers the risk that effective therapies fail due to logistics or payment issues.

• Overfunding and thematic crowding remain possible, but leaders with scale, differentiated technology, or infrastructure positioning are more likely to consolidate rather than be displaced.

Key Risks

• Isotope Supply Crunch: If the ramp-up in isotope production does not keep pace with demand, the growth of radiotherapies will hit a hard ceiling. Many radiopharmaceuticals use exotic isotopes that are currently produced in gram or millicurie quantities worldwide. For example, Actinium-225 (an alpha emitter powering several next-gen therapies) is in extremely short supply, global production can meet only a fraction of potential patient demand. Efforts are underway to scale production (new reactors, particle accelerators, even projects to extract Actinium from Thorium stockpiles), but success is not guaranteed. Any setback, a reactor project running years late, technical difficulties in isotope extraction, or geopolitical issues (some isotopes come from reactors in Russia or Europe that could face disruptions), could leave eager patients and drug developers empty-handed. This would significantly slow adoption and could cause investors to lose faith if revenue ramps are bottlenecked by supply constraints rather than science.

• Logistical and Clinical Adoption Barriers: Even with supply in hand, operational logistics pose a risk. Radiopharmaceuticals have very short half-lives (hours to days), meaning the whole chain from production to patient is time-sensitive. Hospitals need special facilities (hot labs, radiation shielding, trained nuclear medicine physicians) to receive and administer these drugs. In areas without established nuclear medicine departments, adoption might lag simply due to infrastructure. Training and staffing are another issue, nuclear medicine specialists are not as common as, say, medical oncologists or surgeons. If healthcare systems are slow to invest in the necessary facilities or if there is resistance among physicians (some oncologists might be unfamiliar or uncomfortable with radiopharma initially), the rollout of new therapies could be slower than forecast. Essentially, the risk is delivery: many patients might benefit, but getting the product to them reliably is non-trivial. This could translate into lower sales than expected for novel drugs in their first few years, disappointing the market.

• Emerging Competing Therapies: The oncology landscape is highly competitive and dynamic. A risk to radiopharma is that some of the same targets being pursued (PSMA for prostate, somatostatin receptors for neuroendocrine tumors, etc.) are also being targeted by other therapeutic modalities. For instance, there are antibody-drug conjugates (ADCs) and bispecific antibodies in development for prostate cancer that target PSMA as well, these deliver a toxin or engage the immune system to kill tumor cells, rather than radiation. It’s possible that one of these approaches yields similar efficacy without the complexity of handling radioactivity. If an easier-to-administer, equally effective therapy emerges for the same patient population, it could cap the market for the radiopharmaceutical alternative. Additionally, conventional external beam radiation technology isn’t standing still, new techniques like proton therapy or targeted intraoperative radiation could address some tumors that radiopharmaceuticals target. While we believe radiopharma will often combine well with other treatments (not just compete), it’s a risk that, in some cases, it may lose out to a more convenient competitor.

Figure 12: While radiopharmaceuticals show great promise, they must compete with existing treatment modalities on both efficacy and safety.

• Regulatory and Perception Risks: Using radioactive materials in medicine naturally attracts regulatory scrutiny. Radiopharma trials have to navigate not just the FDA’s usual hurdles for drugs, but also radiation safety regulations from nuclear agencies. A risk factor is that regulatory requirements could become onerous, for example, requiring additional long-term safety monitoring or complex handling protocols that slow down trials and approvals. Furthermore, any safety scare could taint public and physician perception. If, hypothetically, a late-stage trial of a radiotherapy revealed an unexpected side effect (say, higher rates of secondary cancers years later, or severe organ toxicity in some patients), it could cast a pall over the whole field. Because radiation fears have a visceral element (the word “nuclear” can alarm people), a single high-profile incident might have outsized impact on sentiment. It’s worth noting that one of the companies in our basket, Actinium Pharmaceuticals, encountered an FDA request for more evidence of survival benefit despite meeting its trial endpoints – highlighting that regulators are cautious and want clear proof that benefits outweigh risks. Such hurdles could delay approvals or require costly additional studies.

• Financing & Execution Challenges for Smaller Players: The innovative biotechs driving this theme (aside from the big pharmaceutical anchors) are mostly pre-profit companies that rely on investor funding or partnerships. If the biotech funding environment deteriorates (for instance, due to rising interest rates or a broad risk-off sentiment), these companies might struggle to raise the capital needed for Phase 3 trials, manufacturing build-out, or commercialization. A prolonged capital winter could force dilutive financings or even shelve certain programs, stunting the theme’s progress. Additionally, executing in this field is harder than in traditional biotech – it’s a marriage of pharma and nuclear engineering. There could be unexpected setbacks in scaling production or in quality control (e.g., a manufacturing batch failing to meet purity standards could halt a trial). Companies might find it challenging to recruit enough specialized talent (radiochemists, medical physicists) to support their programs as they grow. In short, the road from promising Phase 1 data to a marketed product is long and fraught with operational hurdles; not all current players will make it to the finish line. For our basket, that’s why diversification is key – but it’s a risk that a couple of the smaller names could falter or even fail along the way, which would hurt the basket’s overall performance if not offset by others succeeding.

• Macro and Geopolitical Risks: Given the global nature of the supply chain (isotopes, equipment, and even patients for trials often cross borders), broader macro risks apply. Trade restrictions or geopolitical tensions could impact isotope supply (for example, some isotopes come from Russia or require Russian enrichment services; sanctions or conflict could cut that off). Currency fluctuations and inflation can affect the cost of building new production facilities or running trials across multiple countries. While these are not core to the thesis, they could create headwinds or volatility in timelines and costs.

Table of Contents

1. Introduction

2. Top 5 Ideas of the Year

3. Worst 5 Ideas of the Year

4. Biggest Fades

5. What This Year Reinforced

Introduction:

This year some ideas worked exceptionally well. Others didn’t. Both outcomes matter.

What follows is a transparent review of the top 5 performing ideas and worst 5 performing ideas published this year, measured from publication to end of year. All links are included so readers can review the previous articles.

A dedicated section for Biggest Fades sits at the bottom, ideas I watched work in real time without ever stepping in. In many ways, these missed opportunities are more painful than the ideas that didn’t work.

Top 5 Ideas of the Year

These ideas shared a common thread: structural tailwinds + mispriced narratives + asymmetric setups.

1. January 26 Calls on LC US +342.1%

This idea focused on a perceived valuation and classification mismatch rather than a directional macro call. The central observation was that LendingClub continued to be valued like a traditional regional lender, despite already operating as a profitable, AI-enabled digital bank with scale.

At the time of publication, LendingClub was originating more than $2 billion in loans per quarter, growing volumes at a double-digit rate, yet trading near tangible book value. The market largely overlooked the company’s underwriting infrastructure, which leverages one of the deepest consumer credit datasets in the industry: over 15 years of labeled outcomes and more than 100 billion data points. Incremental AI capability was being added through small, targeted acquisitions, improving credit performance and operating efficiency.

These advantages were already visible in reported results. Credit losses declined materially year over year, whole-loan sale pricing improved, and margins expanded, even as peers faced deteriorating credit trends. The company’s strong capital position further supported the durability of the business model.

The idea ultimately benefited from the market beginning to reassess LendingClub as a “second-derivative” AI credit platform, one where artificial intelligence was not a speculative future initiative, but a system already embedded in underwriting and economics. As that recognition improved, valuation followed.

In hindsight, the strength of this idea came from focusing on what was already working in the data, rather than what management hoped to build, and identifying where market perception lagged operational reality.

2. The Moat That Closes With Every Mortgage +75.7%

This was not a simple call on falling rates or a housing rebound. The core thesis was that Rocket Companies was being mispriced as a cyclical mortgage originator, when in reality it is in the process of becoming a vertically integrated, end-to-end housing platform.

The acquisitions of Redfin and Mr. Cooper, alongside the simplification of Rocket’s Up-C structure, materially changed the company’s earnings durability and strategic optionality. Redfin added top-of-funnel traffic and a national brokerage footprint, while Mr. Cooper brought scale servicing, recurring cash flow, and customer lifetime value. Together, they create a flywheel for retention, cross-sell, and structurally lower customer acquisition costs, something the market was slow to appreciate.

Operationally, Rocket continued to execute. AI-driven automation improved productivity and lowered costs, market share was gained even in a weak housing environment, and the balance sheet remained fortress-like. Importantly, the thesis also acknowledged macro uncertainty: servicing income and a flexible cost base helped offset higher-for-longer rate risk.

To further balance rate exposure, the idea was paired with copper producers (ERO CN and CS CN), a deliberate hedge against re-accelerating inflation or rising rates. This created a more robust expression across both rate-sensitive and inflation-sensitive regimes, rather than a one-way macro bet.

In hindsight, the strength of the idea came from treating Rocket as an emerging housing ecosystem, not a levered duration trade, and from structuring the idea to survive multiple macro outcomes.

3. Tariff to Table (LS) +74.6%

This idea examined the U.S.–U.K. Economic Prosperity Deal (EPD) as a rare, discrete policy catalyst capable of reshaping cross-Atlantic flows in both agricultural commodities and digital services. Rather than treating the agreement as a headline geopolitical event, the research focused on how specific quota creation, tariff removal, and regulatory harmonization would create winners and losers across adjacent industries.

The framework centered on expressing the downstream effects of regulatory clarity, where new access, cost advantages, or competitive pressure were likely to emerge, while contrasting businesses positioned to benefit from the agreement with those facing incremental margin or valuation risk.

U.S. Agricultural Export Beneficiaries
The EPD introduced a preferential duty-free ethanol quota for U.S. producers and removed tariffs on U.S. beef exports into the U.K. This materially altered export economics for select U.S. agricultural firms. The idea highlighted companies with modern production assets, export readiness, and scale advantages, where incremental international demand could flow directly into margins. Management commentary at the time reinforced that regulatory-aligned international demand was becoming a more meaningful earnings driver.

U.K. Domestic Protein Pressure
On the other side of the trade, the agreement increased competitive pressure on U.K.-centric protein producers. The introduction of tariff-free U.S. beef created the risk of import-driven price competition in a market already facing cost inflation. Businesses with limited geographic diversification and exposure to domestic fresh-protein pricing were identified as structurally more vulnerable to margin compression as supply dynamics shifted.

Digitized Trade Enablement vs. Consumer SaaS
Beyond physical goods, the EPD placed explicit emphasis on digital trade provisions, including paperless customs processes and harmonized digital standards. The research differentiated between platforms that directly enable cross-border commerce, through tariff calculation, VAT handling, and compliance workflows, and consumer-facing software businesses with limited linkage to trade infrastructure. The former stood to benefit from increased cross-border activity and regulatory normalization, while the latter faced the risk of relative multiple compression as capital rotated toward policy-aligned platforms.

In hindsight, the strength of this idea came from treating the EPD not as a macro headline, but as a microeconomic shock that reallocated value across supply chains and software layers. By focusing on second-order effects, who gains access, who absorbs competition, and who enables the system, the framework captured dispersion that broad market narratives missed.

4. Bandwidth Wars (LS) +58.8%

This idea examined the growing bifurcation within the global connectivity ecosystem. While “networking” is often discussed as a single sector, the research argued that the market was increasingly splitting into two distinct camps: companies enabling the next wave of bandwidth expansion and convergence, and legacy vendors tied to slower-growth, increasingly commoditized segments.

Rather than making a directional call on technology or equity markets broadly, the work focused on relative fundamentals inside the connectivity stack, who is positioned to benefit from secular demand for bandwidth, and who faces structural headwinds as architectures evolve.

Enablers of the Bandwidth Expansion
The research highlighted companies supplying critical technology for modern connectivity buildouts, including virtualized broadband access, optical transport, and high-speed datacenter interconnect. These businesses sit directly in the path of several durable trends: DOCSIS 4.0 and fiber upgrades, hyperscale cloud and AI datacenter expansion, and the shift toward software-defined, higher-capacity networks.

Across this group, common characteristics emerged: exposure to customer capex that is strategic rather than discretionary, improving product mix, and demand driven by physics (speed, power, latency) rather than refresh cycles. Product roadmaps centered on next-generation technologies such as virtualized access platforms, 800G+ optical links, and high-speed electrical interconnects designed for AI clusters. In many cases, fundamentals reflected this positioning through stronger growth, margin expansion, and improving balance sheet profiles.

Legacy Connectivity and Structural Pressure
In contrast, the research identified connectivity vendors whose core exposure remains tied to mature or saturated markets such as consumer Wi-Fi hardware, SMB networking, or traditional enterprise campus networks. While some of these businesses are well run and profitable, their end markets face slower growth, heightened competition, and increasing commoditization.

These companies tend to rely on incremental product updates, cyclical refresh demand, or cost discipline rather than participation in the fastest-growing areas of network spending. As customer budgets increasingly prioritize broadband infrastructure, cloud interconnect, and AI-driven capacity expansion, legacy segments risk being deprioritized.

Why Dispersion Matters
The central insight of this idea was that connectivity demand is simply reallocating. Bandwidth requirements continue to rise from the last mile to the core, but value accrues unevenly depending on where a company sits in the architecture. Firms enabling convergence, virtualization, and scale stand to benefit disproportionately, while those anchored to box-based or fragmented niches face relative erosion.

By focusing on this internal dispersion rather than the sector as a whole, the idea captured performance differences driven by technology relevance, customer mix, and capital allocation priorities, rather than broad market direction.

In hindsight, the success of this idea came from recognizing that the future of connectivity belongs to those enabling greater bandwidth, efficiency, and integration, and that legacy hardware models, while often optically cheap or operationally stable, remain structurally disadvantaged as the connectivity cycle evolves.

5. The Skeleton Crew +48.7%

This idea focused on a simple but non-consensus observation: the next wave of automation would not be led by abstract AI models alone, but by physical robots, increasingly humanoid in design, enabled by mature sensors, compute, and motion control systems. The value was expected to accrue not to any single robot form factor, but to the upstream suppliers providing the indispensable “bones and guts” of every robotic system.

Rather than speculating on which humanoid platform might win, the research emphasized infrastructure inevitability. Every general-purpose robot, regardless of brand or design, requires dense edge compute, high-reliability power management, precise sensing, motion control, and rigorous semiconductor testing. These needs exist independently of marketing narratives or end-user hype.

Why the Timing Mattered
The work argued that several forces had converged to push robotics out of the lab and into commercial deployment: persistent labor shortages, rising wages in logistics and manufacturing, improvements in real-time AI inference, and declining hardware costs. Together, these factors created a tipping point where general-purpose robots became economically viable rather than experimental.

Upstream Enablers Over End Products
The core insight was that the most durable value would likely accrue to companies supplying:

• Edge AI compute and robotics software stacks

• Analog, power, and sensing components embedded in every joint and actuator

• Semiconductor test, packaging, and inspection tools required to manufacture advanced robotics chips

• Optical and display technologies that support perception, interaction, and monitoring

These layers benefit regardless of which OEM ultimately dominates humanoid deployment.

Early Identification of the Robotics–Test Nexus
Notably, this research highlighted Teradyne’s dual exposure to robotics well before it became widely discussed. The analysis connected Teradyne’s ownership of Universal Robots and MiR, among the most widely deployed collaborative automation platforms, with its dominant position in semiconductor test. As robot intelligence increases, so does chip complexity, and every advanced processor, sensor, and memory component must be validated. This “picks-and-shovels” exposure to both physical automation and the silicon underneath it was identified early, ahead of broader commentary elsewhere (wink wink).

A Stack, Not a Bet
The strength of the idea lay in treating robotics as a stack. Compute providers, analog leaders, packaging specialists, optical equipment makers, and test vendors all occupy chokepoints with high switching costs and long design cycles. Many of these businesses already generate revenue from adjacent markets (cloud, automotive, industrial), giving them resilience while robotics adoption scales.

In hindsight, the effectiveness of this idea came from separating inevitability from uncertainty. The exact shape of humanoid robots remains fluid. The need for compute, power, sensing, testing, and integration does not. By focusing on the latter, the research captured exposure to the automation inflection without relying on speculative end-state winners.

Worst 5 Ideas of the Year

These ideas failed for different reasons: regime mismatch, narrative persistence, or being right too early (also called being wrong until further notice). Each one highlighted a weakness worth addressing.

1. LS Cash Flow Cosmetics (LS) -18.5%

This idea examined a structural divergence in global beauty, contrasting digital-first, science-led operators with diversified geography against a legacy prestige player reliant on travel retail and wholesale distribution.

The core thesis remains intact, but the idea proved too early. While channel pressure in travel retail and capital allocation constraints were correctly identified, the market proved willing to look through near-term disruption, restructuring, and China uncertainty for longer than expected.

Digital-native and IP-driven brands continue to deepen moats through data, proprietary formulations, and embedded distribution, while legacy prestige models face slower channel recovery and rising competition. The disconnect lies in timing of recognition, not in the underlying analysis.

This was a reminder that structural edges can take longer to surface in consumer sectors, and that patience, not conviction, was the missing ingredient here.

2. Hot vs Cold Storage (LS) -24.1%

This idea focused on a structural shift in data infrastructure: AI workloads increasingly favor low-latency, high-throughput storage over traditional hard-disk–based capacity. The analysis correctly identified flash, high-bandwidth memory, and modern networking as long-term beneficiaries, while legacy disk-centric models faced rising relevance risk.

The issue was timing and relative scale. Hyperscalers proved more willing than expected to extend HDD lifecycles for deep-archive and cold tiers, delaying the point at which latency economics fully dominated procurement decisions. At the same time, early AI-driven storage demand was simply too small, relative to the massive installed HDD base, to force immediate displacement. Capacity roadmaps and pricing discipline from disk vendors further slowed near-term disruption.

That said, the underlying physics have not changed. AI clusters remain constrained by data movement, not raw capacity, and capital spending continues to migrate toward flash, NVMe fabrics, and high-speed networking. This was a case where the destination appears right, but the path is unfolding more slowly and unevenly than anticipated.

3. Rare by Design (LS) -28.1%

This idea was built around a clear structural view: cash-generative, orphan-focused biopharma businesses with pricing power and execution discipline should outperform a broad biotech index dominated by speculative, pre-revenue names. Fundamentally, that view held, many rare-disease companies continued to grow revenue, expand margins, and raise guidance.

The problem was the hedge. In practice, biotech performance over the period was driven less by fundamentals and more by macro flows, sentiment shifts, and risk-on/risk-off positioning. Broad biotech beta moved in ways that overwhelmed relative dispersion, blunting the intended offset and compressing spreads.

In short, the selection was largely right, but the pairing failed in a regime where capital flows mattered more than cash flows. This was a reminder that even well-founded relative frameworks can struggle when sector-wide sentiment dominates fundamentals.

4. A Payments Trifecta -47.3%

This idea centered on high-quality payments infrastructure businesses with strong moats, recurring revenue, and durable compounding characteristics across processing, point-of-sale hardware, and cross-border rails. Fundamentally, the businesses continued to execute.

The primary issue was valuation, not quality. These companies entered the period priced for near-perfection, leaving little margin for error as rates stayed higher for longer and investors rotated away from premium compounders. Even solid execution failed to offset multiple compression.

In hindsight, the key lesson was clear: great businesses can still be poor ideas if paid for at the wrong price. This was not a breakdown in fundamentals, but a reminder that quality does not immunize against valuation gravity.

5. TRUECALLER AB-B -72.2%

This idea was based on Truecaller’s unusually strong engagement (utility-level DAU/MAU), improving mix toward recurring revenue (subscriptions + business), and a belief that a negative earnings reaction was more optics than fundamentals.

What went wrong was not that engagement disappeared, it was that the stock proved far more sensitive to non-fundamental overhangs than expected:

• Reporting optics and incentive-cost noise mattered more than anticipated. Even if economically non-cash, share-price-linked compensation and GAAP volatility created persistent confusion and kept investor confidence fragile.

• Macro/FX sensitivity hit perception. With meaningful exposure to India/MENA/Africa and SEK reporting, currency and risk sentiment became a bigger headwind than the thesis underweighted.

• “Quality” wasn’t enough without a clear near-term re-rating catalyst. The business continued to evolve, but the market didn’t reward the transition fast enough, and multiple compression did the rest.