Are closures good long-term, though?

Obviously, the most effective way to catalyze a multiple here is to just keep chopping wood. Keep driving traffic (blue line) and keep taking 4% price (reasonable in an inflationary environment) and we’ll see profitable growth. Management disclosed on the Q4 call this morning that the next 12 months of closures would be about 200 units, and that those units skew toward $130-135K AUV— roughly half the system-wide average of $256K. This means each closure removes a disproportionately small amount of royalty revenue relative to the unit count it represents; these bad units effectively dilute the accelerating comps story for the company as a whole. The brand-level data confirms this broader theme: Supercuts, the highest-comp brand, is under-represented in total closures (38.7% of closures vs. 46.9% of the base), while the other brands closed at a disproportionately higher share. If this pattern continues, the remaining closures in FY27 should cost roughly $1.6M of royalty on my math. RGS could conceivably accelerate beyond 200 closures, specifically targeting the SmartStyle tail (since that brand is running -4.5% comps and is the most Walmart-dependent, lowest-quality chunk of the portfolio). The constraint is that Walmart’s own lease terms and the $150K revenue threshold referenced on this morning’s earnings call puts a floor on how fast RGS can walk away from those units without triggering buyout obligations.

Now the above is quite important. Ostensibly, these stores should be 100% margin accretive, but they’re not. They can potentially cost RGS more in collections effort, legal costs, bad-debt write-offs on unpaid royalties, or lease-guarantee exposure than the royalty they generated. The 10-K explicitly flags that “some franchisees have stopped paying rents as they come due” and that Regis remains liable as guarantor on certain subleased locations. A bad $130K-AUV franchisee is a meaningfully higher credit risk than the system average, and more likely to fall behind on royalties, more likely to default on a sublease RGS is on the hook for, and more likely to consume collections/legal resources disproportionate to the revenue it generates. Closing these units is necessary to eschew future liabilities and potentially place franchise talent in better setups, or just halt sinking resources/trainings into hopeless efforts.

The other lever is SG&A. FY26’s $4.7M reduction in SG&A (from $46.8M to $42.0M) came from three sources: lower exec comp, lower franchise brokerage expense, and lower education event costs, partially offset by rising company-owned G&A as Alline (wholly owned stores) came onto the corporate balance sheet. Corporate headcount and event spend should shrink with the footprint as further closure means less territory to cover. It seems that so far this hasn’t impaired the franchisee-support function the whole royalty model depends on. The brokerage-expense line should keep shrinking mechanically as well, long as new development stays minimal (only 8 net franchise units were built system-wide in FY26). So there are two small, structural tailwinds there. The offsetting Alline-related company-owned G&A drag should ease now that the acquisition is fully lapped. I think $1-2M of further net G&A improvement is a reasonable ceiling ask for FY27.

Refinancing, finally

This is the catalyst with the most torque over the NTM. Bill Charters, a Sabal Capital special sits veteran who joined the board specifically for this, he even got a shoutout on this morning’s call. Current debt sits at SOFR+9% (roughly 12.5% all-in), with 4.5 points PIK’d rather than cash-paid. Certainly this is a distressed-workout rate from the 2024 near-bankruptcy refinancing. The catch is that eliminating this PIK toggle means higher near-term cash interest even at a better total rate, since today’s structure defers roughly 1/3rd the cash.

A realistic refinancing lands around SOFR+600-750bps, or roughly 9.5-10.5% all-in given current SOFR near 3.5%. It’s crazy to me that to kind of go for this pricing, though. I think this is why Charters is involved. It’s not apples-to-apples as this is a smaller cap, smaller tranche situation, but high yield option-adjusted spread today is like, +300. That would be the sea change that could actually move the needle. Ever 100 bps of savings is almost exactly $1M of FCF that flows directly to shareholders, or another 34 cents a share on a $30 stock.

TL;DR:

I have $28-30M in EBITDA over the next twelve months. I have $120M of gross debt accounting for the PIK. Depending on cash flow let’s say they pay down $20M of that so we see $10M of annualized cash interest. That’s $16-18M of FCF to equity accounting for minimal capex needed. That’s a 20% FCF yield over the next twelve months with ample room to pay down the debt and also re-invest in the brands. AND all of that’s without getting better ROI on the current $42M of SG&A that drives top-line, plus relatively distressed credit (think CCC) in a growth/turnaround environment. I mean, 7% cash interest rate would be a home run, but that seems unlikely. Have you seen a national ad spot for Supercuts recently? Neither have I, so where is this SG&A capital being allocated? How can the company improve asset turns? This is the fundamental question that will make or break the stock.

INVESTMENT ANGLE

InterGroup is a $72M equity value stub sitting underneath a heavily mortgaged San Francisco hotel and a eclectic portfolio of real estate assets. After twice defaulting on the debt tied to its Hilton San Francisco Financial District hotel and drawing going-concern language from its auditors, INTG refinanced in April 2025, returned to listing compliance, and has posted improving numbers for 3 straight quarters as San Francisco tourism recovers. The risks are obvious— heavy single asset exposure, thin trading/zero liquidity, and heavy leverage with the new mortgage/mezz debt both maturing around April 2027.

WHAT IT DOES

InterGroup is the diversified holding company controlled by longtime (42 years!) chairman and CEO John V. Winfield. The hotel exposure comes indirectly — InterGroup holds a 76% economic exposure to Portsmouth Square, Inc., which owns and operates the ~550-room Hilton San Francisco Financial District Hotel. INTG also consolidates some of the debt recourse to the FiDi Hilton via an intercompany loan to the Portsmouth Square entity.

Separately, InterGroup directly owns a portfolio of 16 apartment complexes, 2 single-family houses as investments, 1 commercial property, and 2 acres of unimproved land in Maui.

HOW IT MAKES MONEY

Hotel revenue (room nights, food and beverage, and ancillary hotel services in San Francisco) and rental income from the apartment and commercial portfolio are the two operating revenue streams. Pretty basic stuff. They own the Hilton flag and license it, sell room nights and accommodations.

P&L at the hotel is driven by core KPIs, RevPAR and occupancy. Basically how much they can charge per room, and how many rooms get rented during a relevant period. The other commercial assets are auxiliary to the hotel operating business and don’t contribute to revenue.

DIRECTION OF BUSINESS FUNDAMENTALS

The trend is turning positive after a rough stretch; the Hotel is in the Financial District/Chinatown are of SF which is notorious for having been hollowed out of post-COVID. However, since the new mayor Daniel Lurie took office, there has been a recovery. Hotel KPIs corroborate the recovery: fiscal 2025 occupancy of 92% (ADR $218, RevPAR $200) improved further in the March 2026 quarter to 94% occupancy, $306 ADR, and $287 RevPAR — hotel revenue that quarter actually exceeded the comparable pre-pandemic (2019) quarter slightly.

The property was assessed at approximately $167M in 2024, according to reporting around the hotel’s refinancing. That was ~7% below its 2013 appraisal. The rest of the real estate portfolio earns ~$7M annualized in NOI over the past 8 years. Numbers aren’t apples-to-apples as we have some acquisitions and dispositions. But we care about the numbers on the books today.

We get to a NAV via the following:

76% INTG stake in Portsmouth × (Hotel value + restricted cash − hotel mortgage − hotel other notes (this includes the intercompany ) + (rest of RE portfolio NOI / cap rate − RE mortgage debt) + marketable securities + cash at the INTG holdco level.

The cap rates across the top show what happens when we capitalize the remain real estate assets at a given rate, and the hotel value assumption is down the leftmost column. Basically, it’s a bet on San Francisco. The hotel is the lion’s share of value here, and I think that maybe the hotel is cheap at $167M. So something in the way of $38-$50 in NAV makes the most sense.

RECENT CATALYSTS

The key 2025 catalyst was the April refinancing of the hotel debt: a new $67M first mortgage plus a modified $36.3M mezzanine loan at a fixed 7.25% rate, replacing debt on which subsidiary Portsmouth Square had twice defaulted. The next catalyst to watch is a refi: both the new mortgage and mezzanine loans mature around April 2027 (with one-year extension options), so continued occupancy and rate strength in San Francisco over the next year will matter for how easily that debt gets renewed or refinanced again.

How much better will SF downtown get? How much better can the hotel do? What about the other California real estate. There’s not a lot of NAV upside here for the lack of control and liquidity. But the real estate is there and it seems to continue to appreciate. So there’s some meat on this bone, but it seems like the easy money has been made post-restructuring, and the turnaround story is mostly complete. Would have been a good contrarian bet 2 years ago.

INVESTMENT ANGLE

Ring Energy is a leveraged, single-basin oil producer whose stock trades like an option on spot crude prices. The equity has swings between $2 and $1 and back within the past year. The upside rests on debt paydown, growing free cash flow, and a pivot toward further horizontal laterals in an under-shaled part of the Permian. The bear case is commodity-price exposure, continued shareholder dilution (a $60M of stock sold in May 2026), and a balance sheet still working off acquisition debt. This is a levered play on the Permian, enough said. If you like the Permian it’s hard not to like this:

WHAT IT DOES

Ring Energy is an independent oil and gas exploration and production company focused almost entirely on the Central Basin Platform and Northwest Shelf of the Permian Basin in West Texas, with a smaller Delaware Basin position. Unlike Permian peers chasing shale plays, Ring’s legacy strategy centers on conventional reservoirs, though it has increasingly layered in horizontal drilling with longer laterals to improve well economics.

As of December 31, 2025, REI had:

• 111,714 gross / 96,234 net acres

• 919 gross / 758 net producing wells

• 153.3 million BOE of proved reserves

• Reserves that are 59% oil, 19% natural gas and 22% natural gas liquids

HOW IT MAKES MONEY

The company is a roll-up. It’s biggest deal was when they acquired Lime Rock. RE acquired 17,700 net acres in Andrews County / Central Basin Platform (CBP). The assets had approximately 2,300 BOE/d of production, >80% oil, from ~101 gross wells. Management described them as “low-decline assets”. The final consideration was approximately $79.3M cash and 6.45M REI shares, so roughly $100M total consideration. And this was another bolt-on to REI’s existing CBP position:

The downside is they’re mostly hedged through ‘26, but the hedges mostly roll off in 2027. So it’s low cost reserves and the hedge book will be gone next year.

DIRECTION OF BUSINESS FUNDAMENTALS

Company just lifted guidance and still remains at ~half its PV-10. There’s virtually no sell-side coverage here, even though production has doubled in the last 5 years (through acquisitions). The FCF has been fantastic, but as with every O&G play they’re subject to spot pricing and their hedge book, which is repressing NTM FCF.

RECENT CATALYSTS

There’s a good story about being a non-capital return play here in the Permian. Everyone in O&G is in capital return mode at this point, and the wildcatters are the one putting more holes in the ground. They are putting almost 100% of their OCF into capex over the next twelve months by their own admission. So it’s not for the faint of heart.

You really want hedges to roll off and the capex story to be cleaned up. Management raised second-half 2026 capex guidance to $80–100 million and targets ~10% production growth in 2027. But its a wildcatter, so we’re not really expecting much in the way of discipline. The $60M raise in May funded a revolver paydown, cutting leverage to ~1.7x and lifting liquidity to about $226M.

I really like this stock today, actually. We need people drilling!

INVESTMENT ANGLE

Culp is trading near multi-year lows after a two-year restructuring program and losses. Longs are banking on $22.5M in newly realized cost savings finally showing up in margins and a bedding segment that’s growing again; the bear case is typical shitco stuff: continuing negative EBITDA in a housing- and furniture-sensitive end market with no clear demand recovery yet (CULP cites housing starts).

Microcap activist 22NW first became involved in 2024, obtaining a board seat. In June 2025, Culp and 22NW entered a new multi-year agreement that gave 22NW continued board representation and resulted in additional independent directors and the creation of a strategy committee. They’ve made a lot of progress on the margin profile of the business.

WHAT IT DOES

Family-founded in 1972 and based in High Point, NC, Culp manufactures and sources mattress fabrics and sewn covers for the bedding industry, plus upholstery fabrics for residential and commercial furniture. It operates a global manufacturing and sourcing footprint spanning the U.S., China, Vietnam, Turkey, and Haiti, with roughly 900 employees. All design and manufacturing facilities are in the US, but they source materials from various countries outside the US.

In 2020, Robert G. Culp IV became CEO. He’s the grandson of the founder and represents the third generation of the Culp family to run the company. He joined Culp in 1998 and worked his way through the organization.

HOW IT MAKES MONEY

Revenue splits ~evenish into two segments: Bedding ($116.6M, 57% of 2026 sales) and Upholstery ($86.9M, 43%). Customers are mattress and furniture manufacturers who buy fabric as a direct input; Culp holds an estimated 20-25% share of the roughly $500-600M domestic mattress fabric market and 8-10% of the larger $1.5-2.5B upholstery fabric market. About 69% of sales are U.S.-based, with the balance international. Its upholstery customers are companies like La-Z-Boy, Ashley Furniture, and major bedding manufacturers on the bedding side.

The value chain gives Culp some bargaining power but it’s still a spread business, which is why they have sourcing operations all over the globe. Culp buys yarn/materials → manufactures fabric → sells fabric by the yard/sells sewn covers (end products)→ mattress manufacturer uses it in mattress → Culp gets paid. Its bedding fabrics generally sell for ~$2 to >$18/yard, depending on the product. Upholstery fabric has a higher pricing floor (~$5-15).

DIRECTION OF BUSINESS FUNDAMENTALS

Fiscal 2026 revenue was $203.5M, down 4.6% year over year, with adjusted EBITDA of $-4.7M — below the threshold needed for executive incentive comp. Bedding sales rose 2.4% for the year and accelerated to +12.5% year-over-year in Q4, while upholstery fell 12.5% for the year on soft home-furnishings demand. The company no longer breaks out segment-level operating income after consolidating its divisions into one platform.

While it’s a bad business gross margin-wise, but there’s evidence that SG&A is bloated ($34.7M or 17% of revenue) and that operating leverage in 2021 resulted in peak earnings. And the company is dead serious about cutting costs and capturing operating leverage:

The strategy has become own strategically important U.S. manufacturing + Haiti cut/sew operation, outsource commodity production, retain design/product/development, finishing, QC and customer relationships in the US.

RECENT CATALYSTS

There’s a lot going on here and I really like the stock. Unfortunately it’s way too illiquid.

• They’ve got some tariff receivables from the government and manufacturing

• They’ve got $40M in real estate ostensibly not being put to its highest and best use

• Plenty of NOLs to shield EBIT

• Pricing continues to move higher despite competition? Very interesting for a commodity business.

I like the direction of the green arrow. I mean there’s a lot to like. Activists are still involved, the results are still coming, and the company wants to continue to push pricing. It’s one of those companies you just want to buy after reading their investor deck and understanding they have a good grasp on everything. And they’re at the bottom of their cycle. But I would reiterate- it is way too illiquid for most anyone.

It is my quarterly reminder that 13-Fs are useful to look at, particularly folks that have an articulated, long-term strategies for unlocking idiosyncratic value in undermanaged assets!

As we continue the RYAM saga post-earnings, we can look at how event-driven/activist folks that are involved tweaked their positions. Previous posts can be found here.

Briefly, we left on a few key notes:

• Management is currently posturing to either maximize value in a negotiation by pumping their pricing/operating leverage. This is what amounts to a default management position (“all is well”), but its worth noting the new CEO is a turnaround guy.

  • I would echo the Mill Pond point that the board has had 10 years post-spin to figure this out and they haven’t

  • The entrenched board is obviously biased toward managing independently and needs to hear from shareholders that their time is up

• Replacement cost, constrained capacity, and potential acquirer AIP’s ability to integrate the assets (particularly in this tariff environment) show that the company is undervalued.

• Their profitability sucks, but there are green shoots on the Cellulose Specialties side of the business, which drives over 100% of incremental EBITDA.

So the key here is convincing the board that they’re doing a bad job and need to sell. But the board also let CEO Scott Sutton walk in April in light of the board-initiated the strategic review. Sutton left equity upside on the table, meaning either he thought the bid would come to fruition, or the turnaround plan was helpless. I don’t think it was the latter based on his earnings call commentary in March:

RYAM reported a 2025 net loss of $420 million a week after private equity firm American Industrial Partners disclosed it made an offer to buy the company in November, before Sutton joined the company in January.

RYAM’s board of directors rejected the offer but AIP said it was still interested in RYAM if the board changes its mind.

Sutton was asked about that offer by an analyst but he didn’t want to talk about takeover interest.

“I’m probably not going to comment on a specific shareholder or any specific offer except what I will say is, we have plans that will deliver substantially more value,” he said.

“You are going to hear more about those plans as we’re updating them in upcoming earnings calls.”

So it’s a schizophrenic situation. Long story short, is that the stock is going to trade on its underlying business without any M&A premium. Any substantive incremental problems at the core specialties business are going to jeopardize the M&A position, so it’s going to get baked in anyway. The stock can and should trade as if the bad balance sheet/overloaded SG&A will stay as is, but with maybe some upside if the company decides to exit the paperboard/HYP business.

Of the activist funds involved as of 6/30, a few have made substantial changes to their positions. This would have been cut-off after the Mill Pond letter but obviously before Q2 results came out. I’ve cherry picked the list, so it’s not super comprehensive, just a few flags for an update on the situation.

Exits

• Irenic Capital Management LP — sold its entire position, -1,356,166 shares. Fans of this newsletter will know that I follow Irenic’s 13-F closely as their style most aligns with mine. And they’re confirmed 130 IQ+ investors.

  RYAM is a little small for Irenic. They’re doing SNAP, they made a bid for RSVR take-private. That’s probably not a great cope for why they would want to exit RYAM, because Irenic is built to press for exactly this kind of outcome for a public company, and they’ve chosen to get out. So it’s probably an indictment of the current setup, but it could also be chalked up to “it’s complicated and won’t drive P&L for the firm”.

• Park West Asset Management — full exit, -442,000 shares to zero. $1.5B in AUM. they run event-driven/merger-arb strategies, but also have a large cap generic L/S book.

• Saddlerock Capital Management — full exit, -512,482 shares to zero. Not a lot on this firm outside of the tagline on their website “Compounding Capital with Transformational CEOs at Points of Change”. I imagine they must have been banking on new CEO Sutton being involved in the turnaround.

Unchanged

• Condire Management LP — 6,358,877 shares, 9.4% of RYAM outstanding, the largest shareholder in the entire register. Zero share change in Q2. Condire is a deep-value/special-situations shop. They are 13/G and not 13/D. They’re sort of riding the wave here but I would speculate they support a full sale.

• Maple Rock Capital Partners — 3,256,889 shares, 4.81% of RYAM, unchanged. Maple Rock runs credit/special-situations books.

• AIP LLC / AIPCF VIII Cayman Ltd — 3,400,000 shares, 5.03%, unchanged. They’re hanging around the rim. I’m not sure they’re obligated to hold their stake but I imagine they’re ride-or-die with the takeover.

• ****Mill Pond (the 13/D filer)— does not file a 13-F and as such is Unchanged from their “3% of shares outstanding” claim they made in June.

Incremental Adds/Buys

• Nantahala Capital Management — 1,697,726 shares (2.51%), +97,726 shares this quarter, roughly a 6% increase to the position. Nantahala runs merger-arb/special-situations strategies, so they were always playing for a full sale and not really the pure turnaround. So it makes sense they add after the Sutton exit.

• Governors Lane LP — Ex-Eton Park folks. Brand-new position, 306,700 shares. Governors Lane is a credit/special-situations investor; a new entrant during an active strategic review is worth noting.

• Rangeley Capital Partners LP — Shows up as a new position, 104,808 shares, representing 0.75% of their own book. Activist/event-driven guys who are sharp!

Incrementally, the Irenic exit is the most significant. I could be convinced that the stock does have an M&A premium in it today depending on how dire the view of the balance sheet is or what run-rate/peak EBITDA margins look like. And the response to seeing M&A premium already in the stock could mean ditching it because you don’t like the fundamentals.

INVESTMENT ANGLE

Amtech is a semicap equipment maker riding the AI infrastructure buildout. AI-related orders for its advanced packaging reflow ovens jumped roughly 120% y/o/y last quarter and now exceed 40% of its TPS (thermal processing solutions) segment’s sales. A very cute infographic below:

The bull case is a further operating leverage as a market share gainer in TPS reflow for OSATs doing advanced packaging of AI chips. These are big, concentrated customers who have balance sheet capacity (ASE/Amkor). One repeat OSAT (Outsourced Semiconductor Assembly and Testing) customer ordered over 300 reflow units in the TPS segment in 2023. Top line is accelerating, and the company has guided to higher margins this year. Typically this means “buy the stock regardless of anything else”.

The bear case is that the stock already ran hard on this narrative and a recent equity raise diluted shareholders by about 29%. It’s not a thin field of competition either, there are both European and Chinese competitors that compete on price and product specifications. Specs/qualifications create a high barrier to switching for customers, as trying to hotswap ovens will mess with yields and product outcomes. So there’s stickier older business, and fierce competition for winning new business.

Strong near-term revenue visibility is great (Q3 call mentioned further visibility through mid ‘27), but long-term, there could be a thinning pricing power narrative/share risk concerns if cost competition becomes a thing. Good for them getting the secondary deal done. I am not a semicap specialist but I would estimate the business might be coasting on the “everything bottleneck” in semicap equipment, but the numbers are the numbers and they’re growing with the industry.

WHAT IT DOES

Amtech Systems makes capital equipment used in semiconductor and electronics manufacturing. Its main subsidiary, BTU International, builds thermal processing (reflow) ovens used to solder and package chips, including advanced packaging for AI and high-performance processors (GPUs and TPUs). A second, smaller segment (Tempress, P.R. Hoffman) supplies wafer polishing, dicing, cleaning equipment and consumables, along with legacy solar cell diffusion furnaces.

You can think of these reflow ovens as long tunnel furnaces with multiple zones that carry a circuit board through a controlled heating-and-cooling profile that melt the solder paste and tiny solder balls under a chip so they fuse into permanent electrical/mechanical joints between the components and the board, then cooling it evenly enough that it doesn’t warp before the joints solidify.

HOW IT MAKES MONEY

Revenue comes from selling reflow ovens and fab equipment to semiconductor device/module makers and OSAT providers, plus recurring consumables, spare parts, and service revenue tied to its installed equipment base. BTU’s Pyramax and TrueFlat reflow ovens are positioned for high-throughput, high-density advanced packaging.

Consumables and service revenue related to the reflow ovens amount to 20%~ of TPS segment revenue. Overall margins have widened since the AI boom got started on discontinuing low-margin legacy lines and shifting mix toward AI-advanced-packaging SKUs that carry higher margin by product design and incremental service sales.

DIRECTION OF BUSINESS FUNDAMENTALS

Fiscal Q3 2026 (reported early August): revenue of $22.4M, up 14% y/y and above the ~$20M estimates. Gross margin rose nearly 400 basis points to 50%. TPS strength was partly offset by continued softness in Semiconductor Fabrication Solutions, which actually contracted y/y due to cratering demand for SiC (silicon carbide) products. The company guided Q4 fiscal 2026 revenue to $22.5-$24M. Cash and equivalents stood at $83.1 million, boosted by the June 2026 offering. Management on the latest call mentioned potential M&A as a use for the cash on the balance sheet:

While we’re pursuing additional growth capital -- while we weren’t pursuing additional growth capital, we felt the timing was right to opportunistically bolster our balance sheet in anticipation of opportunities to supplement organic growth with synergistic acquisitions.

About 46 cents of every incremental revenue dollar in TPS is flowing through to EBIT, with 30 cents going to R&D and SG&A. Based on the guide, we’re looking at ~$12M of EBITDA on $86M of revenue. The continuing trajectory of incremental margin flowthrough could mean something like $20M of EBITDA on $120M of revenue or higher; if we’re early innings that could widen out massively. If not, we’re at ~15x EV/EBITDA at peak earning power at the top of a cycle. That’s why the chart looks the way it does.

RECENT CATALYSTS

The $60M offering probably means mostly likely M&A, or greenfield capacity for new TPS products. Management also plans to unveil next-generation, higher-density packaging equipment at SEMICON Taiwan in September 2026, a potential demand catalyst to watch.

INVESTMENT ANGLE

UTMD is a profitable, debt-free niche medical device maker trading near ~multi-yearish lows after two straight years of customer-loss-driven revenue declines. The company has been in a post-COVID de-rating, de-growthing process. The key question is whether the OEM/distributor losses are a one-time reset or an early sign of eroding competitive position; I lean toward the former.

Losses in BPM are the big driver here. As a percent of the dollar revenue drop from peak 2021-2022, PendoTECH and BPM kits have been 90% of losses vs. other product lines, offset by growth in other areas. The bet isn’t necessarily winning back the customers in BPM, it’s that the losses are now fully lapped (2026 comps no longer carry that drag) while the stable, high-margin core continues growing modestly.

From the 10-K on revenue losses:

Looking forward to 2026, UTMD expects at a high level of probability that it will not obtain any revenues from its previous OEM customer PendoTECH or its previous China distributor for BPM kits. The combined 2025 revenues to those two customers were $2,458, or 6.4% of consolidated 2025 sales. Management is focused on obtaining revenue growth in other areas to offset those losses, and achieve the same revenues in 2026 as in 2025. But the projection of new revenue growth is at a lower level of certainty than the projected losses. So revenue stabilizes or grows ~slightly from here off a lower base.

Also, a big bet here is what they do with their cash. It seems like with their history they’re likely to be acquisitive.

WHAT IT DOES

Utah Medical Products, founded in 1978 and based in Midvale, Utah, designs, manufactures, and markets specialty disposable and reusable medical devices. Its core hospital-facing product lines cover labor and delivery/obstetrics (vacuum-assisted delivery systems, fetal monitoring accessories), neonatal intensive care (umbilical catheters, feeding tubes, respiratory devices), gynecology/urology, electrosurgery, and blood pressure monitoring/blood management. Branded devices are sold directly to hospitals and through distributors in the U.S. and ~40 other countries.

UTMD rode the COVID-era vaccine and biologics manufacturing boom. Biopharma customers over-ordered single-use bioprocessing components during 2021–2022 to secure supply amid shortages, then sharply cut orders as they worked down inventory once capacity fears eased.

HOW IT MAKES MONEY

Revenue comes from two streams: branded device sales to hospitals and distributors worldwide, and OEM/private-label manufacturing, where UTMD contract-manufactures specialty disposables and other devices. Many products serve narrow clinical niches with limited competition; certain FDA-cleared designs are costly to replicate, leading to strong margins and pricing. UTMD has a small direct sales force supplemented by international distributors.

UTMD’s marketing focus is with clinicians, primarily through clinical meetings, trade shows and the internet. In competitive bidding processes, UTMD works with administrators who are responsible for hospital purchasing decisions. UTMD sells on 2 things: design and reliability. Despite having competitors in every vertical, UTMD doesn’t compete on price. There’s a great (but blackpilling) paragraph in their 10-K explaining why:

The longer-term overall cost of care in the U.S. will continue to increase, with quality of care lower, as innovative suppliers are excluded from participating in the marketplace as a result of unnecessary regulatory and other purely administrative burdens. The length of time and number of administrative steps required in evaluating new products for use in hospitals has also grown. As a potential negative factor to future performance, as UTMD introduces new products it believes are safer and more effective, it may find itself excluded from certain customers because of the existence of long-term supply agreements for existing products. UTMD may also be unable to establish viable relationships with other medical device companies that do have access to users but lack an interest in the Company’s solutions, or demand too great a financial or administrative burden.

DIRECTION OF FUNDAMENTALS

One cool quote from the 10-K: “UTMD was formed as a Utah corporation in 1978. UTMD sold stock to the public one time in 1982 for $1,750 (before offering costs of $321). Since 1992, UTMD has returned $156 million in the form of share repurchases, and an additional $90 million in cash dividends, to its public stockholders.”

Because UTMD didn’t cut fixed manufacturing overhead as fast as biopharma OEM volume fell, margins compressed on top of the revenue decline post-COVID. Management flagged this explicitly going into 2024 as biopharma OEM sales were projected to decline significantly without a comparable decline in overhead. The magnitude shows up in their EBITDA trajectory, with the peak in 2021/2022, now with LTM EBITDA of $12.3M.

RECENT CATALYSTS

The revenue decline traces mainly to the loss of two large customers: OEM pressure-transducer sales to biopharma customer PendoTECH and a China-based blood-pressure-monitoring distributor. Replacement revenue from new OEM biopharma customers is ramping more slowly than management expected. Separately, ongoing Filshie Clip product liability litigation has elevated legal expenses through 2026, though management expects those costs to ease in the second half. The company continues to pay a quarterly dividend ($0.31/share) and repurchase stock through the downturn. Nothing imminent on the horizon, just continuing execution.

INVESTMENT ANGLE
Kewaunee offers rare exposure to lab-construction and biocontainment equipment in a sub-$150 million market cap, with the NuAire deal being a transformative, revenue-diversifier beyond project-based furniture into recurring equipment sales. The bear case: declining backlog, margin compression post-acquisition (+ general digestion issues with NuAire), and reliance on lumpy, timing-sensitive projects make near-term results volatile and hard to forecast. I would emphasize lumpiness here, almost exclusively because big labs do big projects intermittently

I like slides like this though:

WHAT IT DOES
Kewaunee Scientific designs, manufactures, and installs laboratory, technical, and biological safety furniture and equipment. Headquartered in Statesville, NC, with manufacturing there and in India, the company sells to life science, pharmaceutical, education, government, healthcare, and general research customers worldwide. Core products include steel and wood lab cabinetry, fume hoods, modular and movable workstations, epoxy resin countertops and sinks, and technical furniture systems.

In November 2024 the company acquired NuAire, a Minneapolis-based maker of biological safety cabinets, CO2 incubators, ultralow freezers, and pharmacy compounding isolators, for $55M. NuAire is the North American share leader in biosafety cabinets, and the deal broadened Kewaunee from furniture into biocontainment equipment. NuAire brought higher-margin, highly recurring, specialized equipment revenue into KEQU’s lumpy/capital-intensive furniture manufacturing business.

What’s cool about the capital allocation here is that KEQU paid off the seller-financing notes for $23M post-close. The deal was done entirely with debt and cash, and a chunk of that debt was already paid off.

HOW IT MAKES MONEY
Revenue comes from designing, building, and installing custom laboratory casework and equipment for large projects like university labs, pharma facilities, hospitals, government research sites. They also do smaller equipment sales through distributors and direct channels. The business runs on long-cycle project orders tracked through an order backlog, plus a growing product-sales stream from NuAire's biosafety equipment. Two reporting segments — a domestic Lab Products Group (LPG) and an International segment — reflect this mix of U.S. project work and overseas (including Middle East and India) construction-linked sales.

KEQU sells through three channels. 1.) is independent dealers that showroom and have relationships with certain universities or hospitals. 2.) is general contractors that handle the new construction and come to KEQU for their “lookbooks” that can be sold to end customers. And 3.) is wholesale soup-to-nuts turnkey projects, especially internationally.

This does imply to some correlation with the research lab/drug development lab capacity, of which we are amply supplied. Vacancy rates for Boston, NYC, SF, and San Diego range from 20-30%. So the idea of an imminent grand build out of lab space isn’t really on the horizon, but remodels are possible.

FINANCIAL TRAJECTORY
Fiscal 2026 (ended April 30) sales rose 17.3% to $282.0 million from $240.5 million, driven largely by NuAire's first full year of contribution. EBITDA was flat year-to-year. It’s a good illustration of the lumpiness to just look at their slide deck for this:

RECENT CATALYSTS
Year-end backlog stood at $165.9 million ($118.8 million LPG, $47.1 million International), down from $214.6 million a year earlier, partly due to the cancellation of two India projects. Management has pointed to continued NuAire integration and balance-sheet improvement as priorities heading into fiscal 2027, but the debt shouldn’t be a deal breaker.

The two major cancellations in India tanked the stock. This stock is cheap if you think we’re at a cyclical trough and that the backlog shrinkage can be won back. It’s expensive if you think lab space is oversupplied.

Inside you are two wolves:

1.) I love volatility. It’s where the opportunity is. Retail get shaken out of trades and even veteran PMs have to sell down when autocorrelation picks up and momentum takes over. It’s how all your money is going to get made— buying into volatile situations created by non-rational actors. If your stocks are taking a pounding, it’s just folks being irrational, be greedy when others are fearful!

2.) I hate volatility. It’s scary, might get me fired, and there’s a very real psychological cost to waking up to +5%/-5% swings that can drain you and destroy your marriage. If your stocks are taking a pounding, it’s evidence that you’re a bad PM! Losers average losers. Winners average winners. Just buy the stuff that goes up. It’s so easy to trade, just sell the bad stuff and buy the good stuff.

We all wake up and deal with this. It’s not fun, but can be mitigated by hard and fast “rules” that you can enforce upon yourself. Most benchmark-huggers have some kind of rule for how much active share they take, and all pod shops have centralized risk books that will give you the axe if you’re not judicious with risk.

When we traffic in Shitcos, all of these problems get amplified. Liquidity dries up, bad news gets priced in a matter of microseconds, and companies just kind of sit there and do nothing. As perennial failures, these companies are used to just totally blowing up and not caring. We talked about this, but it’s worth repeating that these things are 50-60 vol instruments. The wild swings come with the territory.

Here is a chart of TLT:

You can see that today’s dramatic rally is in no way a break from the trend— we are seeing higher interest rates for longer. And it hasn’t really affected the risk curve just yet. But it should— equities should be priced off of long-term treasuries. It’s financial gravity, and it’s been talked about by academics and operators ad nauseum and will be until kingdom come. We can’t sustain record high equity multiples with long rates at multi-decade highs without earnings growth. Now, as of today, we are seeing lots of earnings growth at the top of the market. Which is why you’re not seeing an imminent crash. Staples/low growth names are trading sideways, while the AI trade carries everything else. Pernod Ricard, a blue-chip spirits business with a 200-year history, is trading at 12x forward EPS.

Academics would tell you that shitcos are at the furthest end of this risk spectrum. Individual names with heavy leverage, illiquidity, and low multiples will get crushed even further in an environment with higher rates. Debt service costs go up, terminal multiples contract, and volatility goes up as distance-to-default gets narrower. At a basket level, it’s really bad. We continue to see 1-day -30% drops in names that got crowded and overvalued. It’s not fun to wake up to that stuff.

However, one interesting thing to think about, particularly for shitcos with lowish-close-to-ZIRP-era-coupon issuance out there, is that debt buybacks in the open market become something of an interest-alignment exercise for management teams. Buying back your debt in the market at a discount to face value is a great capital allocation decision that creates equity value AND it’s also a way of entrenching management teams and keeping overpaid board members. Finally, alignment! God forbid a creditor committee get involved and rationalize some costs.

Buybacks and dividends don’t necessarily align with managers and boards. It sucks but it’s the truth. They want to use the cash to empire build with dumb acquisitions, or overstaff themselves, or pay their boards more. It’s grim, but there’s a reason these shitcos stay shitcos. It took B. Riley a near-bankruptcy and distressed asset sales to eventually buy back their bonds at 80 cents. Some of these other names have debt trading in the 70-80 cent range. Some of the energy businesses out there are battening down the hatches even in a good environment for E&P investment. Make sure you’re setting your Google alerts for debt buybacks and “gain on extinguishment of debt” announcements!

INVESTMENT ANGLE

The appeal is New York City, baby! The assets are primarily in Brooklyn and Queens, with some stuff on Long Island. The land and improvements are held largely debt-light relative to its likely market value ($~30M total of debt vs. $300M of mkt value), at a crazy discount to market and to peers. The play is waiting out the family or hoping for a cash grab from the younger generations. The risk is, of course, thin trading volume, tight family voting control that limits any outside influence on capital allocation or a sale, no dividend, and poor operations resulting in high costs relative to peers.

So you’re playing for a change of control here in the next 50 years.The good news is the real estate is probably worth 3-4x the current EV of the company. And you have the potential conversion to a REIT if you somehow wrested control from the family. A classic piggy bank NAV story that will never be able to cracked open (maybe?!). It’s sort of like owning a Chinese SOE, except those sometimes pay dividends to minority owners.

WHAT IT DOES
J.W. Mays, Inc. is a Brooklyn, New York-based owner and operator of commercial real estate. Founded in 1924 as a department store chain, the company wound down its retail operations in the early 1980s and redeployed its capital into owning income-producing property. It now holds roughly nine buildings totaling about one million square feet, concentrated along their crown jewel assets in the Fulton Street corridor in Downtown Brooklyn (including 9 Bond Street and 540 Fulton Street), with additional properties in Jamaica, Levittown, Massapequa, Fishkill, and Circleville, Ohio (weird). The company has been controlled by the Shulman family for a century through a multi-class share structure that preserves majority voting control despite modest insider economic ownership.

HOW IT MAKES MONEY
Revenue comes entirely from leasing retail, office, restaurant, and other commercial space to tenants under operating leases. Rents are largely set through typical long-term leases with normal escalators rather than short-term market pricing, giving revenue a degree of stability but limited near-term upside. The company does not develop or flip property; it is a long-term holder, and many of its Brooklyn parcels have been owned for decades, meaning book value understates current market value in a borough where commercial land has appreciated substantially.

FINANCIAL TRAJECTORY
Trailing twelve-month revenue is about $22M, down modestly from the prior year chiefly on vacancy. The company runs at 60-65% occupancy, which is quite terrible. Cash flow is fine, but taxes, insurance, and operating costs have outpaced rent growth. There’s maybe 200K square feet of space that is underused/unused. The balance sheet carries about $30M in debt against $3M in cash. No dividend is paid.

Commercial real estate values continue to march higher in the outer boroughs despite the headlines about “Mamdanistan”. The Brooklyn assets alone on downtown Fulton are worth probably $200M vs. the current EV of $112M. The rest of the portfolio probably adds another $100M ish to it. So we’re talking about buying at 1/4thish of NAV today. Which, for some, is enough to tuck away for a rainy day.

RECENT CATALYSTS
In April 2026, control of an adjacent 112,000-square-foot commercial condo block at 486-496 Fulton Street traded from Pimco to Gindi Capital and Crown Acquisitions, with J.W. Mays remaining landlord of record on the underlying lease. This is where we get the current mark-to-market on the Fulton assets and the corresponding stock spike in April of this year.

Separately, the company renewed a lease at 504-506 Fulton Street with monthly rent stepping up from $30,188 to $34,716 through April 2031. So some green shoots there.

In retail, it is paramount to always cater and retain your core customer. You cannot alienate the demographics that built your business. A good example of alienating one’s customers is Ron Johnson, the pioneer of the Apple Store that went to take on the CEO role at JCPenney. This was a famous failure and is cited as sort of the archetypal backfiring of the famous Henry Ford quote, “if I asked my customers what they wanted they would have said faster horses.” An automobile is, functionally, a faster horse! Or it at least fulfills the same role! All companies should listen to their customers.

Johnson was doing interviews this month talking about the “vision” being fine, just the “implementation” being bad.

This is a difference without a distinction. You lost the core customers, doesn’t matter how or why, or if there were more runway, the core customers left. The JCP analog is important to think about here, because there’s a lot of overlap in demography with Lands’ End. They primarily cater to women aged 45 to 65. The broader overall median customer age sits slightly above 50, with shoppers over 55 making up approximately 45% of sales. Usually, these are high-income, suburban purchasers with average HH incomes exceeding $100,000. I think JCP skews toward lower income, but the age cohort is similar.

If we think about this in terms of a turnaround, JCP was a lazier thesis. Basically, “what if we took the sales per square foot of the Apple Store and slapped that multiple on JCP?”. This eliminated big draws, like couponing, sales, and discounting.

For Lands’ End, it’s different. For one, there’s no virtually no store footprint. One of their advantages in moving to the WHP licensing model is that their brand history is catalog-first, digital-second, with physical stores a distant third. Their core competency in building brand IP and consumer trust without compromising on quality will hopefully travel overseas. And that’s sort of why it’s not really a “turnaround story” per se, it’s more like “new management is here, and they want to grow overseas and with US Xillennials without alienating the core customer.”

There’s no doubt that the Wawa collaboration was inspired by the famous Trader Joe tote that has been taking the world by storm. It’s a hit on the socials but it’s a limited release, so don’t expect much in the way of real top-line impact. But the point is that this is something that can be relatable for a new demographic without trashing the brand value:

It’s why I think we’re at an inflection point in the stock. Eddie Lampert is off the case, his CEO is gone, and Charlie Cole/WHP are using their expertise to grow the brand domestically, not just overseas. And it’s showing up in the web traffic data too:

I should note that my Alt Data expert has warned that the Bloomberg Second Measure data is functioning poorly and it’s best to look at other indicators such as web traffic; above is the Similarweb data.

An interesting angle on the Lands’ End summer catalog is how the Lands’ End demographic thinks in terms of modesty. “Can I wear this to the beach with my kids/parents/grandkids et al?”, or “am I going to want to spend a whole day outdoors in a revealing outfit?”. One of the core takeaways that I get when I talk to the people buying Lands’ End product is that they make quality stuff, but they also make modest stuff, like their swimsuits and outerwear. In the age of Instagram, not everyone wants to “naked dress” in sheer outfits and stringkinis. And fewer brands even stock more modest swimsuits; I think this is one of the core reasons web traffic spiked so high this summer. There is a very real (probably older) customer out there that has a new GLP-1 body and wants to go to the beach, but also wants to leave something to the imagination. For these other digital-native brands, it’s hard to launch a modest product if you’re a brand that already sells to the stringkini demographic. So there’s a niche here for people who don’t like athleisure or revealing outdoors gear, in my view.

I belabor the point, but the stock won’t work without growth. And it will be key in this earnings call to see how they’re growing with the younger American cohorts, since the WHP ramp-up for sourcing and distribution in overseas markets will take time to get going. Overseas is certainly a growth avenue and story I care about as it relates to the story long-term, but it’s still a small chunk of the business. The market will want to see that the brand still has power domestically.

To summarize, there’s a 3-component value prop to the name:

• The first is the Lands’ End stake in IPco, which is the 50/50 JV with WHP Global valued at $600M. This is a multiple of 12x the GMR rate guaranteed, probably closer to a 10x factoring in forward growth and the guidance that LE’s adjusted EBITDA would incorporate $30M of EBITDA from the JV.

  • WHP Global paid $300M for a 50% controlling interest in the IP joint venture.

  • While $300 million cash for 50% implies a baseline $600M total valuation, WHP purchased a controlling operational interest. Under GAAP deconsolidation rules, when a parent company derecognizes contributed intangible assets and retains a non-controlling 50% equity stake, the retained interest is measured at its fair value on the closing date.

  • The fair value of Lands’ End’s retained 50% interest—incorporating discounted future cash flows from guaranteed minimum royalties plus external third-party licensing projections plus option value—was appraised at ~$375.8M.

• You can think of the exchange option is as being marked on the books at $75M, but I just want to illustrate that there’s upside to that. There’s a minimum multiple at which the stake in IPco converts to WHP Topco stock, which is 13x, so it’s good to think of these as barometers for where stuff should trade in this royalty model. So you’re getting a kiss-up on multiple, and some of the growth in LE.

• The third component is the remainco uniform business, the fulfillment infrastructure, the physical minimal locations, and other physical assets. We can back into the remainco EBITDA via guidance:

  • The company’s reported/guided Adjusted EBITDA ($68M–$78M for FY2026) includes Lands’ End’s 50% equity share of the JV’s net profits.

  • Conservatively, this means probably $40M in EBITDA from remainco

  • Give it a 5-6x multiple and we get $200-240M, add this to the $375M and we get $615M, across 30M shares means ~$20 share conservatively.

OK so it’s dirt cheap on conservative assumptions. What will actually get investors interested in the stock?

1.) Emphasis on Eddie Lampert’s exit. This is the emphasis here- despite him being the controlling shareholder of LE, he has constructively sold the brand to WHP, who are going to take the brand global as managers of IPco. Andrew McLean left in June, and outsider Charlie Cole was brought in as CEO.

Cole has already done his first “collab” with Wawa, a popular gas station/sandwich brand in the midatlantic:

I think the fit is cute. It’s important to do these refreshes domestically too to grow the brand with millennials and gen Z. It’s sort of a boomer/late Gen X Americana brand, but it has to room to grow. Their sheets (which can be custom monogrammed) are some of the highest rated across price points.

2.) The overseas opportunity. Post-World Cup and America 250 there’s a growing positive American sentiment overseas, and its the perfect time to launch product in Europe. European eCommerce was +14% in Q1, which is a good start. It’ll take some time for WHP to ramp up supply chain and get distribution. But it’s a long-term white space for the company.

WHP is almost purpose-built to attack that gap. WHP says its platform reaches 80+ countries and 200+ operating partnerships, with $8.5B+ of retail sales across its brands. So we’re here, the growth is here, let’s get excited. Only 7.6% of revenue comes from non-US sources.

Just tell the story simply— there’s a new sheriff in town, and growth is possible!

Cellebrite is tanking today. Deservedly so:

• Sudden, quick CEO change to a recently-hired product executive

• Revenue/ARR guidance taken down

• Sales lead times longer than anticipated, kicked out into the rest of the year

• Product launches (Inseyets and Genesis) seemingly not successful just yet in the face of AI competition

• Global pushback from customers for having ties to Israel

It’s a virtually unmitigated disaster for the growth of the business. Even just going from their high-teens growth profile to a low-to-mid-teens growth profile means the multiple should tank.

Now, all of this probably made a great story on the short side— 1.2M shares short earlier in the summer because something like 8M shares short heading into earnings today. However, outside of the current valuation (~12x-EV/FCF) there’s a few mitigants to the problems that one might have the ambition to fade in order:

• Previous CEO was a private equity hatchet man, and I get the vibe that the new guy might be deliberately kitchen-sinking the guidance and tempering expectations, particularly with respect to the lead-times commentary.

• Everyone f-ing knows that product sales to governmental entities have long lead times! I think this is just tempering expectations and prepping investors for the eventual sales process to civilian Federal branches of government as mentioned on this mornings earnings call

• The genuine red flagginess of having your customers be slow to adopt your AI products is probably the lion’s share of today’s sell-off. It’s almost a tautology— your growth engine, AI products, aren’t working, so your stock should re-rate from a growth multiple to a value multiple.

• If we hyperbolize what management said on the call, about certain agencies pushing back against their foreign ownership rules (they have a controlling shareholder in Japan and are Israeli-based), we might see that become a tailwind if their major holder frees up shares. The Israel-associated-pushback might just fade based on sentiment, if there’s any serious pushback there at all.

CLBT was part of what I flagged a few months ago as “quality SaaS”, or something moat-y that had a chance against vibe-coded junk. There’s a number of reasons this is still the case:

• Evidentiary chain-of-custody in criminal cases is extremely important for the prosecution and CLBT provides the backbone for evidence handling

• Every state has its own procedure for retrieving device information and CLBT tailors their solutions to local procedure

• The UFED device itself is somewhat of a hardware moat, or at least serves as a wall to the walled garden of CLBT solutions

If you think 11-12x EV/FCF is fair for at least a double-digit grower with some product insulation from AI, this thing seems fair at this price. Usually, it takes a slow, meandering, credulous shareholder base to gradually wind down a company’s multiple in a situation like this, but this shareholder base decided “we’re gonna do it all in one day”.

Today’s newsletter is a public service announcement. I am talking to no one in particular. Mostly, I am doing this as a reminder to myself. But recent price action in popular small-cap land is showing just how crowding and borrowing conviction works. You end up with someone yelling fire in the movie theatre. This generally takes the form of some pablum about how “the story changed” or “the risk/reward changed” from one of the original evangelists, informed almost strictly by the stock being down a lot.

It’s a little sad but entirely predictable. This is the game we’re in— even an academic would tell you that 60-vol instruments will have thermonuclear detonation days. Or even just “bleed off the big earnings pop” like we’re seeing in the broadcaster basket I mentioned a week ago. That’s really what I wanted to talk about today.

When I look at small caps, or any idea for that matter, things are best understood in the original Buffettology terms of a “general” and a “workout”. A general being something of quality that will grow and is temporarily undervalued e.g. American Express at 12x earnings in the ‘60s, and a workout being a classic special situation/activist liquidation/merger. Most of the time the cheap stuff we find will be a workout. That means a quick exit after a cheap, illiquid piece of paper gets catalyzed by X event occurring. Even if that event is just a subsidiary growing 200% y/o/y and being a cash cow, or something. The point is that it is not a general. Business durability and quality are not part of the equation, really.

I can’t keep falling in love with these shitcos! Statistically, most of this stuff is junk. And if the thing pops 25% the week after you buy it, well, chances are you should probably sell it. Again, I’m mostly writing this to myself. There’s a number of good reasons to do this:

• You probably didn’t identify the next MNST/compounding machine

• There’s probably an anguished shareholder base that has been waiting for a pop to sell

• The IRR you just made means that statistical mean-reversion is an absolute certainty to crush forward returns

• We are dealing with bad companies, and they generally find ways to squander momentum (stock offerings, buying a new business, over promising and underdelivering)

• A company just doesn’t go from bad→good because you bought the stock, you should still hold an adversarial view until the long-term chart looks like a beautiful asymptote.

  • The corollary to this is that like, a chartist will probably evaluate management more shrewdly than I would as a non-chartist. It’s a funny thing those charts.

The only real valid pushback: 1.) taxes, and 2.) idea velocity. If you’re constantly selling these pops you’ll make money, but you need to recycle capital. So it’s a tough business regardless. The motto sort of has to be “I can always buy this thing back, and its super volatile, so I’ll probably get another bite at the apple, no need to be greedy”.

INVESTMENT ANGLE
PHUN is trading below its net cash position is the headline; this is mechanistic margin of safety if management stops burning cash or returns capital. Which is exactly what the activist (Goldenwise Capital) is pushing for. Against that: the underlying operating business is minuscule (under $3M ARR), losing roughly ~mid-single digits millions of dollars every quarter. This is the third business line that the company has undergone since going public in 2021.

If the activists get there way it’s a quick cigar-butt victory. If the company has their way, the board will continue collecting exorbitant fees while the cash pile dwindles. The market sees it the latter way.

WHAT IT DOES
Phunware is an Austin, Texas-based enterprise software company that builds mobile app and location-intelligence tools for large physical properties, primarily hotels and other hospitality venues, with healthcare cited as an adjacent target market. Its platform, now branded as an AI-enabled "Guest Intelligence Platform," combines mobile app infrastructure, indoor location tracking, and generative AI to help operators engage guests in real time. Recent product launches include AI Concierge (commercially released January 2026), an AI Itinerary Builder, and a forthcoming Location-Aware Data Layer.

PHUN has run into trouble here with competition in just about every vertical, which is why the money gets burned in SG&A. In general app dev, clients prefer cheap, industry-specialized point solutions (Intelity for a hotel, NGP VAN for a political campaign) over a massive custom MaaS framework. Shifts in iOS/Android privacy policies hit third-party SDK-based location data and targeted mobile media hard, giving first-party app builders and larger ad-tech platforms a strong advantage.

The company is best known outside investing circles for building the Trump-Pence 2020 campaign's mobile app, which briefly made the stock a retail-trading favorite; that association has faded as a business driver in 2019 as evidenced by the long-term stock chart. It got memed, it’s now pivoted into this MaaS business.

HOW IT MAKES MONEY
Revenue comes from software subscriptions and services, application transactions, and some advertising-related activity, sold as one- to five-year license or service contracts. In practice this is a very small business: net revenue was just $0.8M in Q2 2026 and $1.3M for the first half of 2026. Gross margin is low for a SaaS business but moving up (69% in Q2 2026, up from 42% a year earlier) because the cost base is mostly software delivery, and the company touts customer retention above 95%. opex (sales, R&D, G&A) of $6.6M in the quarter dwarf revenue, producing an operating loss of $6M and an adjusted EBITDA loss of $5.5M.

FINANCIAL TRAJECTORY
Phunware has a long history of NOLs funded by roughly $100M+ raised since inception via stock offerings (including a 1-for-50 reverse split). The balance sheet holds $92.1M in cash and no debt, versus a market cap of roughly $43-46M. Cash burn is actually accelerating, and just annualizing this quarter means $12M of burn every six months to fund the AI/hospitality pivot; at that burn rate, the current cash balance funds 6-7 more years of operating losses. New CEO Dmitry Kroshka took over in May 2026 to lead the strategic transformation. I am not hopeful. Too much competition and too little domain and selling expertise, not enough differentiation.

RECENT CATALYSTS
In July 2026, activist investor Goldenwise Capital Group (a 6.6% holder) sent Phunware's board a public letter criticizing governance and capital allocation and nominating four director candidates, warning it may run a proxy contest at the 2026 annual meeting if rebuffed. Phunware publicly rejected the letter as lacking a credible plan. Management is also pushing a fresh investor-relations push (webinars, investor presentations) to explain the "2.0 Strategy."

I think the activists have a good play here. The cash is valuable, the operating losses are valuable, and the shareholder base is despondent. There’s an IRS stipulation that operating losses have to be used by similar owners of the business, meaning that ownership can’t move over a certain threshold based on number of 5% owners of the business. It’s interesting that Goldenwise is the only listed 5%+ owner I can see, so there might be no problem here with maintaining the operating losses.

I keep bandying about two ideas in my head that seemingly don’t conflict. The first can best be summed up by this interview that Shkreli did on TBPN:

"every single trader out there seems to make the same mistake over and over again. their position size is probably two to ten times more than it should be."

"most folks don't actually have an edge when they trade. but if they did, they're trading as if they had a 4x or 5x kelly edge."

"even with a 60/40 edge on every trade you make, you'll go bust if you overbet."

"and who has a 60/40 edge in the stock market? nobody has a 60/40 edge."

OK so I agree with a lot of this. People oversize bets for all sorts of reasons. Probably the biggest is just inability to objectively suss out what percent edge they actually have. But it’s important to cut through this a bit. Kelly calls for a 20%-of-bankroll bet on a 60/40 bet with a symmetric payout. That means: a weighted coin, that has a 60% probability of heads and 40% tails, but still pays out $1 on heads and $0 on tails. Practically speaking, we’re talking about bets here where the downside is -100%. Shkreli correctly points out that in most simulations the bankrolls go to zero.

He flips back between 55/45 and 60/40 in the video. But he’s leaving out an important point about payout structure Let’s say we’re betting on merger arb, the offer is signed at $10 all cash, the stock trades at $9. Your estimate of fair value in the event of a deal break is $8.50. So effectively you’re risking 50 cents to win $1. That doubles the size of your bet according to Kelly:

And even at a 50/50 probability, the payout structure calls for a 25% of bankroll bet. I agree with everything Shkreli says, but he ignores that by seeking asymmetry in payout structure- merger arb vs. a +1/-1 payout structure (think a binary option or FDA approval or something) that Kelly sizing becomes more useful, and larger bet sizes become rationalized based on downside caps. In my Leopold take I mention the big asymmetric bets from Munger and Soros:

You do need some kind of fear of God in terms of risk management when you manage OPM, and it seems it was absent here. Soros had 300% of his fund in one FX trade, but it was FX, and it was structured through derivatives, so the downside was like 10% of NAV. When Munger had 120% of his fund in one merger arb idea, it was the local government nationalizing a utility company. The counterparty was the government. And you ended up with equity in a utility if it failed for some insane reason. These trades are qualitatively VERY different than what Leopold was doing, I suspect a quant would find numbers to agree with me.

What I mean to say with that bold is that the math supports big position sizes when the downside is capped. And I would argue it’s easier to find situations where downside is capped vs. Shkreli’s world of biotech, where binary outcomes are the norm. I would call handicapping biotech/FDA drug trial outcomes, or even pricing any binary event as “Shkrelian edge”. That is, I think the probability of drug approval is X, and the market is .5X or something. And the stock becomes worthless if there’s a negative outcome.

Ten years ago, John Huber wrote “What Is Your Edge?”, discussing just how mispriced the largest companies in the world can get:

Huber refers to what I would call Shkrelian edge as “informational edge”. Which is probably another good name for it. Huber would describe his edge as a “time horizon edge”. Which is fine, but I don’t think it captures the entire essence of the truncated payout structure that Huber is actually entering into in his Bank of America case study. Quoting:

The banks highlighted are just two examples of mispriced stocks this year, but there are countless examples of widely followed large cap stocks that become undervalued from time to time. I think they often get mispriced because there is a general perception that the next year or so is going to be very difficult for the company, and there isn’t any real near-term catalyst that will drive the stock price higher. This creates an advantage for those who are willing to deal with short-term underperformance.

What often happens is the short-term underperformance doesn’t even occur. Sometimes the market “advances” gains—the stock appreciates an amount in 6 months that you thought would have taken 3 years.

It sounds impish to say, but we’re just talking about buying low and selling high. Companies trade up and down relative to their own historical forward multiples. Sometimes they’re rich based on good news, sometimes they’re cheap because of bad news. Bad news happens to good stocks. And there might not be a near-term catalyst, but fundamentally you can’t lose much money on the stock if earnings stay steady and the multiple is at or near a historical trough. Consensus, quality businesses have a way of working these things out. The floor on a Bank of America type large cap might not be as concrete as a standing M&A takeover offer, or a net-net price where the company can be efficiently liquidated, but nevertheless Huber is saying you won’t lose much relative to being long the S&P, and you stand to gain a lot if the company re-rates to where it was historically, or even just grows and employs capital efficiently [i.e. buybacks].

Continuing to use Huber’s example:

BAC’s book value is growing at around 6-8% annually, which means the bank will have somewhere around $20 of tangible book value in 3 years. At a modest return on tangible equity of 10% (which is what the company was already doing at all-time low profit margins), the company will have around $2 of earning power in 3 years. I think this would probably be worth 10-12 times earnings or so, meaning the stock would be worth somewhere between $20 and $24 in three years, and you could buy as much as you wanted for $12.

So you could say, (if you’re a long-only benching to the S&P 500) “on a 3-year horizon, in a Kelly format, I stand to make 18.5% of return over the next 3 years if my BAC thesis works. If I park it in the S&P I get 10%/year. And if BAC doesn’t re-rate I probably get a lousy 6% IRR.” So on a relative basis, I’m risking 400 bps of relative IRR if the stock doesn’t re-rate vs. upside of 850 bps of IRR if the stock does re-rate. That’s +3357 bps of alpha vs. -1400 on a 3 year horizon. Full Kelly is a ~5% size at a 30% win probability, which moves up to 10% at 33% probability. All Kelly is saying is that it’s worth doing something when your risk/reward is like this.

Of course this doesn’t take into account wild idiosyncratic events in the stock. I think it’s still a worthwhile underwriting exercise because there are two mitigating factors in the idiosyncratic underwriting— the first is that it’s a 3-year horizon, so even a big SEC fine or something won’t hit book value that much over 3 years. The same could be said about temporary crises at other blue chip companies— 3 years is enough time to work through a bad product launch or soft seasonality.

The second mitigating factor is that we’re relying on a quality screen; it’s a little like that Todd Combs “95% confidence interval that earnings will be higher in 5 years” or whatever he said. If you operate under that assumption, you can get to significant position sizes without huge amounts of Shkrelian edge. Even a 30% chance you’re right about it is worth a 5% position. That’s what I would call Huberian edge: I see what the future could be, I see that I won’t lose much in the way of relative return based on business quality, a trough multiple, and a long time horizon, and the trade is worth putting on".

I find myself agreeing with both Shkreli and Huber— the conclusion is always be on the hunt for asymmetry, both informationally and in payout structures. I suspect Shkreli would push back on this post in multiple ways: he might suggest that “every stock can always go to zero” and he would be right of course. And he’s right that informational edge is much firmer ground than ballparking peak/trough multiples on businesses, and giving confidence intervals to earnings power growth. Every business can always trade through its trough multiple, and things can go parabolic the other way too. He’d probably also say that you’d be fired before you get to see a 3-year horizon to fruition, and Huber is right that all your alpha could come in one 6-month period, either at the tail end of the 3-year period or at the start (good problem to have!).

Evaluating the usefulness of Kelly is important as well. Kelly is super sensitive to win probability of course, but the common theme everyone would agree with is asymmetry is good. You can use Kelly to back into implied win probabilities with position sizes as your starting point as a heuristic. Said another way, just ask Claude, “What win probability is implied by an X risk/reward ratio and a Y% position size”. Kelly is great for poker and binary outcomes. But it starts to fall off in usefulness when payout structures change. My advice to folks is just to focus on coming up with a risk/reward when forecasting, and think hard about conviction when sizing.

INVESTMENT ANGLE
CPS offers a rare pure-play way to bet on niche defense and electronics packaging materials without exposure to a large, diversified prime contractor, and its order backlog. Its unique tech and mostly easy sourcing give it optionality on defense budget growth and exposure to the data center world. That said, the business runs on thin-to-negative margins, revenue is lumpy and program-dependent. The spike you see on the chart is a speculative retail frenzy that juiced the stock in June. The company took advantage of this and issued $9.6M in new stock.

The company was seen as a data center/UAV-exposed micro-cap that could get juiced with the need for new drone tech and drone parts or simply participate in the data center bonanza. The stock sold back off after the issuance and reality sank in that they would have competition for these deals.

Why hasn’t it gone berserk with the data center bull market? There’s no real bottleneck. When a datacenter buys a UPS system or a high-density power module, CPSH only captures a tiny, sub-component level dollar value (the AlSiC baseplate or hermetic substrate). Because they are so far downstream, market participants view them as a general industrial materials supplier rather than a direct infrastructure bottleneck.

WHAT IT DOES
CPS Technologies designs, manufactures, and sells metal matrix composites (MMCs) — engineered materials that combine metals with ceramics to achieve properties neither material offers alone, such as high thermal conductivity paired with a tunable coefficient of thermal expansion, low weight, and high stiffness. Founded in 1984 and based in Norton, Massachusetts, the company has spent over three decades as one of the few independent commercial producers of these materials at scale. The first line of the MD&A says “The Company’s products contribute to the electrification of the green economy”.

Products include AlSiC (aluminum silicon carbide) components used as microprocessor lids, heat sinks, power substrates , and IGBT (Insulated Gate Bipolar Transistor) baseplates (used in hybrids/EVs/data centers instead of copper); hermetically sealed packages built to military and commercial specifications; and HybridTech Armor, which pairs ceramics with a metal matrix for lightweight ballistic protection.

There’s a number of competitors globally in all of these verticals. It’s probably a national strategic interest to make sure the company remains a domestic producers of these critical components (they have no competition in armor), even if our allies like Germany and Japan produce them.

HOW IT MAKES MONEY
Revenue comes from selling finished or semi-finished MMC components to industrial and government customers across electronics, power generation, automotive, telecommunications, aerospace, and defense. Denka is their big Japanese competitor. Historical customers have included large electronics and semiconductor names such as TI, HP, Agilent, and Amkor. A meaningful share of revenue is defense- and government-linked: recent quarters included a Navy SBIR contract extension tied to Amphibious Combat Vehicle armor work and a $4 million hermetic packaging order. The business is order- and program-driven rather than subscription-based, so quarterly revenue can swing with contract timing and defense procurement cycles.

It’s a job shop. They get custom-ish orders, they fill them

FINANCIAL TRAJECTORY
Despite the defense headlines, CPS’s revenue splits 80/20 commercial/defense application. Second-quarter 2026 revenue was $8.3 million, up slightly from $8.1 million a year earlier. First-quarter 2026 revenue had declined to $7.0 million on order timing, missing estimates.

This is a really low margin business historically, which makes sense since they compete with the industrial world with a US manufacturing cost base. Seems like they need to keep developing product and hopefully win more budget from DOD.

Current areas of active development include injection molded alloys, infiltration of composites, and HybridTech Armor: These products have been successfully installed on U.S. Navy aircraft carriers. HybridTech is easily modified by altering ceramic fiber reinforcements, ceramic tile composition, and overall panel thickness. CPS is actively engaged in developing alternate HybridTech designs to meet demand across platforms that demand both extremely low density armor solutions (aircraft) and heavy, high-threat situations (land vehicles).

RECENT CATALYSTS
The company participated in US government small business innovation research (SBIR) and small business technology transfer (STTR) programs. The typical SBIR/STTR funding structure is as follows:

Phase I: Feasibility Study. Initial government funding to evaluate the technical merit, feasibility, and commercial potential of the solution proposed in response to a published need.

Phase II: Prototype Development. Funding for the development of prototypes, engineering models, and testing based to expand upon Phase I results. Enables the small business to refine products and prepare them for commercialization.

Phase III: Commercialization. No direct SBIR/STTR funding. The small business uses commercial or non-SBIR government funding to bring products to market. Successful Phase I/II projects often attract commercial interest or government contracts. From the 10-K:

An important element of the SBIR/STTR program is the “Phase III mandate.” Federal agencies are required to grant non-competitive, follow-on contracts to small businesses for technology developed under Phase I and Phase II contracts, provided the work derives from or extends that earlier work. This provides a mechanism to commercialize technology without additional competition, enabling rapid procurement to satisfy government needs. To satisfy this mandate, federal agencies should, "to the greatest extent practicable," award Phase III contracts to the original Phase I/II awardee, bypassing the need for a separate, formal competition.

This is R&D funding for the company from taxpayers, which helps product development and eventually gets to a “guaranteed” contract. Successful product development means more contracts —> more revenue. With their historically unprofitable management, I’m not even sure that more revenue means more profits. But it’s good for national security that this company exists, it’s just a matter of how much the US government lets it earn.

RYAM reported 2Q results last night. They were good— revenue was up low double-digits, even accounting for average selling price compression of -10%, squaring to the reported +22% in volume. Unfortunately, the business is still clogged up by fresh US capacity in PBD and oversupply of HYP in Asia. Management also noted just overall weaker demand for paper and packaging materials. Cellulose specialty products pricing showed +21% while bulk product pricing collapsed.

The earnings release we’ll review to intuit whether this review actually results in a transaction:

“I am excited to join RYAM at a pivotal time for the Company,” said Daniel M. Krawczyk, President and Chief Executive Officer. “The Board asked me to bring an operational lens, strategic perspective and transaction experience to a clear mandate: maximize value for our shareholders. I am fully committed to concluding the Company’s comprehensive review of strategic alternatives with urgency and discipline and communicating a clear path forward during the fourth quarter. At the same time, we will continue strengthening the performance and market position of the business while keeping the Company on track to deliver against its full-year objectives.

“Since stepping into the role, I have spent significant time with our employees, customers, leadership team and advisors, as well as interested parties participating in our strategic review and other stakeholders. Those conversations have reinforced my conviction that RYAM has substantial untapped value. At the center of that value is a highly differentiated Cellulose Specialties franchise, supported by leading market positions, specialized and difficult-to-replicate assets, deep technical expertise and strong customer relationships built over decades. These durable competitive advantages, together with a tangible pipeline of reliability, productivity and cost initiatives across our manufacturing network, provide clear, actionable levers to unlock value and deliver stronger, more consistent earnings and cash flow.

“Our second-quarter results were ahead of expectations and showed strong sequential improvement, reflecting both the value of our Cellulose Specialties products and the benefits of ongoing operational improvements. Our full-year trajectory remains aligned with our prior expectations, including generating positive free cash flow in 2026. Cellulose Specialties pricing increased 21% compared with the prior-year quarter, reflecting the value our products deliver in our customers’ most demanding applications. Our customers are critical partners to RYAM, and we will continue working collaboratively with them to help differentiate their products and create value in the markets they serve. We also remain focused on improving reliability and prioritizing the products and markets where RYAM has the strongest competitive position.

“To be clear, executing the business plan and completing the strategic review are not competing priorities. Stronger operations, disciplined commercial execution and improved cash generation enhance the value of the Company and the range of alternatives available to the Board. We also continue to advance select biomaterials opportunities that offer attractive investment economics and meaningful value creation. For example, RYAM retains a capital-light interest in the Altamaha Green Energy project through its land and prior investments, preserving potential upside without requiring additional cash equity from the Company. My focus is to build on RYAM’s Cellulose Specialties leadership position, partner closely with our customers and act decisively on the opportunities in front of us, while ensuring that the path ultimately selected captures the full value of this unique platform for our shareholders.”

“The range of alternatives available to the Board”— hopefully this means a price range!

As if responding directly to my previous post on RYAM, CEO Krawczyk noted on the call this morning that “[his] appointment does not alter, delay, or narrow the [strategic alternatives] process”. He also said the review would be complete in Q4 of this year. So we’ll get some color on that, hopefully. Also the call had some talk of becoming a US government supplier of nitrocellulose, a specialty product used in weapons systems— an interesting growth avenue for that higher margin business. The most info I can get on this is that DoD awarded a $136.7M contract to MilliporeSigma in 2021 specifically to establish domestic nitrocellulose membrane production capacity (primarily for medical diagnostic tests). It’s used in small arms manufacturing so there might be more growth there in the wake of the reconciliation bill.

Later on in the call, Krawcyzk also noted his experience and made the pitch that he was the guy to do the turnaround. So I remain unconvinced of an imminent transaction. It might be a negotiation tactic where the company has to posture that it will remain independent, pursue growth avenues and cost initiatives and be super profitable and awesome. It might also be Krawcyzk doing a pitch for himself to stay on after the business is sold— why do the dirty work for PE? Why fire a bunch of people if you can just sell out to a strategic and have them do it all? To get a better price for the whole business, hopefully. But it does seem like there’s low-hanging fruit to be cut (board expenses!).

I imagine the positive reaction to the stock’s movement today is in response to the improvement in underlying product lines and the pricing pickup in specialty products. There’s revenue opportunity, there’s cost opportunity, and margin improvement is the name of the game for these commodity businesses. It might also be pod-shop event-driven guys seeing the Q4 deadline mentioned for strategic review getting involved now that the catalyst is imminent. Probably truncates a distribution of outcomes, and allows the options market to be tapped for some structuring with puts and calls around key prices ($12 top from AIPCF mentioned in part 1). No one’s falling in love with this stock anytime soon- bad business, clearly more valuable private/merged into another, bad balance sheet. But the risk/reward might be appetizing to some heading toward the conclusion of the review.

There are funny stories that get told and retold about legendary trades and investors. It’s worth re-stating that, as this newsletter mentions frequently, circumstances always define everything— everything is always n=1, this time is always different, pattern recognition and analogies are never one-to-one, and one is always working off imperfect information. That being said, it is worth looking at what the greats have done in markets, relatively recently, particularly with equities. Druckenmiller has that famous quote on equities. Bastardized, it’s something like 90% is sector/factor/beta, 10% of the move is idiosyncratic.

In a fairly recent interview (maybe 2024?) he once talked about how he just asked an LLM “for the most liquid ADRs in Argentina” when he liked the new policymaking regime in Argentina. Then he just bought the basket. When Buffett bought his basket of airlines pre-COVID, he bought every single US carrier, without favoring any single one of them, but consciously staying under 10% as to not trigger the short swing profit rule. This is the key feature it seems of the basket advantage that is sort of tautological— of course there is more liquidity. Liquidity is helpful on both entry and exit. Liquidity attracts MORE liquidity. Idea velocity in liquid instruments accelerates as a function of price. That is a smart-sounding way of saying “the more liquid your trade, the quicker the market decides if you’re right or wrong”. It’s probably the key selling point of putting on a basket trade. Look at Leopold’s book— it was essentially two home-run basket trades, just levered a bit too much. That whole episode shows just how powerful on the upside and downside you can get with basket trades: hugely idiosyncratic baskets can trade -30% in calm markets. Said another way— you get the juice when trading these baskets while maintaining liquidity. Maybe not as much as single names, but you get the point.

**Another tautology— a pop-culture-acronym basket, something like FANG, doesn’t necessarily make a basket, but it can after enough positive reinforcement and reflexivity. The original Facebook, Amazon, Netflix, and Google had very different business models. Even Facebook/Google, which could be classed as “digital ads” could be bifurcated as “social media vs. search”. So business model correlation can be overridden by market sentiment at times.

The post is timely because I like a few basket trades, but I hate pretending like I’m an expert in any of these individual businesses. I don’t know which TV station management will do the best in the political cycle. I don’t know which American coal companies have the best cost curve. I don’t exactly know which former blue-chip software businesses will actually be fine in the world of AI. But I know enough to know they’re hated and being bought together shields me from too much in the way of idiosyncratic bombshells. This is essentially how I group my watchlist— European defense, coal/PGMs, busted consumer names, busted SAAS names, TV stations, non-MAG7 MAG7 (things like Tencent).

The downsides of the basket vs. the single name are sort of obvious. You’re never gonna capture the maximum upside long-term. One-hundred baggers generally don’t come in basket form. You also eschew any activism where you’re going to be able to effect change at the company management level. Your positions are naturally smaller so you don’t get the Bezos-style “1,000,000 to 1 payout” for a given company’s product launch or special circumstances. At sort of an ontological level, the basket is antithetical to any business that seeks to be differentiated. You’re saying “actually the product doesn’t matter, you’re a slave to the industry dynamics regardless”. Which is spiritually misaligned with alpha-stockpicking. But nonetheless, lately, I find myself more attracted to these basket-type trades. At the end of the day it’s just about being more ambivalent about how you make your money, rather than focusing 100% of time on finding 100-baggers.

This quest, for the alpha-stockpicker, is rooted in a belief in human agency—that a great CEO matters, that a great product matters. The basket trade is an admission that, most of the time, the tide matters more than the boat. It is an acknowledgment that we are operating in complex systems where exogenous factors (liquidity, policy, reflexivity) regularly overwhelm endogenous drivers (earnings, product-market fit, management). Commodities guys like Goerhing and Rozencjwag get this. By treating it purely as a proxy for an industry or factor or commodity price, we are not admitting defeat. Rather, we are optimizing for the variables where we actually have an edge. We may not have an edge in predicting which busted SAAS company will survive, but we may have an edge in recognizing a group of them, trading at a historic valuation disconnect relative to forward growth, deserve the benefit of the doubt and may benefit from an “AI helps, not hurts” narrative shift.

Ultimately, the goal of investing is not to be ideologically pure, the goal is to make money. A portfolio comprised entirely of single-name, high-conviction moonshots can be a recipe for catastrophic idiosyncratic risk, while indexing is admitting defeat (not to mention the current state index constituents). The basket trade sits in that sweet spot.

INVESTMENT ANGLE

Bridgford is currently a story of margin compression. Revenue is roughly flat, losses are widening, and the dividend cut killed the stock earlier this year. The offsetting factor is the owned real estate across four states, which sits on the balance sheet at depreciated cost. If margins don’t recover organically, this could be the death knell for the operating business, possibly unlocking the value of the real estate. This is a classic “value without a catalyst”- you need a controlled company to fold up shop and sell off real estate assets. Probably a longshot, but worth keeping on the watchlist.

WHAT IT DOES

Bridgford Foods is a controlled, publicly traded food manufacturer founded in 1932, now headquartered in Dallas, Texas, with production facilities also in Anaheim (California), Chicago (Illinois), and Statesville (North Carolina). The company operates two segments: Frozen Food Products (frozen bread dough, biscuits, rolls, and fruit turnovers sold mainly to foodservice and retail channels) and Snack Food Products (beef jerky, dry sausage, meat snacks, and ready-to-eat sandwiches sold to supermarkets, mass merchandisers, and convenience stores). The snack segment generates the majority of revenue.

HOW IT MAKES MONEY

Frozen products move through wholesalers, cooperatives, and foodservice distributors. Snack products are sold both through customer-owned distribution centers and a direct-store-delivery (DSD) network that stocks convenience stores and supermarkets directly. Beef and other meat inputs are the largest cost driver for the snack segment, making margins sensitive to commodity price swings. Bridgford owns its manufacturing and distribution real estate outright rather than leasing it, a structural feature that sets it apart from most small-cap packaged-food peers.

Walmart is 33% of revenue and 9% of receivables. Not a great spot to be in.

FINANCIAL TRAJECTORY

FY2025 revenue rose 3.3% (this was after price pass-through) to $231.0 million, but the net loss widened to $13.4 million from a much smaller loss the year before. The trend continued into FY2026: second-quarter revenue was roughly flat at $50 million, with a net loss of $5.8 million versus $5.0 million a year earlier. Gross margin compressed to 17.5% from 21.9% as beef and overhead rose faster than the price increases the company has been pushing through.

Frozen foods as a whole have taken a hit in the GLP-1 wake. It might be a cyclical thing, but it’s hard to see a turnaround on the horizon. Particular with labor/pension/delivery costs being borne out by the company.

RECENT CATALYSTS

The risk/reward is totally contingent on the value of the company’s real estate. The headquarters in Anaheim, the multi-property manufacturing hubs in Dallas and Chicago, and the Statesville plant are the lion’s share of real estate value. The prime locations in aggregate could be worth as much as $200M compared to a current EV of $65M. BRID sold one facility at 170 N. Green Street in the Fulton Market district of Chicago in June 2022 for $60M to a developer. So it’s truly prime real estate:

BRID suspended its quarterly dividend on February 5, 2026, a rare move for a company that had paid one for decades, signaling a priority on cash preservation. Management has responded to margin pressure with repeated price increases, a shift toward private-label production, and a mix shift toward higher-margin snack items, though these have only partly offset cost inflation so far. This is the bear case— it’s clearly the owner/operators want to keep running the thing. And any monetization of real estate might kill the whole thing, unless it was arranged as a sale/leaseback. But even then, with current margins, the real estate assets are clearly not being put to their highest and best use.