Everybody's Supplier. Nobody Knows.

There’s a company in this market that makes the pressure vessel holding the propellant, the composite tube holding the payload, the wing holding the weapon, and the bored steel body of the weapon itself. It sells those parts to nearly every prime contractor and nearly every new-space name you can list. By design it doesn’t compete with a single one of its customers. It has never once been the story.

It went public a couple of months ago and raised a large primary round. No insider sold a share into the deal. It used essentially all the money to retire debt, taking a heavily levered private balance sheet down to something conservative in one afternoon. Revenue grew sharply last quarter. Adjusted EBITDA grew faster. Backlog hit a record.

Then it reported, and the stock fell hard enough to close below its offering price.

The reason is the most boring thing in finance. When the offering closed, every unvested management incentive unit vested at once, and the accounting rules made the company expense all of it in the quarter. The charge was a nine-figure number. It moved no cash. It changed no contract. It can’t recur, because the IPO itself was the trigger. But it turned a normal quarter into a headline loss of more than a dollar a share against a consensus of roughly breakeven, and the machines did what machines do.

The tape sold an accounting event.

Underneath it, the overwhelming majority of this company’s revenue sits on sole-source or single-source awards, its average customer relationship runs multiple decades, and management believes a large fraction of its already-qualified American production capacity is sitting available right now. All of that in a sector where a July executive order just put prime contractors on a clock to qualify alternative domestic suppliers.

The chief executive told analysts he’s more optimistic about what’s in front of the company than he was three months ago. Nobody wrote it down.

Then there’s this. Four insiders, including the CFO, wrote personal checks for stock at the offering price, and not one insider has sold a share since. The stock now trades below where they bought. And the entire published research coverage on this company comes from the banks that sold you the deal.

That’s the setup. The rest of this is the name, the ticker, the financing stack, the balance sheet details the data providers are getting wrong by a billion dollars, the insider buys, the management dossier, why this beats the defense stock everyone actually owns, the valuation work, and the eight things that would get me out.

No paywall on this one. Read all of it.

If you want the live version, join the Guardian Research chat. The model portfolio is up 82.29% year to date through August 14, and every entry, every trim and every add goes out in the chat the second I do it, not in a Sunday recap after the move already happened. You also get to argue your own ideas with a room of people who do this for a living. This position is in there.

→ Join the chat.

Let me describe the shape of this before I name it, because the shape is the thesis.

In 1995 six companies made solid rocket motors in the United States. Today two do. For fiscal 2026 the Pentagon asked for $24.4 billion of missile procurement. For fiscal 2027 it asked for $70.5 billion, a 188% increase in one budget cycle. PAC-3 MSE goes from 357 units to more than 3,200. THAAD goes from 55 to 857. Tomahawk goes from 55 to 785.

Everyone understands that part. It’s priced. What isn’t priced is the second-order fact, which is that money doesn’t make rocket motors. Motor cases, nozzles and insulation make rocket motors, and qualified physical capacity constrains those, along with the requalification process. Nozzles run seven to ten months of lead time. Ammonium perchlorate has exactly one domestic producer. L3Harris spent $30 million to drive a 1,000% capacity increase at a single motor-case supplier, which tells you how binding the constraint is and how violently primes will pay to relieve it. Breaking Defense laid out the whole crunch in January and it’s worth your time.

Tom Karako at CSIS put the FY27 volumes plainly. They “simply cannot be achieved in a single year.” An analyst quoted in the same reporting was blunter: “solid rocket motors and seekers continue to be really significant bottlenecks, and I’m not sure that there’s a dollar figure that can overcome those bottlenecks.”

Space is the second leg. Falcon 9 flew its 90th mission of 2026 on August 4, an annualized pace near 152 flights, and every one of those expends a second stage. Amazon’s Leo constellation has to put roughly 2,900 more satellites in orbit before its July 30, 2029 license condition. The Space Development Agency has another 470 or so satellites funded and in build. Every one of those spacecraft needs a structure, a honeycomb panel, a deployable, a solar array, and in the maneuvering ones, a tank.

Then in July 2026 the White House signed Executive Order 14415. It gives the Secretary 180 days to identify national-security acquisitions that depend on unreliable suppliers, then requires prime contractors to qualify alternative domestic sources, with contract termination available as the enforcement mechanism. From January 1, 2027, waivers are largely prohibited.

Read that again. The federal government just made second-sourcing a legal obligation with a deadline attached.

Which leaves one question worth answering. Who owns qualified, certified, available American production capacity in these exact product categories, and what’s the market paying for it?

Applied Aerospace & Defense, Inc. trades on the NYSE as AADX. Huntsville, Alabama. 1,542 employees. Eleven facilities across six states. It closed at $18.11 on the last quote I have before publication, and the last exchange close I can confirm is $17.92 on August 14. That’s a $3.12 billion market capitalization and an enterprise value near $3.51 billion once you fix the net debt figure half the data providers still have wrong.

Applied makes the physical hardware everything else bolts to. Three end markets.

Space and Launch Systems covers spin-formed propellant tanks and domes, engine nozzles, nose cones, payload fairings, payload adapters, composite tubes, satellite bus structures, solar arrays, deployables and deorbit systems. It has content on SpaceX Falcon 9 and ramping content on Blue Origin’s New Glenn. The James Webb Space Telescope sunshield came out of the Huntsville operation. So did the largest solar sail ever built, roughly 17,792 square feet deployed, four basketball courts, packed down to the size of a microwave, announced for NOAA in May 2026.

Defense Aviation and Airborne Systems covers wings, control surfaces, fuselage structures, landing gear and rotorcraft drive systems. It has content on Anduril’s Fury collaborative combat aircraft and on Bell’s MV-75 tiltrotor, plus a large installed base generating aftermarket and sustainment work.

C5ISR and Precision Strike Systems covers RF-transparent composite radomes and radar antennas, missile bodies, integrated air and missile defense hardware, and solid rocket motor cases. That segment grew 262% year over year last quarter, from $13.7mm to $49.6mm. Space and Launch grew 59%, from $24.5mm to $38.8mm. Defense Aviation grew 5%, from $75.3mm to $78.9mm.

The prospectus language is the whole business in three lines. Roughly 87% of revenue and 86% of pro forma revenue comes from sole-source or single-source awards. Roughly 89% of revenue and 88% of pro forma revenue is tied to IP-enabled production processes. The average customer relationship spans 39 years.

Thirty-nine years. That’s a qualification moat with a date stamp on it. In flight hardware you don’t switch vendors because someone quotes 4% cheaper. You switch because the incumbent failed, and then you spend two years and a lot of money requalifying. Applied holds sole- or single-source position on 87% of what it sells, which is a customer sourcing decision as much as a technical one, though it works the same way in practice for as long as it holds.

CEO Trip Ferguson framed it on the Q2 call: “We’re prime agnostic. We serve nearly all the leaders and innovators in defense and space, and we don’t compete with our customers by design.”

Customers named in company press releases and on that call include Lockheed, Northrop, L3Harris, Boeing, RTX, BAE, GE Aerospace, Bell, Sikorsky, Sierra Space, SpaceX, Blue Origin, Anduril, NASA and NOAA. Worth knowing that the prospectus itself names no customers, referring only to “blue-chip aerospace and defense prime contractors.” Every name above comes from press releases, trade coverage or the earnings call.

Rating: long.

Price: $18.11 on the last quote before publication. Last confirmed close $17.92 on August 14.

Guardian Research price target: $26.00. That’s 26x FY2027E adjusted EBITDA of $190mm, cross-checked at 22x FY2028E of $216mm. Both routes land at $26.

Bull case: $35.00, at 24x FY2028E adjusted EBITDA of $268mm on a 26.5% margin.

Blue sky: $55.00, at 30x FY2028E adjusted EBITDA of $328mm. That’s Karman’s margin structure on Applied’s capacity.

Bear case: $9.25. The second-half guide misses, margins don’t recover, the multiple compresses to the aerostructures median.

Horizon: 12 to 24 months. Two earnings prints, one lock-up expiry, one budget cycle.

What breaks it: fundamentals, not price. The full list is at the end.

This isn’t a max-size position, and the lock-up is the reason. There’s a dated, bounded supply event in front of this stock in November, and being maximum-size into something you can’t control and can’t vote on is how a good thesis turns into a bad outcome.

I don’t use stop-losses and I’m not publishing one here, and how I size and exit is the same on every name I own. A price is not a thesis. If this trades to $15 because a sponsor markets a secondary into a thin float, nothing about the business has changed and I’m a buyer. If it trades to $22 because momentum funds found it while the margin recovery quietly failed, I’m a seller at a profit. The tape isn’t evidence. The prints and the filings are evidence, and I’ve listed which ones I’m watching and what they’d have to say to get me out.

So lower prices with the thesis intact are an add. Size it so a 30% drawdown on a supply event is an opportunity rather than a problem. That’s a sizing decision you make before the drawdown, not a reaction to one.

Start here, because if you don’t understand it you can’t own this.

On August 12 Applied reported second-quarter revenue of $167.3mm, up 47.4% year over year. Consensus sat under $156mm, so that’s a beat of roughly 7%. The company also issued its first guidance as a public company: FY2026 revenue of $670 to $690mm and adjusted EBITDA of $150 to $155mm, against a stale consensus of $676mm. Technically a raise.

Then it reported GAAP earnings per share of negative $1.04 against a consensus of negative $0.01.

Here’s the whole bridge, straight from the release:

• Q2 GAAP net loss: $(154.0)mm

• Add back income tax expense: $31.5mm

• Add back interest expense, net: $26.2mm

• Add back depreciation and amortization: $15.1mm

• Add back IPO share-based compensation: $110.1mm

• Add back transaction costs: $5.2mm

• Add back integration and restructuring: $2.0mm

• Add back management fees: $0.2mm

• Q2 adjusted EBITDA: $36.4mm, up 38.5% year over year

Share-based compensation in the first half of 2025 was $1.6mm. In the first half of 2026 it was $110.8mm expensed, out of $113.2mm of total cost. A 69x increase, and the 10-Q explains it in one sentence: “Upon the completion of the IPO, all outstanding incentive units vested, resulting in total share-based compensation cost of $113,183 [thousand].”

Pre-IPO profits interests. They vested on the offering. The company had to expense them. Divide $110.1mm by the 148.2mm weighted average shares and you get $0.74 a share. Add it back and the quarter goes from negative $1.04 to roughly negative $0.30. That’s inside a rounding error of the negative $0.25 adjusted figure that got reported in some places.

It gets better one layer down. Management disclosed that roughly six percentage points of the 22.2% Q2 gross margin was IPO share-based comp allocated into cost of sales. Strip that and the underlying gross margin sits near 28%, consistent with FY2025’s 27.9%. The margin collapse BofA cited when it trimmed its target from $24 to $23 is, in substantial part, the same accounting event showing up in a second line item.

Adjusted EBITDA grew 38.5% year over year at a 21.8% margin. The stock fell 17%.

One caveat I won’t paper over. The tax line is real cash. Because the $110mm of stock compensation is largely non-deductible, Applied booked $31.5mm of income tax expense on a $122.5mm pre-tax loss, and that produced a $36.5mm accrued income tax liability that didn’t exist at year-end. The charge is non-cash. Its tax consequence isn’t.

First, the lineage.

This isn’t an old company with a new ticker. It’s two old companies inside a four-year-old private equity vehicle, and you need the lineage to underwrite the balance sheet.

PCX Aerosystems traces to 1900, headquartered for most of its life in Newington, Connecticut. Rotor head assemblies, power transmission gearing, landing gear, engine and airframe hardware. Customers included Boeing, Bell, GE and Sikorsky. Greenbriar Equity Group acquired it in April 2021, concurrently carving Senior Aerospace Connecticut out of Senior plc. PCX then bought Timken Aerospace Drive Systems in 2022, which is where the rotorcraft drive franchise comes from.

Share

Applied Aerospace Structures Corporation was founded in Stockton, California in 1954. A 25-acre campus, roughly 330,000 square feet, building composite and metallic airframe and spacecraft structures. Greenbriar acquired AASC on December 8, 2022. Kevin Bidlack, who ran AASC then, is Applied’s Chief Operating Officer today.

Then the buying started.

• October 2024: Innovative Composite Engineering, White Salmon, Washington. Composite tube manufacturing.

• March 2025: NeXolve Holdings, Huntsville, Alabama. Polyimide films and deployables. Built the James Webb sunshield and won a NASA Collier Award. Became the Advanced Materials and Deployable Systems division. $20.7mm including a $5.0mm earnout.

• November 14, 2025: AASC and PCX merged. AA&D Holdings, LP merged with Rotor Topco, LP. Announced December 3. Over 120 years of combined expertise.

• January 2026: Vestigo Aerospace and its Spinnaker drag sails, for deorbit compliance. $540K. Founder Dr. David Spencer, formerly of NASA JPL, joined as VP of Deployable Systems.

• March 2, 2026: Consolidated Boring, Cincinnati, Ohio. Deep hole boring for missile bodies, gun tubes and motor cases. $374.8mm.

• March 2, 2026: Ultracor, Billerica, Massachusetts. Honeycomb core in carbon, quartz, Kevlar, fiberglass and proprietary Ultraflex. $7.2mm. Vertical integration into its own composite inputs.

Headquarters moved to 355 Quality Circle NW in Huntsville, which arrived with NeXolve and happens to be the address of the American missile-defense industrial base.

This is a Greenbriar buy-and-build that reached its exit window. That’s a description, not a criticism, and it dictates three things you have to underwrite: the goodwill, the leverage and the sponsor’s ownership.

Now the debt. Nobody reads the debt footnote. Read the debt footnote.

Nobody reads the debt footnote. Read the debt footnote.

The original credit agreement dated December 1, 2022 gave them a $130.0mm term loan and a $20.0mm revolver, maturing October 1, 2028. Amendment No. 1 on October 1, 2024 took term loans up to $250.0mm, the revolver to $40.0mm, added $100.0mm of delayed-draw commitments, and pushed maturity to December 1, 2030. Amendment No. 2 on November 14, 2025 took term loans to $645.0mm, delayed-draw to $150.0mm and the revolver to $100.0mm, on the same day the two platforms merged. Amendment No. 3 on March 2, 2026 added $180.0mm of incremental term loans, drew the full $150.0mm delayed-draw and took the revolver to $125.0mm, with $31.1mm drawn immediately and $25.0mm later.

Read that sequence with the dates attached. In November 2025 they merged the platforms and upsized the term loan to $645mm. In March 2026 they levered up another $330mm to buy Consolidated Boring. In May they filed the S-1. In June they IPO’d and paid it all back down. That’s the classic sponsor sequence. You can dislike it. You should still price it correctly, because what the public bought is a de-levered version of an asset that got assembled cheaply.

The terms, as disclosed:

Pricing runs at “term SOFR plus an additional spread based on the Company’s total net leverage ratio, which was initially 6.25%.” It’s a leverage-based grid, and the initial 625bp spread was struck when the company was far more levered than it is now. At 2.7x it should step down meaningfully. The grid tiers aren’t disclosed, which makes this a real, unpriced call option on interest expense.

Amortization runs 0.25% of the aggregate initial term loan principal, quarterly. Trivially small.

Prepayment is optional at any time subject to a penalty up to 1.00% of principal repaid. A soft call, so it’s cheap to refinance.

Maturity is December 1, 2030. Four and a half years out. No wall.

On covenants, the 10-Q says only that the company “was in compliance with all applicable covenants.” The numeric maximum net leverage test, the fixed charge coverage minimum, the step-down schedule and any excess cash flow sweep sit inside the credit agreement exhibit, which I couldn’t retrieve. I’d rather tell you that than guess. At 2.7x against an agreement written for a far more levered company, headroom isn’t the risk.

No interest rate swaps or derivatives appear anywhere in the fair value footnote, so the term loan looks fully floating and unhedged. That’s exposure in a rising-rate scenario and a benefit in a cutting one.

What’s absent from the stack surprised me, and it’s rare for a sponsor-backed IPO. No convertible notes. No preferred stock, with 50,000,000 shares authorized and zero issued. No warrants. No seller notes, because the Consolidated Boring seller took $70.0mm of equity units instead. One earnout in total, a $5.0mm Level 3 contingent consideration liability from NeXolve, unchanged at both June 30, 2026 and December 31, 2025.

Compare that to the average de-SPAC or sponsor exit and it’s remarkably clean. No hidden claim sits ahead of the common. What you see in the share count is what there is.

Where it stands today: total term loan principal of $405.8mm, down from $643.4mm at December 31, 2025. Long-term debt net of $395.2mm, down from $627.0mm. Finance leases of $31.0mm. Cash of $18.1mm. Net debt on management’s basis of $387.7mm, or $418.7mm including finance leases. The $125.0mm revolver is fully undrawn.

They repaid $626.2mm of principal, revolver and accrued interest in one motion. Debt-to-capitalization fell from 0.80 to 0.33. Net leverage went from an estimated 8.8x to a management-stated 2.7x. Treat the 8.8x carefully, because it’s a third-party estimate and the company has never disclosed a pre-IPO leverage number. One independent estimate puts the post-IPO figure at 3.10x rather than 2.7x, a definitional difference over the EBITDA base. Total liquidity today runs about $143mm.

Here’s what the screeners are getting wrong. As of this writing, stockanalysis.com still shows AADX with $1.04bn of total debt and a $4.54bn enterprise value, producing an EV/EBITDA of 45.9x. Those are first-quarter numbers, from before the offering. The correct enterprise value is $3.51bn and the correct forward multiple is 23x. Screen this company on an automated data feed today and you’re looking at a business that no longer exists.

And now the balance sheet, where six things matter and none of them show up on a screener.

Six things matter here and none of them show up on a screener.

Tangible book value is negative. Total shareholders’ equity is $827.9mm, or $4.81 a share, which puts the stock at 3.8x book. But goodwill is $581.4mm and net intangibles are $353.3mm. That’s $934.7mm of purchase-accounting assets against $827.9mm of equity, so tangible book value runs approximately negative $107mm, or negative $0.62 a share. That’s what a roll-up looks like. It means no asset-value floor sits under this stock. The floor is the earnings power of qualified capacity. Anyone underwriting this as cheap-on-book is underwriting the wrong thing.

They paid up for Consolidated Boring, and the purchase accounting is preliminary. CBI cost $374.77mm of total consideration, including $70.0mm of fair-valued equity units issued to the seller, with 8,564,729 shares issued to AA&D Holdings in the process. The March 2026 financing funded CBI, Ultracor and fees together, so don’t read the financing components as a clean bridge to the purchase price. The allocation puts identifiable net assets at $139.7mm, intangibles at $168.1mm, and goodwill at $235.1mm that isn’t tax deductible. CBI’s FY2025 was $105.6mm of revenue and $13.3mm of EBIT, so Applied paid roughly 3.5x trailing sales and, depending on where you put D&A, near 20x trailing EBITDA for a business earning materially below its own corporate margin. That’s a full price. The defense is in the run-rate, because CBI contributed $43.2mm of revenue in four months, which annualizes near $130mm, about 23% above its own prior year. On that basis the price is closer to 2.9x forward sales. Deep hole boring is exactly the capability the missile ramp needs. But this wasn’t a bargain, the allocation is still inside the measurement period, and a downward revision to the intangible split would flow through amortization.

Contract assets are 2.8x accounts receivable, and that’s the working capital story. Accounts receivable came in at $69.6mm, down from $71.4mm. Contract assets, unbilled, came in at $197.7mm, up from $140.8mm. Unbilled receivables rose $56.9mm in six months, of which $44.2mm was a straight operating cash use and the rest came in with CBI. Applied recognizes 82.7% of revenue over time, which means it does the work, books the revenue and waits to bill. Normal for long-cycle aerospace, and also exactly how a growing manufacturer starves itself of cash. First-half free cash flow was negative $103.7mm, which is operating cash flow of negative $82.1mm less $21.6mm of capex. Cash interest paid in the half was $40.8mm.

CFO Jeff McRae’s answer: “We expect much of that working capital to be converted back to cash as second half deliveries occur,” and “We project generating positive free cash flow in the second half of the year.” They framed roughly $36mm of working capital build as reversing. Watch that in the Q3 print above everything else. It requires a nine-figure turn in one half. It’s achievable, because the IPO and transaction costs are gone, interest expense runs at less than half the prior rate, and deliveries are back-end weighted. It isn’t automatic, and $18.1mm of cash is a thin cushion to be wrong on.

A $36.5mm cash tax bill now sits in accrued expenses that wasn’t there in December. Accrued income taxes came in at $36.5mm against nothing at all at year-end. Accrued compensation and vacation ran $15.5mm versus $15.1mm, and other accruals $11.3mm versus $11.2mm. The effective tax rate for the half was negative 18.2%, meaning tax expense on a pre-tax loss, driven by the non-deductibility of the stock compensation and a valuation allowance against part of the deferred tax assets. Management guided the full-year effective rate to roughly 7% and said the long-run rate will be higher. Don’t model 7% past 2026.

Plenty isn’t disclosed, and I won’t pretend otherwise. Federal and state NOL carryforwards, gross deferred tax assets and the valuation allowance balance aren’t in the interim filing. Operating lease right-of-use assets and liabilities aren’t broken out on the face of the balance sheet, so they presumably sit inside other assets of $43.4mm and other non-current liabilities of $50.6mm. There’s no pension or OPEB footnote at all, which is worth a look in the first 10-K given a business with 1900 roots. Customer concentration isn’t disclosed in the 10-Q. Funded versus unfunded backlog isn’t disclosed. Book-to-bill isn’t disclosed. Contract type mix, firm-fixed-price versus cost-plus, isn’t disclosed. Applied is an emerging growth company filing on scaled disclosure with two years of audited financials and no auditor attestation on internal controls. The first 10-K in March 2027 and the first proxy in spring 2027 will roughly double what you can see. Until then you’re underwriting with less information than you’d want, and that’s part of why the stock is cheap.

No impairment. The goodwill rollforward shows $342.5mm at year-end, plus $238.9mm from acquisitions, zero impairment, ending at $581.4mm. For a roll-up that just re-based its equity, skipping the kitchen-sink writedown at the IPO is a small quality signal. They didn’t clear the decks, because they had nothing to clear.

Read this before you buy anything, because it sets the size.

The executed underwriting agreement dated June 2, 2026 runs the restricted period “commencing on the date hereof and ending at 4:00 p.m., New York City time, on the Trading Day … that is the 180th day.” One hundred eighty days from June 2 lands on Sunday, November 29, 2026, which puts practical expiry on or about Monday, November 30. The company hasn’t stated the date. That’s my arithmetic and I’m flagging it.

Morgan Stanley and Jefferies hold waiver authority in their sole discretion, with three business days’ notice to the company and a required press release two business days before effectiveness. Carve-outs include employee equity, option exercises, S-8 registrations, 10b5-1 plans, tax withholding, and strategic transactions up to 10% of outstanding shares.

Greenbriar, through AA&D Holdings, LP, owns 126,786,731 shares, or 73.5% of the company. Public float runs 43.58mm shares, about 25% of shares outstanding, roughly $790mm of value. Twenty-day average volume is about 961,000 shares, or $17mm a day.

On July 15, 2026, and nobody wrote about this, AA&D Holdings made a pro rata distribution of 11,456,787 shares to its limited partners, “including certain directors and officers of the issuer,” at $0.00 consideration, transaction code J. That’s what took the sponsor from 80.2% to 73.5%.

It isn’t a sale. The distributed shares stay locked up. But it’s exactly the housekeeping a sponsor does before it starts selling, and it converts 11.5 million shares from one disciplined holder into a diffuse population of limited partners who each make their own decision on November 30. Against a 43.6 million share float, that population plus a 126.8 million share sponsor block dominates supply and demand in this stock for the next four months.

The Stockholders’ Agreement dated June 4, 2026 sets board nomination rights off the “Original Amount” Greenbriar held at the IPO, not off shares outstanding. Hold 40% or more of that Original Amount and Greenbriar nominates 100% of the directors. At 30% it nominates 40% of them. At 20%, 30%. At 10%, 20%. At 5%, one director plus a non-voting observer.

Do the arithmetic. Greenbriar can sell down to roughly 29% of the company and still nominate the entire board. Applied is a controlled company under NYSE rules, exempt from majority-independent-board and independent compensation and nominating committee requirements. A Registration Rights Agreement dated the same day is the mechanism through which secondary offerings will come, and I couldn’t retrieve its demand count, piggyback terms or standstill.

This is the largest structural negative in the name and I won’t soften it. Buy AADX and you’re a minority partner in a Greenbriar asset, and Greenbriar’s job is to sell. The counter-argument is that this is also why the stock is cheap, and that a marketed secondary at a discount is a known, dateable, bounded event rather than an unknowable one. Karman carries a similarly sponsor-heavy register and is up 22% over 52 weeks anyway, though I haven’t verified how its own lock-up expiry was handled.

How I handle it: a marketed secondary is the base case here rather than the tail, so the position gets sized to survive one and to keep capital available to meet it. A dated, bounded supply event is the rare kind of risk you can actually plan around. The mistake is being fully sized before it happens.

Trip Ferguson runs the company. CEO since November 2025. He was President of Space, Cyber and Directed Energy at AeroVironment from May to October 2025, Chief Operating Officer of BlueHalo from September 2022 through its acquisition by AeroVironment, and Chief Operating Officer of Dynetics from November 2018 to September 2022. Marine Corps officer, attained Captain, three overseas tours. BS Economics with merit from the Naval Academy, MBA from the University of Alabama in Huntsville.

Two honest observations. The five-month AeroVironment tenure is a fast exit post-acquisition, and he has no prior public-company CEO experience. His Form 3 at the IPO showed zero directly-owned shares, and unlike his CFO and three directors he didn’t buy in the directed share program. He almost certainly received units in the July LP distribution, though I can’t confirm the amount. I’d rather see him on the tape with a code P.

Jeffrey McRae is CFO. He was SVP and CFO of Triumph Group from February 2014 until Triumph named a successor in 2016, and before that President of Triumph Aerostructures–Vought Aircraft Division and President of Vought Integrated Programs. Twenty years at BAE Systems and predecessors, including interim President and GM of the Armament Systems Division. Started his career at KPMG. BS Accounting, Michigan State.

Context worth knowing: Triumph itself deteriorated badly during McRae’s tenure there. I couldn’t find any statement of the reason for or exact date of his departure, and I won’t imply a connection I can’t source. What I can source is that he put $500,000 of his own money into this stock at $20 on June 4.

Christopher Rogers is President and Chief Strategy Officer. He joined as Chief Growth Officer in December 2025 and got the President and CSO title effective July 27, 2026, board-approved August 7, announced August 10. Before that he spent 2005 to 2025 at Harris Williams as Managing Director and Head of the Aerospace, Defense and Government Services Group. Marine Corps officer. BS History from the Naval Academy, MBA from Harvard Business School. His compensation wasn’t modified with the promotion.

Applied hired the banker who covered this exact sector for twenty years and made him president eight months later. Read that as an M&A hire, because that’s what it is.

Kevin Bidlack is COO, formerly CEO of AASC through the Greenbriar acquisition and the ICEL and NeXolve deals, 30-plus years, Cal Poly San Luis Obispo with an MBA from CSU Stanislaus. Tom Holzthum is Co-COO, formerly CEO of PCX Aerosystems from September 2021 and Chief Commercial Officer from January 2021, with 25-plus years at Triumph Group including EVP of a $1.1bn Integrated Systems unit, BS Mechanical Engineering from RPI. Kai Kasiguran is Chief Accounting Officer, formerly VP and Controller at Triumph with seven years at Ernst & Young as an audit senior manager, and he’s a CPA. Joe Maisto is Chief Administrative Officer covering administration, contracts, integration and risk, formerly at Whitcraft and Pursuit Aerospace and PwC, Cornell. Matt Brush is Chief People Officer, Amit Shah is CIO, and Dr. David Spencer, the Vestigo founder and a former NASA JPL planetary mission designer, is VP of Deployable Systems.

Three roles are missing. There’s no named General Counsel or Chief Legal Officer. There’s no named Chief Technology Officer. There are no named segment presidents. For a $3bn public company with 87% sole-source government-adjacent revenue and an active M&A program, the absence of a general counsel is a governance gap worth watching. It isn’t disqualifying. It’s a note in the margin with a date on it.

I searched specifically for litigation, False Claims Act matters, restatements and short-seller reports touching AASC, PCX or Applied and found none. The 10-Q reports no material legal proceedings, only boilerplate about routine government procurement audits.

The board runs eight members. David King chairs it, and he was CEO of Dynetics from 2015 to 2020, Group President of Dynetics at Leidos from 2020 to 2022, and earlier held senior leadership at NASA. He sits on Compensation and on Nominating and Governance. Ferguson holds the CEO seat. Noah Roy is a Greenbriar Managing Partner, there since 2008, previously an MD in Goldman Sachs’ aerospace and defense group from 2003 to 2008, and he sits on Compensation and Nominating. Noah Blitzer is a Greenbriar Managing Director in A&D, there since 2011, previously at Citi, and he was the deal principal on both PCX in 2021 and AASC in 2022. He chairs both Compensation and Nominating and Governance. Jack Morris is a Greenbriar director, there since 2015, previously at Barclays, and sits on Audit. Susan Lynch is independent, was CFO of V2X from 2019 to 2023, sits on the boards of Crane Company and Onto Innovation, and chairs Audit. James Katzman is independent, was SVP of Corporate Development at GE and GE Aerospace from 2021 to 2025, is a retired Partner of Goldman Sachs, sits on the boards of Brinker International and the Hershey Trust, and sits on Audit. Scott Goldstein is independent, is SVP and Fellow at Parsons, was Chief Scientist at Anduril from 2021 to 2023, and is a retired USAF Major General with dual EE degrees, a PhD from USC, 200-plus publications and five patents.

Three of eight seats are Greenbriar, and Blitzer chairs both Compensation and Nominating.

Look at the other five. The chairman ran Dynetics and before that held senior leadership at NASA. The CEO was Dynetics’ COO under him, which is a pre-existing working relationship worth knowing about. Katzman ran corporate development at GE Aerospace and is a retired Goldman partner, which makes his an M&A seat rather than a decoration. Goldstein was Anduril’s Chief Scientist and a two-star Air Force general. Lynch was a public-company defense CFO. That’s a board assembled to buy things and to sell the company. You don’t put a retired Goldman partner who ran GE Aerospace corp dev and a former Anduril chief scientist on a build-to-print structures board unless the plan is bigger than build-to-print.

Now to the people who bought it.

On June 4, 2026, settlement day, four insiders bought stock at the $20.00 offering price through the directed share program.

• Jeffrey L. McRae, Chief Financial Officer. 25,000 shares at $20.00. $500,000. Held 25,000 after.

• James C. Katzman, Director. 25,000 shares at $20.00. $500,000. Held 29,250 after.

• Susan D. Lynch, Director. 8,000 shares at $20.00. $160,000. Held 8,000 after.

• Scott Goldstein, Director. 500 shares at $20.00. $10,000. Held 4,750 after.

That’s 58,500 shares and $1,170,000 across four buyers. Zero sellers. Zero open-market sales by anyone since the IPO.

Two of those carry weight. The CFO wrote a half-million-dollar personal check for stock in the company whose books he keeps, on the day the offering closed. Katzman, the retired Goldman partner who ran corporate development at GE Aerospace, matched him. The share-based comp charge was mechanically triggered by the offering itself, so the accounting outcome was knowable on the day they bought, though I can’t establish what any individual knew or when.

One thing I want to be careful about: the prospectus reserved 1,625,000 shares for the directed share program, and my sources conflict on whether director and officer purchases through it are lock-up restricted. I couldn’t resolve it against the 424B4, so I’m not leaning on it. What isn’t in dispute is that four insiders wrote personal checks at $20 on the day the deal closed.

The stock is $18.11. You can buy it 9.5% cheaper than the CFO did, with the bad quarter already behind you.

Against that, the CEO bought nothing. His Form 3 showed zero directly-owned shares. I’d rather he had. It’s the one piece of the alignment picture that doesn’t fit and I won’t explain it away.

Now the ownership side, which is the most striking data in the file and the reason “nobody is talking about it” is a measurement rather than a flourish.

• AA&D Holdings, LP (Greenbriar). 126,786,731 shares. 73.5% of the company.

• Vanguard Portfolio Management. 1,273,937 shares. 0.75%.

• T. Rowe Price Group. 1,256,748 shares. 0.74%.

• Vanguard Capital Management. 1,185,115 shares. 0.69%.

• Capricorn Fund Managers. 310,000 shares. 0.18%.

Strip out the sponsor and you’re left with about 4.0 million shares, of which 2.46 million is Vanguard index product. Active managers hold roughly 1.6 million shares, about 3.6% of the float, and essentially all of it is T. Rowe. No hedge fund. No crossover. No named long-only of consequence. One third-party report claims a 3.25mm-share Temasek position that I couldn’t reconcile against the primary holder tables, so treat it as unconfirmed. MarketBeat’s summary runs one sentence: “Applied Aerospace & Defense has minimal institutional ownership at this time.”

The attention side matches. Seven analysts sit in the S&P Global consensus and eight firms publish a target. Kratos has 21. AeroVironment has 20. Rocket Lab has 18. Karman has 11. Seven of the eight underwrote the deal, with UBS the only publishing firm outside the syndicate, and six of the initiations printed on June 28 and 29 at the end of the research quiet period. QuiverQuant shows no WallStreetBets mentions, no retail sentiment and no social data for the ticker. I found one independent newsletter write-up and no Reddit discussion. Short interest runs 1.57mm shares, about 3.6% of float, 1.36 days to cover, so there isn’t even a short thesis being expressed.

Nobody talks about this stock. It has existed as a public equity for seventy-six days, 73.5% of it sits locked in a private equity vehicle, and almost everyone paid to have an opinion on it also sold it to you. The whole setup runs on obscurity. Crowded names don’t rerate, which is the same reason the names I size up tend to be ones nobody has heard of.

Every retail investor who wants defense tech owns Kratos Defense & Security Solutions, which trades as KTOS. Here’s why I think Applied is the better construction, and Kratos gets its due first, because the bull case there is real.

Kratos has an LTM book-to-bill of 1.3:1, a record $2.084bn backlog, a $15.0bn bid pipeline, $1.24bn of net cash, hypersonics revenue compounding from about $200mm in 2025 to a stated $700mm-plus in 2027, the MACH-TB 2.0 franchise, the Prometheus solid rocket motor joint venture with Rafael, and GE Aerospace’s name on a second-source JASSM engine. Kratos in 2026 has a genuinely different revenue mix than Kratos in 2019. Defense Rocket Systems grew 50.2% organically last quarter, Turbine 43.3%, Microwave 29.5%.

Now the arithmetic, with AADX first and KTOS second in each line.

• Price on August 17, 2026: $18.11 versus $63.29

• Market cap: $3.12bn versus $11.88bn

• Enterprise value: about $3.51bn versus $10.64bn

• TTM revenue: $575.9mm versus $1,523mm

• Adjusted EBITDA margin: 21.8% in Q2 versus 8.3% in Q2, and 8.1% for the Kratos first half, down from 8.4%

• GAAP operating income in Q2: negative $96.2mm, all of it share-based comp, versus negative $1.6mm on $458.8mm of revenue

• FY2026E free cash flow: guided positive in the second half versus guided negative $85mm to negative $105mm

• Cumulative FY21 through FY25 free cash flow: not applicable, Applied was private, versus negative $219.9mm

• Share count change over 12 months: plus 33%, which is the IPO itself, versus plus 15.1%

• Share count change over 5 years: not applicable versus plus 49.8%

• Share count change over 10 years: not applicable versus plus 186.6%

• Backlog to TTM revenue: 1.96x, or 1.65x on GAAP RPO, versus 1.37x

• EV to forward adjusted EBITDA: about 23x versus about 61x

• Insider activity over six months: four buys and zero sells versus 197 trades, all of them sales, including 800,000 shares by the CEO for roughly $63.4mm

• 52-week price change: not applicable, Applied IPO’d in June, versus negative 7.6%, and negative 27.6% year to date

Five things fall out of that.

Kratos is a 9% EBITDA-margin manufacturer valued as a technology company. Applied earns 21.8% and trades at roughly a third of the multiple. Gross margin at Kratos went from 27.74% in 2021 to 22.86% in 2025, down 490bp across the best defense demand environment in forty years. Applied’s underlying gross margin, adjusted for the IPO charge, sits near 28%.

Free cash flow at Kratos has never worked, and management’s own guidance says it still won’t. Negative in four of the last five fiscal years. Negative $128.9mm on a trailing basis. When Kratos raised revenue guidance by $145mm in August, it simultaneously cut operating cash flow guidance by $20mm and made free cash flow guidance worse. Growth there consumes cash. Unbilled receivables rose $104.3mm in the first half alone. Applied has the same working-capital dynamic at 21.8% margins instead of 8.3%, so the growth pays for itself sooner.

Dilution is the Kratos business model. $2.26bn of stock issued in thirty months. The February 2026 offering priced 16.4 million shares at $84.00 and those buyers are down about 25%. Authorized shares just went from 195 million to 245 million, which is headroom for more. Here’s the punchline: revenue per share was $6.70 in 2019 and $8.11 today, a 2.7% compound annual growth rate across seven years of the best story in defense. Revenue in 2012 was $969mm. Revenue in 2025 was $1,347mm. Thirteen years at 2.6% a year.

Backlog coverage favors Applied. Its $1.13bn contract backlog covers 1.96x trailing revenue and, per the CFO, “roughly about half of that backlog should convert to revenue in 2027.” Be precise, because $1.13bn is management’s contract-backlog metric and GAAP remaining performance obligations are $947.6mm, 19% lower, or 1.65x trailing revenue. Use RPO when you want the contractually firm number. Kratos’s $2.084bn covers 1.37x. On either measure Kratos has less contracted visibility than its narrative implies.

Then the insiders. Four Applied insiders bought $1.17mm of stock and none has sold. At Kratos, every one of 197 insider transactions in a six-month window was a sale, including 800,000 shares by the chief executive for roughly $63 million.

I’m not short Kratos and I wouldn’t be. The hypersonics franchise is real and the balance sheet is genuinely a fortress. But the market in 2026 has been clear about what it will and won’t pay for, and the evidence sits in the tape. Over 52 weeks: Astronics up 234%, Moog up 134%, Ducommun up 124%, Park Aerospace up 102%, AAR up 85%, Curtiss-Wright up 43%. All profitable, cash-generating hardware suppliers with real capacity. Against that: Kratos down 7.6%, AeroVironment down 26.6%, Firefly down 42.6%, AIRO down 57.4%. All cash-burning narrative names.

The market pays for margin and cash conversion. Kratos trades like a re-rating supplier while earning like a de-rating one.

Here’s the peer set, with enterprise value, TTM revenue, revenue growth, EBITDA margin, EV/Sales and EV/EBITDA.

• AADX, Applied Aerospace & Defense. $3.51bn EV, $575.9mm revenue, plus 47.4% in Q2, 21.8% margin, 5.2x FY26E sales, 23.0x FY26E EBITDA.

• KRMN, Karman Holdings. $9.01bn EV, $589.6mm revenue, plus 50.2%, 26.1% margin, 15.3x sales or 12.2x forward, 58.6x EBITDA or 41.2x forward.

• DCO, Ducommun. $3.39bn EV, $865.1mm revenue, plus 9.2%, 14.8% margin, 3.9x sales, 26.4x EBITDA.

• PKE, Park Aerospace. $722mm EV, $76.2mm revenue, plus 20.1%, 22.3% margin, 9.5x sales, 42.5x EBITDA.

• CW, Curtiss-Wright. $26.38bn EV, $3.65bn revenue, plus 10.5%, 23.1% margin, 7.2x sales, 31.3x EBITDA.

• BWXT, BWX Technologies. $17.13bn EV, $3.51bn revenue, plus 22.7%, 13.6% margin, 4.9x sales, 35.9x EBITDA.

• MOG.A, Moog. $15.15bn EV, $4.32bn revenue, plus 15.8%, 14.7% margin, 3.5x sales, 23.9x EBITDA.

• ATRO, Astronics. $4.35bn EV, $942.1mm revenue, plus 14.5%, 15.8% margin, 4.6x sales, 29.2x EBITDA.

• HEI, HEICO. $54.26bn EV, $4.91bn revenue, plus 18.8%, 28.0% margin, 11.1x sales, 39.5x EBITDA.

• KTOS, Kratos. $10.64bn EV, $1,523mm revenue, plus 25.5%, 5.7% margin, 7.0x sales, about 61x forward adjusted EBITDA.

• YSS, York Space Systems. $1.46bn market cap with EV not published, $405.0mm revenue, plus 59.7%, negative margin, 3.6x price to sales.

Applied’s EV comes from $3.12bn of market capitalization plus $387.7mm of net debt. The FY26E multiples use guidance midpoints of $680mm revenue and $152.5mm adjusted EBITDA. Peer data comes from stockanalysis.com statistics pages and company releases as of the August 17 close.

Karman is the comp. Same customers, same product adjacency in payload adapters, fairings, nozzles, tanks and composites, nearly identical scale. Karman guides FY2026 to $730 to $745mm of revenue and $215 to $222.5mm of adjusted EBITDA. Applied guides to $670 to $690mm and $150 to $155mm. Karman trades at 12.2x forward sales and 41.2x forward EBITDA. Applied trades at 5.2x and 23.0x. That’s a 57% discount on sales and a 44% discount on EBITDA to the company it most resembles.

Part of that gap is earned. Karman runs a 30% EBITDA margin against Applied’s 22%, has 95% institutional ownership against roughly 3.6% of Applied’s float in active hands, has no lock-up cliff in front of it, and doesn’t carry a 73.5% sponsor. Part of it isn’t earned, and closing even half of it is worth a very large number.

Here’s what Applied’s stock is worth today, on today’s guided EBITDA, at each peer’s multiple.

• Aerostructures median, 13.8x: $9.96, down 45%

• Moog, 23.9x: $18.89, up 4%

• Ducommun, 26.4x: $21.10, up 16%

• Astronics, 29.2x: $23.58, up 30%

• Curtiss-Wright, 31.3x: $25.44, up 40%

• BWXT, 35.9x: $29.51, up 63%

• HEICO, 39.5x: $32.69, up 81%

• Karman forward, 41.2x: $34.20, up 89%

• Park Aerospace, 42.5x: $35.35, up 95%

Applied is priced at Moog’s multiple. Moog is a fine company growing 15.8% at a 14.7% EBITDA margin. Applied grew 47% last quarter, 19.8% organically, at a 21.8% margin, in a faster end market, with 40% of its factory floor empty.

The private market pays more than any of this. Tinicum and Blackstone took Senior plc private in April 2026 at 15.2x adjusted EBITDA. Tinicum bought TriMas Aerospace for $1.45bn at roughly 18x LTM adjusted EBITDA in November 2025. The same buyer, twice, in six months, for qualified A&D structures capacity. Applied at 23x forward sits above the sponsor take-out band and far below the public quality band, which is where a business with the right assets and the wrong margin sits.

The budget isn’t the thesis. The bottleneck is.

The FY2027 request puts $70.5 billion against missile procurement, up 188% from $24.4bn, split Army $36.6bn, Navy $22.6bn, Air Force $11.3bn. PAC-3 MSE goes from 357 units to 3,203, though reporting on that line isn’t consistent, with one outlet at 3,203 and another at 2,798, the difference being a mandatory tranche that needs separate congressional approval. THAAD goes from 55 to 857. Tomahawk from 55 to 785. AMRAAM from 464 to 1,811. Standard Missile from 166 to 540. GMLRS runs 4,824 rockets. PrSM runs 1,134.

Unlike prior budget spikes, this one is getting locked with contracts. In a six-week window across July and August 2026, Lockheed signed a $58.62 billion seven-year PAC-3 MSE multiyear with a commitment to triple capacity by 2030. Raytheon signed a $22.9 billion seven-year Tomahawk deal taking output from 60 a year toward more than 1,000, a 17x ramp. Northrop signed a $3 billion seven-year framework to become the second supplier of Patriot rocket motors and to supply THAAD mid-body shells. L3Harris signed frameworks to quadruple THAAD propulsion and nearly triple PAC-3 MSE motor output.

RTX chief executive Chris Calio explained on the Q2 call exactly why that matters to a supplier: “If you can go out and give a supplier a seven-year firm order, they will lean forward, they will make the investment.”

The primes are building a merchant supply base from scratch, in public. Anduril is teaching existing nozzle suppliers to make motor cases on their composite-winding machinery. Ursa Major is developing composite cases to bypass steel. L3Harris spent $30 million to drive a 1,000% capacity increase at a single case supplier.

That last number is the most important datapoint in this entire report. It establishes three facts at once. Case capacity is the binding constraint. Primes will fund and qualify merchant case suppliers directly. And the return on relieving that constraint gets measured in multiples rather than percentages.

Now it’s a legal obligation. Executive Order 14415 gives the Secretary 90 days to develop strategies to “expedite testing and qualification of new domestic sources” and 180 days to identify acquisitions that depend on unreliable suppliers and require contractors to qualify alternatives, with contract termination available as the remedy. From January 1, 2027, waivers for non-compliant material are largely prohibited.

That inverts the usual reading of Applied’s 86% single-source concentration. The bear reads it as concentration risk. Management reads it as a starting point, and they’re right. Chris Rogers put it this way: “Our expectation is that [single-source revenue] will go down in part because a lot of the opportunities we’re seeing to dual source are very attractive.” Second-sourcing in 2026 is a demand tailwind for a qualified independent. The 86% falling is the mechanism of growth.

The airframe leg is inflecting too. On June 17, 2026 the Air Force ordered both the General Atomics FQ-42A and Anduril’s FQ-44A Fury into production, four months ahead of schedule, as a deliberate dual-source strategy. Anduril’s Arsenal-1 in Ohio is rated at 150 Fury aircraft per year and rolled its first Ohio-built airframe in late July. The FY2027 request carries the first-ever CCA procurement line at $996.5mm plus $1.3bn of RDT&E, and the Air Force Secretary’s near-term goal is “more than 150 combat-capable CCA by end of the decade,” against a long-run ambition of 1,000 to 2,000 that a prior Air Force Secretary floated in April 2024 and that isn’t current policy. Separately, Bell’s MV-75 fielding got pulled from 2030 to 2027, with 334 aircraft planned through FY2040 and the first procurement dollars, $266mm, in FY2027.

Applied has content on Fury. Applied has content on MV-75. Neither is meaningful in the current revenue base.

Now the counterweights, because you’re not getting only one side.

Blue Origin is grounded. On May 28, 2026 a New Glenn first stage exploded during a static fire, destroying the booster, an attached fueled second stage, and severely damaging LC-36, Blue Origin’s only operational New Glenn pad. Repairs or an alternative site “could take more than a year.” Management said on the call that “all indicators and direction from Blue Origin have been to continue moving at pace,” and the manifest behind it is real, with up to 27 Amazon Leo missions, seven NSSL Lane 2 flights and Blue Moon. New Glenn launch-structure content is at risk for twelve months or more, and it’s a named growth driver in this story.

Golden Dome’s 2027 funding isn’t appropriated. The FY2027 requirement is $17.5bn, roughly 97% of it dependent on an unpassed $350 billion reconciliation package, with only about $400mm in the base budget. On August 12, 2026, the same day Applied reported, General Guetlein said publicly: “If [lawmakers] don’t figure it out, there is no Golden Dome because there is no funding.”

The whole defense topline is a spike rather than a plateau. The FY2027 request is $1.5 trillion, up 44%. CSIS projects funding declines 16% in real terms from FY2027 to FY2028 and then runs flat with inflation. Anyone capitalizing 2027 growth into a terminal multiple is making an error. I haven’t.

NASA is going the other way. The FY2027 NASA request is $18.83bn against $24.44bn enacted, down 23%, the lowest in 66 years inflation-adjusted, with Science down 46%. Only Artemis and exploration grow. Civil space isn’t the tailwind.

Now the part that sits in no sell-side model.

Here’s the part that sits in no sell-side model, because it isn’t a forecast. It’s an asset.

Ferguson on the Q2 call: “We operate 11 purpose-built U.S.-based facilities across 6 states with more than 1.5 million square feet of production capacity.” And separately on the same call: “We estimate roughly 40% of that capacity is available today, measured across floor space, equipment capacity and workforce flexibility.”

Do the math with me. Applied guides to $680mm of revenue out of roughly 900,000 used square feet, which is about $756 of revenue per utilized square foot.

Read the quote carefully first. The 40% is a blended availability estimate across floor space, equipment and workforce. It isn’t 600,000 vacant square feet and I won’t pretend it is. But take it at face value and map it one-for-one onto square footage, and the illustrative arithmetic says $453mm of additional revenue could get produced inside buildings the company already owns, on equipment it already owns, in facilities already qualified, certified and audited. That’s my calculation on management’s estimate, not a company figure.

Illustrative total revenue capacity inside the existing four walls: roughly $1.13 billion, about two-thirds more than the company is guiding to this year, without building anything.

The incremental economics on filling qualified idle capacity aren’t average economics. They’re contribution economics, because the fixed overhead is already being carried. At a 25% incremental EBITDA margin, filling that space adds $113mm of EBITDA. At 30%, $136mm. At 35%, $159mm.

Holding the caveat above, Applied’s existing physical plant running full at plausible incremental margins would support something in the range of $265mm to $310mm of adjusted EBITDA. The company guides to $152.5mm this year and the market pays 23x for that.

You’re being handed that option for free, and it’s why the capital intensity question resolves in Applied’s favor. Capex guidance for 2026 runs roughly $50mm, about 7% of revenue, and construction in progress is up 3.6x to $14.6mm, which is the capacity build management described. Kratos is spending $250 to $270mm of total investment in 2026 and guiding to negative free cash flow to buy growth. One company has to build the factory. The other one already has it, empty, waiting for a purchase order.

In an industry where qualified capacity is the binding constraint, and where an executive order compels primes to qualify second sources on a 180-day clock, having 600,000 certified square feet already sitting there isn’t a footnote. It’s the asset.

Now the quotes nobody wrote down.

Collectively these describe a company that thinks 2027 is much bigger than 2026, said so out loud, and got no credit for it.

Ferguson on demand: “We are incredibly excited about 2027 and beyond. We are seeing the same demand signals that many of the primes and our customers are seeing.”

Ferguson on his own conviction after the print: “Our job for the balance of this year is execution. And I’ll tell you that I’m even more optimistic about what is in front of this company today than I was a quarter ago.”

Rogers, the man they just made president, whose entire prior career was valuing these assets at Harris Williams: “What stands out most is the sheer scale of the integrated opportunity that sits in front of us.”

Rogers on the capex already being spent: “We’ve made the capacity investments these ramps require, and we’re integrating our capabilities more tightly to drive speed and performance.”

Rogers on the acquisition and the pipeline: “CBI is delivering the synergies we underwrote and our M&A pipeline remains exciting and very active.”

Ferguson on content opportunity with the primes: “There remains vast white space.”

Management on the cross-sell that only exists post-merger: “The customer comes to us for capability and finds 3, and we win work we simply couldn’t have won as separate businesses.” Transcript sources disagree on whether that’s Ferguson or Rogers, so attribute it to management.

McRae on 2027 visibility: “Roughly about half of that backlog should convert to revenue in 2027.”

McRae on the margin recovery being mechanical: “As those programs start to mature through the second half of the year, we will see margin expansion there,” and “Most programs will reach our target margin profile within the first several units.”

McRae on cash: “We project generating positive free cash flow in the second half of the year.”

Ferguson on the strategic frame: “Applied is purpose-built for this mission, and we’re confident our strategy provides an opportunity for asymmetric upside.”

That last one is a chief executive using the word asymmetric on his first earnings call as a public company. Either he’s reading the same setup I am, or it’s a coincidence.

Three routes here, and they don’t all agree. I’ll show you where they disagree.

Start with the reverse DCF, because it’s the most useful number in this report.

Build a straightforward five-year model from guidance. FY2026 revenue of $680mm at a 22.4% adjusted EBITDA margin, then $790mm, $900mm, $1,010mm and $1,120mm at margins of 23.0%, 24.0%, 24.8% and 25.5%. That’s a 13.3% revenue CAGR, below BofA’s own 14% assumption, and roughly 310bp of margin expansion over four years, below BofA’s 350bp. Tax at 21%. Capex declining from 7% to 5.5% of revenue. Working capital at 12% of incremental revenue.

Discount at a 10.5% WACC, back out $350mm of net debt at year-end 2026, and solve for the terminal multiple embedded in today’s $3.51bn enterprise value.

• Assume a 16x exit multiple, and the market is implying FY2030 adjusted EBITDA of $291mm

• Assume 20x, and it’s implying $233mm

• Assume 24x, and it’s implying $194mm

Sit with the middle row. At $18.11, on a 20x exit multiple, the market prices Applied to generate $233 million of adjusted EBITDA in 2030.

RBC’s price target, the most explicit published model on the name, already underwrites $230 million of adjusted EBITDA in 2028.

The market is discounting the sell-side’s 2028 number to 2030. It took two full years out of the ramp in three trading sessions, on a non-cash charge. That’s the mispricing, stated as precisely as I can state it.

Now the DCF, honestly. Same model, exit-multiple method, sensitized. At a 9.5% WACC: $16.04 on a 14x exit, $18.30 at 16x, $22.81 at 20x, $27.33 at 24x. At 10.5%: $15.42, $17.60, $21.96 and $26.31. At 11.5%: $14.83, $16.93, $21.13 and $25.33.

I’ll say the unpopular thing. The DCF doesn’t make this look like a screaming bargain. At a 10.5% WACC and a 16x exit you get $17.60, roughly where it trades. You need a 20x exit to reach $22 and 24x to reach $26.

That’s a feature of the analysis. This is a relative-value and capacity-option story rather than a discounted-cash-flow story. The DCF is back-of-envelope, and the full version belongs in the three-scenario template. What it does tell you is that at $18.11 you aren’t paying for optimism. You’re paying a 16x terminal multiple on a supplier whose closest peers trade at 26x to 42x, in a sector where the private market clears at 15x to 18x for lower-growth assets. The downside already sits in the price. Nobody setting that price has imagined the upside.

Third route, comps, which is where the number comes from. Base case FY2027 adjusted EBITDA of $190mm, which is $790mm of revenue at 24.0%, net debt down to about $290mm, about 176mm shares. At 26x, Ducommun’s current multiple with a modest premium for growth, that’s $26.42. Cross-check on FY2028: $216mm of adjusted EBITDA at 22x, net debt about $215mm, 177mm shares, gives $25.63. Two independent routes, two years apart, landing within eighty cents of each other.

Guardian Research price target: $26.

Now the four scenarios.

Bear case, minus 49%, gets you $9.25. FY2028 revenue of $760mm at a 20.0% margin on a 13.0x multiple. The second-half 2026 guidance misses. Margin recovery doesn’t arrive. A continuing resolution stalls 2027 program starts. New Glenn stays grounded. The sponsor markets a large secondary into a weak tape at a discount. The multiple compresses to the aerostructures median. This is the scenario I’m underwriting against.

Base case, plus 42%, gets you $25.63. FY2028 revenue of $900mm at a 24.0% margin on a 22.0x multiple. Guidance gets met. Margins normalize as early-stage programs exit learning curves. Backlog converts at the stated 50% rate into 2027. The lock-up clears with a manageable secondary.

Bull case, plus 97%, gets you $35.65. FY2028 revenue of $1,010mm at a 26.5% margin on a 24.0x multiple. Idle capacity starts filling. Dual-source awards convert under EO 14415. C5ISR keeps compounding on the munitions multiyears. Margin closes half the gap to Karman. The multiple rerates to the quality-supplier band.

Blue sky, plus 207%, gets you $55.53. FY2028 revenue of $1,150mm at a 28.5% margin on a 30.0x multiple. Applied becomes a designated merchant second source on one or more major motor-case or interceptor programs. Fury and MV-75 reach rate with Applied content. Margin reaches Karman’s structure on Applied’s capacity. The market decides it’s a Karman rather than a Ducommun, or a strategic decides the same thing and buys it.

Scenario mechanics so you can audit me: price equals FY2028 adjusted EBITDA times the multiple, minus net debt, divided by shares. Net debt and share count differ by scenario, because a company that misses deleverages more slowly and dilutes more. Bear uses $330mm and 178.0mm shares. Base uses $215mm and 177.0mm. Bull uses $150mm and 176.0mm. Blue sky uses $60mm and 176.0mm. Run them all on the base balance sheet instead and you get $9.95, $35.08 and $54.34. The conclusions don’t change.

Note the shape of that distribution. The base case, which assumes management merely does what it guided, implies 42% upside, and at a deliberately punitive 19x it still gets you to $21.97, or 21%. The bull case isn’t heroic. It requires filling factory space that already exists. And the blue-sky case doesn’t require inventing anything. It requires Applied to earn the margin its nearest comparable already earns.

Now the sell side, and why to discount it.

• Baird, Peter Arment, Outperform, $30, on 2028E EBITDA

• Jefferies, Sheila Kahyaoglu, Strong Buy, no published target

• RBC Capital, Kenneth Herbert, Outperform, $24, on 19.5x 2028E adjusted EBITDA of $230mm

• Stifel, Jonathan Siegmann, Buy, $24, on 2027E EBITDA

• UBS, $24

• BofA, Ronald Epstein, Buy, $23, cut from $24 on August 13 and 14, on 23x 2027E EV/EBITDA, modeling a 14% revenue CAGR from 2025 to 2030 and about 350bp of margin expansion

• Morgan Stanley, Kristine Liwag, Equal Weight, $24, raised from $23 on August 13, calling risk-reward balanced

• Wolfe Research, Outperform, $23

Consensus runs $25.14 across the seven analysts in the S&P Global set. MarketBeat, using a slightly different eight-firm register, shows $24.63. Range is $23 to $30. Every published target sits above the current price. The post-earnings action wasn’t uniformly negative either. BofA cut a dollar, Morgan Stanley raised its target on August 13 while staying neutral, and Baird didn’t touch its $30.

Now discount all of it. Seven of these eight firms underwrote the offering, with UBS the lone exception. Six of the initiations printed within a day of each other at the end of the quiet period, and there’s very little independent voice in the coverage. Use the sell side as a sanity check on direction and as a description of what the underwriters told institutions in June. Don’t use it as an edge. There isn’t one in a research report every buyer already read.

One more thing to consider. Greenbriar is a private equity firm holding 73.5% of an asset it has owned for four years. The board contains a retired Goldman partner who ran GE Aerospace corporate development. The president is the banker who covered this sector at Harris Williams for twenty years.

At 18x FY2027E adjusted EBITDA of $190mm, less $290mm of net debt, over 176mm shares, a whole-company take-out clears at roughly $17.78. At 20x, $19.94. That’s the honest answer and it isn’t encouraging. A financial-sponsor take-out at strip multiples creates no value from here, which tells you something important. The exit that pays is a strategic buyer or a public-market rerating. A strategic paying a control premium on a 24x forward multiple gets you to the high twenties, which is the realistic M&A ceiling and happens to sit right on my price target.

Now where Wall Street has it wrong.

On “it missed earnings by a dollar.” GAAP EPS was negative $1.04 against a consensus of negative one cent, and $0.74 of that gap is a single non-cash line, the one-time vesting of pre-IPO incentive units at the offering, which the 10-Q describes in a sentence and which won’t recur. Revenue beat by about 7%. Adjusted EBITDA grew 38.5%. Guidance came in above the standing consensus. That consensus was last updated July 1, before the quarter, before the guidance, before any of it.

On “margins are deteriorating.” Gross margin printed 22.2% against a prior-year 28.2%, and the 28.2% is BofA’s figure rather than a company disclosure, since Applied doesn’t publish a Q2 2025 gross margin. The company-sourced comparable is FY2025’s 27.9%. Management disclosed that roughly six percentage points of that decline is the same IPO share-based comp allocated into cost of sales, which puts underlying gross margin near 28%. The adjusted EBITDA margin decline, 21.8% versus 23.2%, is real, and it’s explained. Roughly 20% of first-half revenue came from early-stage development programs still working down learning curves, which reach target margin “within the first several units.” That’s a mix problem with a known half-life.

On “it’s a private equity roll-up with negative tangible book.” Both true, and this is where I concede the most ground. $934.7mm of goodwill and intangibles against $827.9mm of equity is no fortress, and purchase accounting on the largest deal is still preliminary. The operating question is whether the assembled asset works, and the evidence says it does: 87% sole-source revenue, 39-year average customer relationships, 21.8% adjusted EBITDA margins, a record backlog, and a cross-sell management can point to by name. A roll-up that produces a differentiated qualified-capacity footprint behaves differently from one that produces a bigger version of the same commodity.

On “the sponsor owns 73.5% and the lock-up expires in November.” Correct, and it’s the reason this isn’t a max-size position. It’s also a known, dateable, bounded event, which is the rarest kind of risk in equities. The question that matters is whether the underlying business can absorb the supply, and a business guiding to positive free cash flow, growing EBITDA 38%, at 23x forward, generally can. I’d rather own a good business with a supply overhang I can put on a calendar than a mediocre business with none.

On “nobody covers it because there’s nothing there.” Seven analysts, almost all underwriters, seventy-six days after listing. That’s the normal state of a newly-public mid-cap before the second and third waves of coverage arrive. The coverage is the re-rating catalyst: non-underwriter initiations, the first 10-K in March 2027 with full segment and concentration disclosure, the first proxy in spring 2027, and the index inclusions that follow a larger float.

I own this. Here’s what would make me wrong. I publish the ones I get wrong too, so this list is the one I’ll be graded against.

The second-half guidance is a bridge too far. To hit the midpoint, Applied needs second-half revenue of $378mm against a first half of $302mm, up 25% half over half, and an adjusted EBITDA margin of 23.7% against 20.9%, a 279 basis point step-up. Management’s explanation, early-stage programs exiting learning curves, is credible and specific. It’s still a demanding half, and a company one quarter into public life missing its first guide would take the multiple down hard. This is the largest near-term risk and the Q3 print in November is the test.

Customer concentration. Third-party summaries of the S-1 risk factors put the top three customers at roughly 59% of FY2025 revenue with the largest at 31%. I couldn’t verify those against the primary filing and I won’t treat them as established. What’s confirmed from the prospectus is that about 87% of revenue is sole- or single-source. That cuts both ways. It’s the moat and it’s the concentration. Losing or de-scoping one large program would be material, and there’s no disclosure in the 10-Q to size it. The first 10-K fixes this.

The lock-up and the sponsor. On or about November 30, 2026, a 73.5% sponsor block plus 11.5mm shares already distributed to limited partners face a 43.6mm share float. A marketed secondary at a discount is the base case rather than the tail. Morgan Stanley and Jefferies can waive early with three days’ notice.

Cash is thin. $18.1mm on the balance sheet against first-half free cash flow of negative $103.7mm and a newly-accrued $36.5mm cash tax bill. The $125mm revolver is fully available and total liquidity runs about $143mm, which is adequate. If the second-half working capital release doesn’t happen they’ll draw the revolver, and drawing a revolver two quarters after an IPO looks bad and gets priced that way.

Program-specific risk. Blue Origin’s New Glenn is grounded with its only pad damaged and recovery possibly exceeding a year. Golden Dome’s FY2027 funding is about 97% dependent on an unpassed reconciliation bill. A continuing resolution delays new program starts. The FY2027 defense topline is a spike that CSIS projects declines 16% in real terms into FY2028.

Integration. Five acquisitions in twenty-four months, two of them closed on the same day in March 2026, with the largest still inside its purchase-accounting measurement period. Morgan Stanley flagged “early risks associated with a recently-combined company” when it initiated at Equal Weight. Roll-ups fail at integration far more often than at deal selection.

Governance and disclosure. Controlled company. Emerging growth company on scaled disclosure. No proxy yet, so no visibility on executive compensation, PSU targets or severance terms. No named General Counsel. No named CTO. Credit agreement financial covenants aren’t public.

They paid up for CBI. Roughly 3.5x trailing sales and near 20x trailing EBITDA for a business earning below the corporate margin, funded entirely with debt three months before filing the S-1. If CBI’s growth rate reverts to its historical trend rather than its four-month stub, there’s a $235mm non-deductible goodwill balance sitting on top of it.

One I’m dismissing: the “cash runway under one year” flag at least one screening service applies here. It annualizes a first half that contained a $110mm one-time charge, $19mm of transaction costs, a full quarter of interest on a $1bn debt stack that no longer exists, and a working-capital build management has committed to reversing. Less likely than the headline suggests.

Now the exit criteria.

No stop. No level. These are things the company would have to report, and any two together are enough.

Q3 revenue below roughly $175mm, or a fourth-quarter implied ramp that needs a bigger sequential jump than the one management just failed to deliver.

The margin step-up doesn’t arrive. Second-half adjusted EBITDA margin failing to clear about 22.5% against 20.9% in the first half, with the early-stage-programs explanation repeated rather than resolved. This is the single most important one. The entire base case rests on that 279bp of recovery being mechanical.

Free cash flow doesn’t inflect, and the revolver gets drawn to fund operations rather than a deal. That converts a thin cash balance into a financing problem, and a financing problem in a 73.5%-controlled company isn’t one you get a vote on.

Backlog declines sequentially, or the “roughly half converts to 2027” figure gets walked back on a later call.

A top-three customer or program is lost or de-scoped. At 87% sole-source I can’t diversify around this, and the first 10-K in March 2027 is when I’ll finally be able to size it.

FY2027 guidance comes in below about $780mm of revenue, or guides margins flat. Management has said “incredibly excited about 2027.” If the number doesn’t match the adjective, believe the number.

A goodwill or intangible impairment, or a material downward revision to the CBI purchase price allocation. $235mm of non-deductible goodwill on a business bought at a full price with debt is where a roll-up thesis usually dies.

Greenbriar dumps the entire block into a marketed deal at a steep discount, which would be information about what the sponsor thinks rather than about supply.

None of those is a price. All of them are disclosures. And note how they resolve: the first three all get answered in a single print in November. This isn’t an idea you have to hold blind for two years to find out about.

Here’s what I’d ask on the Q3 call.

On the guide: what’s the Q3 and Q4 revenue split inside the $378mm implied second half, and how much of the 279bp margin step-up is program mix versus the share-based comp comparison rolling off? What proportion of second-half revenue is already on firm purchase order today?

On cash: quantify the working capital release. Is the $36.5mm tax payment inside the “positive free cash flow in H2” guidance or outside it?

On the debt: what’s the current applicable margin on the term loan after the deleveraging? The initial spread was SOFR plus 625bp. If the grid delivers, say, 200bp of step-down on $406mm of principal, that’s roughly $8mm a year of pre-tax income nobody has modeled, though the 200bp is my illustration rather than a disclosed tier.

On backlog: funded versus unfunded. Book-to-bill for Q2 and Q3. What’s the dollar value of the roughly 50% converting to 2027, and how much of the balance sits under framework agreements versus firm orders?

On concentration: confirm or correct the top-three and largest-customer figures. Give us government versus commercial. Give us firm-fixed-price versus cost-plus.

On capacity: what revenue does the 40% available capacity represent at full use, and what incremental capex activates it? What’s the incremental EBITDA margin on incremental volume through existing plant?

On dual-sourcing: how many active pursuits, in what dollar magnitude, on what award timeline? Has EO 14415 produced inbound from primes?

On capital allocation: with the revolver undrawn and the M&A pipeline described as “exciting and very active,” what’s the leverage ceiling for the next deal? Is there any buyback authorization?

On the sponsor: anything the company can say about the lock-up and Greenbriar’s intentions.

The chart first, and why it isn’t the risk framework.

The chart is eleven weeks long. There’s no pattern in it and I won’t invent one. What follows is context on where supply and cost basis sit.

The IPO priced at $20.00, a dollar below the top of the $18 to $21 range. It opened at $20.75, printed $19.01 on day one, down 4.95% from the offer, and sank to $17.54 by June 5. It then ran hard into the analyst initiations, putting in its highest close of $23.85 on July 1, gave all of it back to a low close of $17.29 on July 22, rallied again to $21.75 on August 11, the day before earnings, and broke, closing $17.92 on August 14. It’s been a violent, two-way tape on thin volume, including a 20.0% session on June 11 and a negative 10.4% session on July 8. The full post-IPO intraday range is $16.57 to $24.24.

Levels worth knowing:

• $16.57 to $17.29. The intraday low and the lowest close since listing. Where the July flush found buyers.

• $17.92 to $18.11. Where it sits, roughly 9.5% below where the CFO and three directors bought.

• $20.00. The IPO price, and effectively the whole book’s cost basis. Expect supply on the first test from underwater June buyers.

• $21.75. The pre-earnings high. Reclaiming it says the market forgave the charge.

• $23.00 to $24.24. The post-IPO high and the bottom of the sell-side target range. Through here means the coverage caught up.

• $26.00. Target.

To be explicit about how I use those numbers: I don’t, as triggers. There’s no level at which I sell this because it went down. Eleven weeks of price history on a 43-million-share float with no institutional base isn’t a signal about anything. It’s a record of who happened to be trading. Those levels are context for where supply sits and where I’d rather add. The risk framework is the list above, and every item in it is a number the company reports.

One structural note on the tape does matter. Average daily volume runs roughly 961,000 shares, about 2.2% of a 43.6 million share float. Liquidity is thin, so work orders patiently. A market order in this name moves it. And the same thinness that let 17% come out of the stock on a non-cash charge moves it the other way when the second wave of coverage arrives.

Now the calendar.

• Q3 2026 results, around mid-November. The test of the 25% half-over-half revenue ramp and the 279bp margin step-up, plus the first look at whether free cash flow inflected.

• Lock-up expiry, on or about November 30. The 73.5% sponsor block plus 11.5mm distributed LP shares against a 43.6mm float. Risk and entry opportunity in the same event.

• Non-underwriter coverage initiations, Q4 2026 into H1 2027. The first independent voice is the re-rating mechanism.

• FY2027 guidance, with Q4 results around March 2027. The first look at what management thinks 2027 is.

• First 10-K, around March 2027. Full customer concentration, contract type mix, funded and unfunded backlog, tax attributes, lease detail, credit agreement covenants. The information asymmetry closes here.

• First proxy, around spring 2027. Executive compensation, PSU hurdles, employment agreements. If stock-price or EBITDA hurdles exist, that’s where management tells you its own targets.

• FY2027 appropriations, ongoing through late 2026. The $350bn reconciliation tranche determines Golden Dome and roughly half the munitions ramp.

• CCA production quantities finalized, before the end of fall 2026, per General Dale White’s stated timeline.

• New Glenn return to flight, late 2026 at the earliest. Restores a named growth driver.

• M&A, any time. The pipeline is “exciting and very active,” the $125mm revolver is undrawn, and a banker is president.

• Index inclusion, post-lock-up. A larger float is the precondition.

• Term loan repricing, any time. Undisclosed, unmodeled, potentially worth about $8mm a year pre-tax.

And the bottom line.

Applied Aerospace & Defense makes the parts everything else in American aerospace bolts onto, and it holds sole-source position on 87% of them across relationships averaging thirty-nine years.

It IPO’d eleven weeks ago at $20 and used every dollar to take net leverage from an estimated 8.8x to a stated 2.7x. No converts, no preferred, no warrants, no seller notes, and one $5 million earnout.

It grew revenue 47% last quarter and adjusted EBITDA 39%, then reported a $154 million GAAP loss because $110 million of pre-IPO incentive units vested on the offering and had to be expensed at once. The charge moved no cash and won’t recur. The stock fell 17% in three sessions and closed below its offering price.

The CFO and three directors bought $1.17 million of stock at $20 on the day the deal closed. Nobody has sold a share.

Seven analysts cover it. Almost all of them underwrote it. Active managers hold about 3.6% of the float. There’s no Reddit thread, no retail sentiment data, and one independent newsletter has written it up. The data providers still show an enterprise value built on a debt balance extinguished in June.

Six hundred thousand square feet of already-qualified American production floor sits available inside a company whose customers are being ordered by executive order to qualify second sources on a 180-day clock.

At $18.11 the market pays 5.2x this year’s revenue and 23x this year’s EBITDA for that, against Karman at 12.2x and 41x. On a reverse DCF it’s pricing 2030 adjusted EBITDA of $233 million, a number the sell side already underwrites for 2028.

The base case requires nothing more than management doing what it just guided, and it implies 42% upside. On a deliberately punitive 19x it still implies 21%. That’s the asymmetry.

The thesis breaks if the second half misses, if the margin recovery doesn’t arrive, if free cash flow doesn’t inflect, or if 2027 guidance doesn’t match the adjectives management is using about it. Not one of those is a price. They’re all disclosures, and the first three get answered in one print in November.

Below eighteen I’m adding. Twenty-six is the target. Thirty-five is where it goes if the factory fills. If it trades to fifteen on a secondary with the numbers intact, that’s the trade rather than the exit.

Guardian Research price target: $26.

If you want to see how I’m actually holding it, the chat is where the book lives. Up 82.29% year to date through August 14. Every entry, every trim, every add, posted as it happens. Come argue with me about the margin recovery.

→ Join the chat.

Disclosures

This report is for informational and educational purposes only. It doesn’t constitute investment, legal or tax advice, and it isn’t an offer or solicitation to buy or sell any security. The information here comes from publicly available sources believed to be reliable, including SEC filings, company press releases, earnings call transcripts and third-party market data, but accuracy and completeness aren’t guaranteed. All investing involves a high degree of risk and losses can be substantial. Do your own due diligence and consult qualified professionals before making any investment decision.

The author holds a long position in the common stock of Applied Aerospace & Defense, Inc. (NYSE: AADX) and may add to or reduce that position at any time without notice. Guardian Research has received no compensation from the company, from Greenbriar Equity Group, from any underwriter of the company’s initial public offering, or from any related party in connection with this report. Guardian Research and its authors aren’t registered investment advisors or broker/dealers and aren’t responsible for any investment decisions made by subscribers.

Certain figures here are Guardian Research calculations derived from cited primary sources rather than figures reported by the company. That includes the November 30, 2026 lock-up expiry date, enterprise value, net debt, incremental margins, revenue per square foot, all scenario and DCF outputs, and the reverse-DCF implied 2030 EBITDA. Several items aren’t disclosed by the company as of this writing and are identified as such in the text, including credit agreement financial covenants, the SOFR pricing grid, funded versus unfunded backlog, book-to-bill, customer concentration in the interim filing, and executive compensation. Market data is as of August 17, 2026. All AADX valuation math is struck on $18.11, the last quote available to me, against a last confirmed exchange close of $17.92 on August 14, 2026. Portfolio performance of 82.29% year to date is measured from December 31, 2025 through August 14, 2026.

Past performance doesn’t indicate future results.

Guardian Research, August 18, 2026