Strive's $700 Million SATS IPO: Auditing the Bitcoin Treasury Shell

On the morning the book priced at $90 a share, one number did most of the work: $700 million, upsized from an earlier, smaller target. In Bitcoin terms that is a rounding error — the major spot venues clear multiples of it before the European session closes — but inside the narrow corridor where a company can still sell primary equity to public investors and wire the proceeds into a single volatile asset, it is a large number. The size is not the story. The structure is.

Strive's SATS is not a protocol. It is not a token, not a chain upgrade, not a governance experiment. There is no sequencer to audit, no contract to review, no supply schedule to model. It is a ticker on a public exchange wrapped around a corporate balance sheet, and the balance sheet is designed to hold bitcoin. Every framework I normally apply — contract-level review, unlock calendars, governance attack surface, liquidity depth — has to be re-mapped onto corporate finance instruments: a registration statement, an underwriting syndicate, a lock-up schedule, an at-the-market facility, a fair-value accounting election.

That re-mapping is precisely where crypto-native readers lose the thread. They ask whether SATS is bullish for Bitcoin. That is the wrong question, and it produces the wrong answer in both directions. The right question is whether the equity capital markets window that makes this structure accretive stays open long enough for the arbitrage to compound. A treasury company is not a Bitcoin position with a stock wrapper; it is a capital-raising machine whose fuel happens to be Bitcoin. We do not predict the wave; we engineer the hull.

Context: how a software balance sheet became an asset class

The template was set in August 2020, and it was not invented by crypto people. A mid-cap software company with flat revenue discovered that it could issue convertible debt at a coupon near zero, because the buyers of that debt were not credit investors — they were volatility buyers paying for the embedded call option. The proceeds went into bitcoin. The equity then re-rated to a premium over the underlying coin, which created a second, self-feeding engine: sell shares above net asset value, buy more coin, lift the coin-per-share figure, feed the narrative, repeat.

That two-engine design is the whole model. The first engine is cheap liability-side capital — converts, preferred, low-coupon debt. The second engine is equity issuance at a premium. Neither engine has anything to do with the operating business, which is why the operating business is always described in the prospectus as a legacy segment.

For four years the model was effectively a monopoly. One balance sheet absorbed the demand of every public-market allocator who wanted bitcoin exposure inside a brokerage account but could not, or would not, open a crypto venue. That monopoly ended in January 2024. The spot ETF approval converted bitcoin exposure into a commodity product priced at roughly 19 to 25 basis points a year, with no premium, no key-person risk, no lock-up, no accounting election, and intraday liquidity. From that moment, the treasury company premium stopped being a default and started being a claim that had to be defended.

By 2025 the supply side had exploded. The Japanese entrant, the medical-device company, the small-cap shells, the dozens of listed vehicles across three continents — every one of them sells the same product to the same marginal buyer. The category went from one name to a crowded sector in under eighteen months. That is the environment Strive is pricing into.

And the disclosure set here is thin. What we have is a headline: an upsized $700 million offering at $90 per share, described as bitcoin-backed securities, framed as a corporate treasury strategy. We do not have the post-IPO share count, the use-of-proceeds language, the underwriter list, the lock-up schedule, or the historical financials. Based on my audit experience, I treat any structural conclusion drawn from five data points — three of them without independent sourcing — as a hypothesis, not a finding. So the honest deliverable here is not a verdict on the trade. It is the audit framework you would run the moment the registration statement lands.

Core: the arithmetic of the shell

Start with the only hard numbers available. A $700 million primary raise at $90 per share implies roughly 7.78 million new shares. That is the numerator of nothing, because the denominator — total shares outstanding after the offering — is undisclosed. Without it, market capitalisation is uncomputable, and without market capitalisation there is no mNAV. The single most important valuation variable in this entire structure is missing from the disclosure set. That is not a criticism of the reporting; it is a statement about how early we are.

What mNAV is, and why it decides everything: divide the equity's market value by the market value of the bitcoin the company holds, net of debt. A reading of 1.0 means you pay a dollar for a dollar of coin. A reading of 2.0 means you pay two dollars for a dollar of coin, and you are therefore paying a dollar for the expectation that management can keep issuing equity above that level.

The accretion math is simple and unforgiving. If a company trades at 2.0x mNAV and issues $100 million of stock, it receives $100 million of cash for shares that the market values at $200 million of implied asset backing. It buys $100 million of bitcoin. Existing holders now own a marginally larger slice of a marginally larger pile — coin per share rises. That is genuine value transfer from new shareholders to old ones, executed at the new shareholders' consent. It is not fraud. It is the same mechanic as a company issuing stock at a high multiple to buy a low-multiple business, except the acquired business here is a bearer asset with no cash flow.

Run it in reverse and the elegance disappears. At 1.2x mNAV, the same $100 million issuance moves coin per share by a fraction of the amount, the narrative support weakens, and the market begins to ask why anyone would pay 20% over spot for a managed pile of coin when the ETF charges 20 basis points. At 0.9x, issuance is dilutive and the engine seizes. The treasury company is a machine that only works while its own equity is expensive. That is the entire structural insight, and everything else in this article follows from it.

The flywheel is a leverage contract, not a business

Strip the branding and what remains is a single-line item: bitcoin per share, multiplied by spot, multiplied by a sentiment coefficient. There is no operating cash flow, no pricing power, no switching cost, no technical moat. If the prospectus confirms a pure treasury mandate, then reported revenue is effectively interest on cash and nothing else. The value proposition sold to allocators is not a business — it is a levered, optionally-financed claim on a bearer asset, packaged in a security their compliance department will actually approve.

That is a legitimate product. It is also a product with a known failure mode, and I have audited the failure mode. When the algorithmic peg on the largest stablecoin in the 2020-2022 cycle began to wobble, the desk I ran exited 48 hours before the break, not because we were smarter than the market, but because our stress model asked a question nobody else was asking: what happens to the collateral chain if the reflexive loop runs in reverse? The answer was that the loop had no floor. Reflexivity is symmetric. The mechanism that converts rising prices into rising coin-per-share converts falling prices into falling coin-per-share, and it does so with the same mechanical indifference.

We do not predict the wave; we engineer the hull. For SATS, the hull is the ratio between equity price and coin backing. Watch that ratio, not the price of bitcoin.

The decay function nobody models

Here is the part the pitch deck omits. Premium compression is a return event even when the underlying asset is flat. Model a simple case: bitcoin unchanged over twelve months, mNAV drifts from 2.0 to 1.3 as the sector adds supply and as ETF substitutes absorb the marginal allocator. The equity holder loses 35% while the coin holder loses nothing. Add a 25% drawdown in spot and the equity's decline compounds to something in the range of 50-55%. That is the double beta: exposure to bitcoin plus exposure to the market's willingness to pay a premium for bitcoin.

There is no hedge for the second component. Options on the equity price the first. Nothing prices the second except the equity itself. This is why I insist on separating the two in any position-sizing exercise: a treasury company should be risk-budgeted as bitcoin beta plus a short position in sector sentiment, not as a bitcoin proxy with a management fee.

Where the money actually lands

Follow the fees. The underwriters take a gross spread on $700 million — on a deal of this size, plausibly in the low-to-mid single-digit percentage range, which is real money before a single coin is bought. The custodian takes an annual basis-point fee on the coin, and that fee scales with the treasury, not with the equity price. The exchanges take listing fees. The lawyers and auditors take the rest.

The most certain cash flows in a bitcoin treasury IPO belong to parties who never touch bitcoin price risk. That is not cynicism; it is a structural observation about where the alpha sits in the value chain, and it should inform how you think about the trade's risk-adjusted return versus the risk-adjusted return of simply owning the custodian's equity.

Custody concentration deserves its own line item. Institutional bitcoin treasuries are held through a very small number of qualified custodians. In 2022 I led the forensic work on a nine-figure integration failure and produced a fifty-page report on cascading algorithmic-stablecoin failure that three regulators later cited; the lesson that survived that exercise was not about code. It was about dependency graphs. A treasury company's dependency graph is brutally simple: one asset, one or two custodians, one capital markets window. Three nodes. Any one of them failing takes the whole structure with it, and none of them is under management's control.

The accounting tail

The 2025 fair-value accounting regime for digital assets changed the presentability of this entire sector. Under the older impairment model, a company marked coin down in a drawdown and never marked it back up until sale, which produced balance sheets that understated reality and earnings that looked like a slow bleed. Fair value through net income removes that distortion. Coin marks now flow directly into reported earnings each quarter.

Two consequences follow, and both are underappreciated. First, the earnings series becomes a mirror of bitcoin's price path, which means quarterly EPS becomes a volatility readout rather than a performance measure. Second, mechanical screens — value filters, earnings-stability filters, quality factors — will systematically misclassify these companies in both directions, and index-driven flows will follow those misclassifications. Anyone modelling flows into this sector on the assumption that fundamental screens behave normally is modelling the wrong thing.

There is a tax tail as well. Unrealized marks are an accounting event, not a cash event, and the interaction between fair-value gains, corporate minimum tax provisions, and buyback excise rules is exactly the kind of detail that turns a clean capital-markets story into a messy quarterly surprise. Read the tax footnote before you read the press release.

The lock-up calendar is the only calendar that matters

Founder and early-investor shares in an IPO of this type typically carry a six-to-twelve month lock-up. That creates a known, dated supply event. For a company whose float is small relative to its total share count, the expiry is not a footnote — it is the dominant liquidity variable of the first year.

The float also determines index eligibility, and index inclusion determines passive demand, which in turn supports the premium that makes the flywheel turn. This is a circular dependency that works beautifully in one direction. In the other direction, a compressed float with a dated unlock and a stalling issuance engine is a textbook setup for a disorderly repricing.

The competitive map, honestly drawn

The category leader holds well over 400,000 bitcoin and has spent five years building financing relationships that a new entrant cannot replicate in a single quarter. The Japanese entrant owns the Asian retail channel. The mid-cap medical and software names own nothing except a ticker and a story. And then there are dozens of 2025-vintage vehicles, most of which will be indistinguishable from each other within eighteen months.

Strive's differentiation is not technical and cannot be. What it plausibly has is brand distribution and political capital — an asset manager built explicitly around shareholder-pressure and anti-ESG positioning, now extending that franchise into a listed treasury vehicle. In capital formation, brand is a real asset: it lowers the cost of attracting the first $700 million. It is also non-transferable, non-amortizable, and entirely dependent on the continued salience of the founder's public profile. That is a moat made of attention, and attention has a half-life.

Contrarian: the correlation is conditional, not structural

The consensus framing treats these vehicles as high-beta bitcoin. The data supports that in rising markets, which is exactly when nobody checks the assumption. My contrarian read is different: a treasury company's correlation to bitcoin is conditional on the equity capital markets window being open, and that conditionality is the real risk factor.

The underlying exposure is not bitcoin. It is a carry trade on the primary issuance window — long the ability to sell equity above net asset value, short the volatility that closes that window. When volatility spikes and risk appetite contracts, two things happen simultaneously: bitcoin falls, and the accretion engine stalls precisely when management would most want to issue. The two hedges you would naturally want cancel each other out. That is not beta. That is a short-volatility position with a bitcoin-shaped surface.

The marginal buyer reinforces this. The buyer of SATS shares on day thirty is not a bitcoin believer who read the whitepaper; that person bought the ETF. The marginal buyer is an equity flow investor — a momentum allocator, a small-cap generalist, a thematic fund that needs a ticker rather than a wallet. These buyers do not price coin. They price relative strength, float, and narrative momentum. When their flow reverses, it reverses on equity-market signals, not on on-chain ones. Which means the correct monitoring dashboard for this position has more in common with a small-cap equity risk desk than with a crypto desk.

One more observation, and it cuts against the sector's self-image. Capital allocation across the digital asset stack is currently mispriced in a very specific way: balance-sheet shells can raise $700 million in a week on a five-paragraph disclosure, while the operators actually building infrastructure are bleeding. Rollup operators in particular are running proving costs that only pencil out if gas returns to bull-market levels — the unit economics of a proving cluster at current fee levels are ugly, and everyone in the space knows it and nobody funds the fix. Capital is flowing to the wrapper and starving the engine. That is a classic late-cycle allocation signature, and it is the strongest argument for treating this IPO class as a positioning tool rather than a long-term compounding vehicle.

We do not predict the wave; we engineer the hull. The hull here is a spreadsheet with four columns: coin per share, mNAV, float, and the lock-up date. Everything else is commentary.

Takeaway: what to watch, in order

Three signals will resolve this thesis long before the price does. First, the use-of-proceeds language in the registration statement: whether it commits capital to coin acquisition or leaves discretion for acquisitions, buybacks, or working capital. Second, the first disclosed coin-per-share figure, which converts mNAV from a concept into a number you can track weekly against the sector leader's premium. Third, the at-the-market facility, if one is filed — because an ATM is the flywheel's throttle, and its existence tells you whether management intends to compound coin per share or merely to hold a static pile.

Everything else — the anti-ESG branding, the political adjacency, the headline size — is distribution, not structure. Distribution gets a deal priced. Structure determines whether the equity is worth holding twelve months after the lock-up expires.

So here is the question I would put to any allocator being shown this deal in a sideways tape: when the marginal dollar of institutional bitcoin exposure can buy a 20-basis-point ETF with intraday liquidity and no key-person risk, what exactly is the 1.8x mNAV premium paying for — and who is on the other side of that trade when the window closes?