Strive's Bitcoin Treasury Is a Feedback Loop, Not a Business

The most informative line in the TD Cowen note was not the price target. It was the phrase about warrant pressure easing. A 27% target increase reads like conviction. It isn't. It's an admission that a supply overhang is clearing — that the stock now trades above the strike prices embedded in its own capital structure. That is a statement about float mechanics, not about Bitcoin, and not about Strive's ability to compound BTC per share. Warrants are not a routine financing artifact. They are a fossil. They appear in SPAC mergers and PIPE transactions, and they persist on the balance sheet long after the promotional deck that justified them has been deleted. When an analyst flags warrant pressure, he is telling you the company's float has been structurally capped by paper issued to early backers — and that the cap is lifting. Everyone read the headline as bullish on BTC. The actual information was about a dilution schedule. Strive is a Bitcoin treasury company. The label is doing a lot of work. There is no protocol, no smart contract, no token, no chain. The code here is a capital structure, and it executes on a quarterly clock rather than a block clock. The archetype is MicroStrategy. The mechanism, stripped of narrative: issue equity above net asset value per share, use proceeds to buy Bitcoin, watch BTC per share rise, watch the stock re-rate on the higher per-share backing, issue again. Repeat. The premium is not a valuation of the company. It is an input to the machine. Remove it and the machine has no reason to exist. Strive is a late entrant. The source material is four data points — aggressive buying, a planned raise, an analyst target hike, warrants clearing. That is thin. I have spent enough time on thin disclosures to know what they usually conceal: a company whose entire operating history is a sequence of financing events. So let's analyze it the way I'd analyze a contract with no audit — find the invariant, then test what breaks it. The invariant is an identity, and it is simple: Market cap = BTC holdings × mNAV mNAV is the modified net asset value multiple — market cap divided by the market value of the coin on the balance sheet. Above 1.0 is a premium. Below 1.0 is a discount. Every treasury company lives or dies on this single variable, and unlike a hash rate or a TVL figure, it is not measured. It is voted on. Write the flywheel as pseudocode and the structure becomes obvious: while (mNAV > 1.0) { issue_equity(at premium); buy(BTC); btc_per_share++; } That loop is a positive feedback circuit with no internal governor. No circuit breaker, no cooldown, no parameter a developer can flip when conditions invert. The only exit condition is the one nobody schedules: mNAV crossing below 1.0. When that happens, the loop doesn't slow. It reverses. Issuing equity below NAV to buy coin is accretive to the seller and destructive to the existing holder — you are selling a dollar of backing for ninety cents and calling it growth. A rational board stops issuing. But a board that stops issuing stops growing BTC per share, and the stock was priced on BTC per share growth. So the equity falls, mNAV falls further, and the thing that funded the buying becomes the thing that forbids it. The code doesn't break. It inverts. I watched this exact architecture fail in 2022. When I reverse-engineered TerraUSD's seigniorage logic, the finding was not that the math was wrong. The math was internally consistent right up until the moment the loop flipped. What was missing was a brake. No mechanism could slow the mint-and-burn engine once peg pressure exceeded the model's assumptions, because the mechanism itself was the source of the pressure. Reflexivity is not a bug in these designs. It is the design. Strive's loop runs on a slower clock than Terra's did — quarters instead of blocks. Slower loops feel safer. They aren't. They give participants more time to construct a rationale for something they would have recognized instantly at 3 a.m. in a mempool. The accounting makes this worse. Under FASB ASU 2023-08, crypto holdings are marked at fair value through income. Strive's income statement becomes, functionally, a chart of the Bitcoin price with a corporate wrapper. Earnings are noise. Revenue, if any exists, is a rounding error against mark-to-market swings. Two numbers matter: BTC per share and mNAV. Only one of them arrives on a predictable schedule. Which brings us back to the warrants. Exercise brings cash in and shares out. The net effect on existing holders depends entirely on where the strike sits relative to the market. TD Cowen calling the pressure eased is a polite way of saying the market is now above the strike — that exercises are accretive rather than dilutive. That is a real, technical, short-horizon positive. It is also an admission that the company's float was, until recently, suppressed by its own history. Here is the part the sector doesn't want to price: there are now dozens of these vehicles. Metaplanet, Semler, and a lengthening list of small caps that discovered Bitcoin as a refinancing strategy. They are not competing for a growing pool of buyers. They are slicing the same scarce institutional allocation into smaller and smaller fragments, each claiming a premium on the same underlying coin. The premium is a function of scarcity of access. Access is no longer scarce. Now the part the bears get wrong, including the version of me that would have shorted this in 2021. The premium is not automatically irrational. A company that can genuinely issue equity at two times the value of its assets and convert proceeds into assets at one times is performing a real arbitrage on its own paper. That is value transfer from new shareholders to old ones, and it is legal, disclosed, and mathematically sound. MicroStrategy did it for years and the accretion was real. This is the one thing a spot ETF structurally cannot do — an ETF cannot issue itself at a premium, because creation is arbitraged away within a day. A closed-end treasury company can hold a premium for months. So the bull case is not stupid. It rests on two assumptions, and only one of them is about Bitcoin. The first is that BTC rises. The second is that management issues paper only when accretive and stops when it isn't. That second assumption is a governance claim, not a mathematical one. It depends on a board resisting the easiest growth lever available to it, in a market that rewards size and headlines. I have audited enough withdrawal logic to know how that usually resolves. They built on sand; I built on skepticism. There is one number to track on Strive, and it isn't the target price or the buying spree. It's mNAV. Above 1.0, the loop compounds and the analysts look prescient. Below 1.0, the same loop eats its own shareholders and nobody sends a note. Cold logic cuts through the noise of FOMO, but it doesn't predict the timing of a sentiment reversal — only the consequences. The flywheel will not announce its failure. It will just stop, and the prior quarter's buying will be reclassified, retroactively, as the top. Watch the premium. Everything else is a press release.