If you model a Bitcoin treasury company as a levered proxy for spot BTC, your model is wrong by roughly a factor of two — and it is wrong in both directions.
Here is the arithmetic that should stop you. Strategy's average acquisition cost sits at $75,441 per coin across 848,000 BTC, a cumulative outlay of roughly $64 billion. Strive's average is $84,422 across 29,462 BTC. Spot Bitcoin is trading near $86,000 — up nearly 7% over thirty days, still about 32% below its $126,000 all-time high. Two treasuries, two cost bases, one asset. And yet the return profile of neither equity is a direct function of Bitcoin's price. It is a function of a ratio that never appears on a Bitcoin chart.
That ratio is the market-to-NAV multiple. Everything else — the volatility warning, the management-execution warning, the amplification ratios executives quote in interviews — is downstream of it. When Andrew Webley, CEO of The Smarter Web Company, told Stephen Livera that investors "can't have performance and no volatility," he was describing the symptom. The disease is structural, and it lives in the spread between what a share is worth and what the company behind it holds.
I want to disassemble that spread, because the current market commentary treats it as a sentiment indicator when it is actually a mechanical transmission channel.
Context: What a Treasury Company Actually Is
Strip the branding and a Bitcoin treasury company is a shell with three moving parts: a pile of BTC, a capital structure, and a management team that decides how the first interacts with the second. The Bitcoin is the collateral. The capital structure is the engine. Management is the operator, and unlike a spot ETF — which is a passive, rules-based claim on the same collateral — a treasury company has discretion, and discretion is a liability as much as an asset.
The standard flywheel works like this. The company trades at a premium to its net asset value, meaning the market values each dollar of BTC held at more than a dollar. Call that multiple mNAV. When mNAV is above 1, the company can issue new shares at the premium, convert the proceeds into Bitcoin, and the BTC-per-share metric rises without any Bitcoin needing to be mined or bought by existing holders. Each issuance is accretive to the holders who did not sell. The premium funds its own persistence. This is the entire thesis, and it is elegant, and it is fragile in a way that is precisely quantifiable.
Strategy's scale makes the mechanism legible. A 334 BTC purchase for $28.7 million is a rounding error against 848,000 coins, but the accumulation cost basis — $75,441 — is the number that matters for solvency optics. Strive's 2,000 BTC purchase for $169 million at $84,422 is a different bet: a smaller book, a higher cost basis, and a much more aggressive posture toward the same flywheel. Strive CEO Matt Cole has argued his firm can out-return the incumbent through a higher amplification ratio — 51.4% versus roughly 25% — and he has attached a price target of $400,000 to $500,000 for Bitcoin by late 2029 to that claim.
I have spent enough time inside capital-structure models to know that when someone leads with an amplification ratio, the honest question is not "how high can it go" but "what does the ratio do when the premium closes." So let me build the model.
Core: The Math of Amplification, and Its Inverse
Define the variables cleanly. Let N be the number of shares outstanding, B the BTC held, and P the price of Bitcoin. Net asset value is B times P. Market capitalization is the share price times N. The multiple is market cap divided by NAV:
mNAV = (share price × N) / (B × P)
When mNAV exceeds 1, the accretive issuance condition holds. Suppose the company issues ΔN new shares at the market price and deploys the full proceeds into Bitcoin at price P. The new BTC per share is (B + proceeds/P) / (N + ΔN). Accretion occurs when that exceeds B/N. Solving the inequality yields a single, clean condition: accretion requires mNAV > 1. Not mNAV > 1.1, not mNAV > 1.5 — the boundary is exactly 1.0, and the magnitude of accretion scales linearly with (mNAV − 1).
This is where I stop trusting the headline numbers. The "amplification ratio" that Strive quotes is a BTC-yield metric — the change in BTC-per-share divided by the change in Bitcoin's price over the same window. It is a realized, backward-looking statistic. It tells you what the flywheel produced when the premium was open and the issuance window was unblocked. It tells you nothing about what happens when mNAV crosses below 1, because the same equation that makes issuance accretive above 1 makes it dilutive below it. The amplification is symmetric. Cole's 51.4% is a statement about the past tense of a reflexive loop.
Here is the part that reads like a vulnerability to me. I first learned this shape of bug in 2017, reverse-engineering the 0x Protocol v1 order-signing logic, where an integer overflow in a boundary condition could drain a liquidity pool during high-frequency trading. The lesson was not "overflow is bad." The lesson was that systems fail at their edges, and the edge is always the arithmetic boundary where a sign flips. For a treasury company, that boundary is mNAV = 1. Above it, issuance is a money printer for shareholders. Below it, the identical operation is a wealth transfer away from them. Same code path, opposite outcome. Logic prevails, but bias hides in the edge cases — and the mNAV boundary is the edge case that every treasury-company pitch deck quietly skips.
The second transmission channel is the convertible bond. Many of these treasuries fund BTC purchases not with common equity but with zero-coupon convertibles — debt that pays no interest, converts to equity at a strike above the current price, and is bought primarily by volatility arbitrage funds that delta-hedge the position. This is not a bug in the structure; it is the deliberate sale of volatility. The company is short its own implied volatility and long Bitcoin. As long as realized volatility stays high and the equity stays above the conversion strike, the arbitrage funds stay hedged, the debt stays cheap, and the flywheel keeps turning.
But a short-volatility position has a familiar payoff shape, and I watched it detonate during the 2020 DeFi Summer when I modeled Uniswap V2's constant product invariant, x × y = k. The AMM was celebrated for its theoretical efficiency. What my report actually quantified was the slippage cost borne by large traders in thin pools — the gap between the clean formula and the messy execution. Convertible-funded treasuries have the same gap. The clean model says cheap debt and accretive equity. The messy reality is that when Bitcoin gaps down, the delta-hedges unwind in the same direction as the price move, the equity falls faster than the NAV, mNAV compresses, and the issuance window that funded everything slams shut.
Speed is an illusion if the exit door is locked. A treasury company can buy Bitcoin in seconds. It cannot un-buy the capital structure that financed the purchase. The convertibles do not disappear; they convert, or they get refinanced at a worse rate, or they force dilution exactly when dilution is most expensive to shareholders.
Now layer on the volatility Webley flagged. A spot Bitcoin holder experiences Bitcoin's drawdown directly. A treasury-company shareholder experiences it twice: once through the NAV, and again through the multiple. If Bitcoin falls 30% and mNAV compresses from 2.0 to 1.2 in the same window, the equity does not fall 30% — it falls roughly 58%, because (1.2/2.0) × (0.7) = 0.42. That is the amplification, working in the direction investors do not cheer. The executives who celebrate a 51.4% amplification ratio on the way up are, by the same arithmetic, running a book that delivers a comparable multiple on the way down. There is no asymmetric version of this structure. The lever does not care which way the load moves.
The management-execution risk that Webley described — "Bitcoin has no management; a Bitcoin treasury company, the management could really, really mess it up" — is not a soft governance concern. It is a hard, quantifiable variable in the model. Every capital-structure decision is a bet on the future path of mNAV. Issuing too aggressively at a thin premium bakes in dilution. Issuing too little lets the premium decay unused. Taking on convertibles with a strike set too low converts cheap debt into expensive equity. Buying at the top of a cycle raises the cost basis — Strategy's $75,441 and Strive's $84,422 are not neutral facts; they are the levels below which the narrative of "we are winning" becomes arithmetically awkward to defend.
And here the comparison to the broader crypto economy sharpens. Liquidity mining programs subsidize a number — TVL — that collapses the moment the subsidy stops, because the users were never the product; the emission was. Treasury companies subsidize a different number — BTC-per-share — through the premium, and the premium is a subsidy paid by new shareholders to old ones. Stop the issuance and the accretion stops. The metric that justified the entire structure goes to zero. This is the same reflexivity wearing a corporate charter instead of a farm contract, and it should be read with the same suspicion.
The current market context makes this concrete. Bitcoin was rejected at $87,000 on Monday after a weak US jobs report had pushed it above that level on Friday — its first time above it in ten days — and it fell back below $84,000 within hours, triggering nearly $600 million in liquidations. That is a 3.5% intraday range producing a nine-figure cascade. For a spot holder, it is noise. For a treasury company trading at a premium, it is a live stress test of the multiple, and the multiple is where the leverage lives. The jobs report did not change any company's BTC holdings by a single satoshi. It changed the market's willingness to pay above NAV, and that is the only input that matters to the flywheel.
Contrarian: The Blind Spot Is Not Bitcoin's Drawdown
Everyone is watching the wrong variable. The consensus risk narrative for treasury companies is "Bitcoin could crash and the equity would crash harder." That is true, and it is also the least interesting version of the risk, because it is symmetric and visible and therefore priced.
The under-priced risk is a regime change in mNAV that persists independently of Bitcoin's price. A premium is a belief about future accretion. If the market decides the flywheel is slowing — because the cost basis is too high, because the convertibles are maturing, because the amplification ratio stops printing — the multiple can compress while Bitcoin is flat. In that scenario, spot holders break even and treasury shareholders lose, not because Bitcoin failed, but because the wrapper failed. The ledger is immutable; the balance sheet is not. Bitcoin's monetary policy is enforced by code. A treasury company's policy is enforced by a board, and boards change their minds.
There is a second blind spot: the passive-flow reflexivity. When these equities get added to major indices, index funds buy them mechanically, which lifts the price, which lifts the premium, which makes issuance accretive, which buys more Bitcoin, which justifies the index weight. This loop is not driven by any fundamental view of Bitcoin's utility. It is a flow-driven feedback that runs until the flows reverse. I saw the same structure in my 2022 audit of Arbitrum's fraud-proof mechanism, where I modeled how validator collusion could delay finality indefinitely despite high throughput claims. The lesson generalized: a system can look healthy at the throughput layer while the trust layer quietly degrades. A treasury company can look healthy at the NAV layer while the mNAV layer quietly inverts.
And I will say the quiet part about the target prices. A $400,000 to $500,000 Bitcoin by late 2029 is a forecast that, if correct, makes almost every treasury company look brilliant and every model of its downside irrelevant. That is exactly why it is an unreliable input. Forecasts that flatter the structure are the ones most likely to be repeated by the people running the structure. I have no objection to the number. I object to it being used as a substitute for the mNAV math, because the math does not care whether Bitcoin reaches $500,000. It cares whether the premium is open when the company needs to issue.
Takeaway
The next twelve months will not test these companies on Bitcoin's price. They will test them on the multiple. Watch mNAV, watch the cost basis relative to spot, watch the convertible maturity wall, and watch whether the amplification ratio is still being quoted in the present tense or quietly shifted to the past. The tell is not in the BTC line of the balance sheet. It is in the spread between two numbers that management controls the framing of and shareholders control the pricing of. When that spread inverts, the exit door locks from the inside — and the question every holder will have to answer is not how much Bitcoin the company owns, but how much of it they actually own after the financing unwinds.
