The mNAV Stress Test: Which Bitcoin Treasury Companies Survive a Two-Year Winter

Strategy sold 32 Bitcoin and the internet declared a funeral. We read the filings instead. The senior claims on the world's largest Bitcoin treasury are covered down to roughly $25,700 per coin, and the companies actually near the edge are not the ones trending. The solvency math, company by company, with the stress test live on this page.

Decentralised News · Updated June 12, 2026 · Original solvency model · Reading time 16 min · Tool included

Which Bitcoin treasury companies actually survive a two-year winter? The filings give a clearer answer than the discourse. Strategy, the company at the centre of the June 2026 panic, holds 845,256 BTC against senior claims of roughly $21.7 billion, meaning Bitcoin would need to fall below about $25,700 before its coins stopped covering every dollar of debt and preferred stock. Meanwhile companies with one-tenth the attention carry cost bases 40 percent underwater and trade below the value of their own coins. The market is stress-testing the wrong names with the wrong math.

So we built the model the discourse is missing. The DN Treasury Stress Index scores any digital asset treasury company from 0 to 100 on four published dimensions, asset coverage, obligation runway, cost-basis stress, and dilution dependency, under any Bitcoin price path you choose. Every company input is a datestamped board value from filings and reports, every formula is printed on this page, and the live Bitcoin price seeds the scenario. This is a Receipts piece: when the next treasury scare hits, and it will, the number it needs is already here.

What the panic got right, and what it got wrong

The June 2026 narrative had three exhibits: Strategy's first Bitcoin sale since 2022, viral theses comparing the company to Terra Luna, and a stock down nearly 60 percent in a year while short sellers circled, with CNBC reporting a coordinated short campaign against MSTR on June 5. Each exhibit deserves an audit.

Right. The structural anxiety is legitimate. Strategy's own Form 8-K of June 1, 2026 confirms it sold 32 BTC for $2.5 million between May 26 and 31, at an average $77,135, with proceeds earmarked for preferred dividends, the first sale since 2022 and a genuine symbolic break with "never sell." The dividend machine is real and large: the Q1 FY2026 release puts preferred equity outstanding above $13.5 billion in notional value, and at the blended coupon implied by the company's own disclosures, roughly 10 to 11 percent across STRK at 8 percent, STRF and STRD at 10 percent, and the variable-rate STRC, that is approximately $1.4 billion a year, about $117 million a month, in perpetuity, in cash. The common stock's collapse from a 52-week high of $457 to the $115 to $121 range is also real, and it matters mechanically, because at-the-market equity sales are the engine that funds everything, and the June 8 filing shows the company still leaning on that engine, raising $181 million of common in the first week of June while buying 1,550 BTC at an average $65,332.

Wrong. The scale narrative fails arithmetic. A $2.5 million coin sale against a $117 million monthly obligation is 2 percent of one month's bill; the other 98 percent came from the $900 million USD reserve the company established in December 2025 and from securities issuance. The Terra Luna comparison fails structurally: Luna's collapse was a reflexive death spiral in which the asset backing the system was the system's own printed token, while Strategy's obligations are backed by Bitcoin it does not issue, and its preferred stock is perpetual, no principal ever comes due, with the nearest convertible debt maturity in September 2028. And the insolvency framing fails on coverage: at $62,000, the coin stack is worth roughly $52.4 billion against $21.7 billion of total senior claims, 2.4 times coverage, with $692.5 million in cumulative preferred dividends already paid across 23 consecutive distributions per the Q1 release. Whatever this is, it is not Luna.

Missed entirely. The forced-sale scenario the doomers fear for Strategy already happened, at a different company, and almost nobody scored it. MARA Holdings sold roughly 15,000 BTC for about $1.1 billion in March 2026 to repurchase a portion of its convertible debt at a discount, cutting its stack to roughly 38,700 coins. That is the contagion event, a top-five corporate holder liquidating nearly a third of its treasury into a falling market, and the market absorbed it without a cascade. The precedent cuts both ways: forced sales are survivable, and they are also no longer hypothetical.

The receipts: five companies, three kinds of death

Five findings, each one a correction to the discourse.

1. The claims break price is the number nobody quotes. Divide total senior claims by coins held and you get the Bitcoin price at which the treasury merely equals what is owed ahead of common shareholders. For Strategy that is $21.7 billion over 845,256 coins: $25,672. Bitcoin at $62,000 sits 59 percent above it. Every bankruptcy thesis that does not engage with this number is a vibe with a thumbnail.

2. The dividend machine runs about 37 years on coin sales alone. The ugliest honest scenario, equity markets closed, reserve exhausted, every dividend funded by selling Bitcoin at $62,000, costs roughly 1,890 coins a month, 0.22 percent of the stack. Run it forward and the treasury funds its preferred obligations for over three decades before exhaustion, ignoring price changes in both directions. Death two is far away. But note what that scenario does: it converts the equity story from "Bitcoin per share always rises" to a slow bleed, which is death one operating in plain sight even while solvency is never in question. The three deaths are different, and this is why conflating them produces both the doomer error and the defender error.

3. The most stressed balance sheet is the least discussed. On these inputs MARA, not Strategy, carries the tightest coverage, roughly 1.2 times at $62,000 after its March sale, with a claims break price in the low $50,000s, uncomfortably close to spot. The mitigant the model deliberately excludes is that MARA mines revenue every day, which a pure treasury company cannot. The honest statement: among pure treasuries, nobody profiled here is near death two; the one name near its break price has an operating business the model does not credit.

4. Cost basis is the silent differentiator. Metaplanet's $104,106 average cost, per its own Q1 2026 report of 40,177 BTC acquired for $4.18 billion, is the highest in the cohort, leaving it roughly 40 percent underwater at $62,000, versus Strategy's 18 percent. Its fixed obligations are tiny, so it cannot die death two, but a high cost basis is what manufactures death one: the deeper underwater the treasury, the harder the equity premium falls, the more punitive every yen of issuance, and Metaplanet already touched mNAV 0.99 in October 2025, the first major treasury to trade below its coins. Survival is not the same as a functioning flywheel.

5. The sector's premium era is over, and that is the actual story. K33 Research counted 26 of 168 listed Bitcoin holders trading below the value of their coins in late 2025, with the sector's average mNAV compressing from 3.76 toward parity since, and the most extreme case, the NAKA vehicle, falling 96 percent from peak to roughly 0.7 times its coins. The treasury model's first act, issue expensive paper, buy cheap coins, repeat, required a premium that no longer exists at most firms. The second act, surviving as a discounted closed-end Bitcoin fund with obligations, is what this stress test measures.

Run any company through any winter

The instrument below seeds the scenario with the live Bitcoin price, lets you set a trough price and a winter duration, and walks the selected company's board inputs through it: coverage at the trough, runway against obligations, the recomputed mNAV with an editable equity-sensitivity assumption, and the survival classification with the binding constraint named. Every input is editable, so when filings update, or when you disagree with an estimate, the model updates with you.

Three winters, modeled honestly

The base winter: $55,000 to $72,000 chop for two years. Strategy's reserve plus open ATM carries the dividend stack without drama; the binding constraint is death one, not death two, as continued issuance below the old highs grinds Bitcoin-per-share accretion toward zero. Metaplanet's flywheel stays jammed by its cost basis, Twenty One's low claims keep it inert and safe, and MARA's coverage stays thin enough that a second debt-driven coin sale is a live possibility. Sector mNAV compression continues; survival is general, shareholder pain is selective. This is the same base case our DN Short Risk Score assigns 45 percent probability at the market level.

The deep winter: $40,000 trough, eighteen months. Strategy's coverage falls to roughly 1.56 times, still solvent by a wide margin, but the model's stress reading climbs into the strained band as the reserve alone covers under a third of the duration and equity issuance turns brutally dilutive. The honest mechanism to watch is the preferred stack itself: those instruments yield 10 percent-plus precisely because the market prices this scenario, and a sustained deep winter would reprice them harder, raising the cost of every future dollar of digital credit. MARA's modeled coverage breaks below 1.0 in this scenario, making it the cohort's contagion candidate, with the offsetting mining revenue the model excludes as its real-world lifeline.

The breaking winter: $25,000, the thesis killer. At Strategy's claims break price, every treasury in the cohort except the near-zero-debt names is in coverage distress simultaneously, and forced sales become sector-wide. The model's purpose is not to predict this, our cycle work puts terminal outcomes like this in the tail, but to locate it: the number is $25,700, not $60,000, and the 36,000-strong army of posts implying otherwise have been arguing with a price 59 percent below the market.

For readers whose conclusion from all this is that they want the asset without the capital structure, the spot-versus-wrapper decision, the comparison work lives in our DN Exchange Fit Engine, and accumulating spot directly on a derivatives-capable venue such as Bybit or OKX removes both the premium question and the dividend question from your personal balance sheet entirely, with self-custody via a hardware wallet completing what a treasury company structurally cannot offer: coins with no claims ahead of you. For readers whose conclusion is the opposite trade, shorting the weakest names into a panic that this article argues is mispriced, run the setup through the DN Short Risk Score first; crowded shorts on solvent companies are how squeezes are born, and the June short campaign against MSTR is exactly that configuration.

What would change our mind

• A missed or deferred preferred distribution at Strategy. Twenty-three consecutive distributions is the track record; the first break ends the benefit of the doubt and moves the company's reading two bands regardless of coverage.
• The USD reserve falling below three months of obligations without a corresponding raise, which would signal the issuance engine is no longer willingly fed.
• Bitcoin sustaining below roughly $31,000, the level at which Strategy's claims coverage drops through 1.2 times and the fortress framing in this article stops being arithmetic and starts being hope.
• Coin sales exceeding 1 percent of any profiled treasury in a quarter for obligation funding rather than opportunistic debt retirement, the MARA precedent crossing from balance-sheet management into death three proper.
• A second major treasury joining NAKA below 0.75 times mNAV for a sustained period, which would mark the discount regime as structural rather than cyclical and invalidate the model's dilution-recovery assumption.

None of these conditions is met as of June 12, 2026. Each is checkable in the tool above by editing the relevant board input, which is the point of publishing the model instead of the opinion.

The honest bottom line

The treasury-company question everyone is asking has a vibes answer, Saylor sold, therefore it is over, and a filings answer, the senior claims on 845,256 Bitcoin break at $25,700 and the dividend reserve covers most of a year before a single coin or share needs to move, and the distance between those answers is where both the panic and the complacency live. The real findings are less tweetable: the premium era that made these companies money machines is over and is not coming back at current discounts; the difference between shareholder pain and insolvency is the entire analysis, and almost nobody is making it; the one cohort member that already faced the forced-sale scenario survived it; and the balance sheets closest to their break prices are not the ones with the viral threads. The stress test is above. Set your own winter. The model does not care whose side you are on, which is exactly what makes it worth citing when the next scare arrives.

Frequently asked questions

The filings say no. As of June 2026, Strategy's 845,256 BTC cover its $21.7 billion of debt and preferred claims 2.4 times at $62,000 Bitcoin, its $900 million reserve covers seven to eight months of fixed obligations, its nearest debt maturity is September 2028, and it has paid 23 consecutive preferred distributions. Insolvency math does not begin until Bitcoin approaches roughly $25,700.

Per its June 1, 2026 SEC filing, Strategy sold 32 BTC for $2.5 million at an average $77,135 to help fund preferred stock dividends, its first sale since 2022. Against a roughly $117 million monthly obligation, the sale covered about 2 percent of one month, making it symbolically significant but financially trivial; the same week the company bought 1,550 BTC.

mNAV is enterprise value, market capitalization plus debt plus preferred notional, divided by the value of the company's Bitcoin. Above 1, issuing shares to buy coins adds Bitcoin per share; below 1, issuance dilutes. It is the metric that determines whether the treasury flywheel functions, which is why sector-wide compression from an average of 3.76 toward parity is the defining event of this cycle.

The Bitcoin price at which a company's coin holdings exactly equal its debt plus preferred claims: total senior claims divided by coins held. For Strategy that is roughly $25,700 as of June 2026. Above it, every senior dollar is asset-covered; below it, the capital structure is underwater. It is the single most clarifying number in any treasury solvency debate.

On pure balance-sheet inputs, MARA shows the thinnest modeled coverage, roughly 1.2 times at $62,000 after selling about 15,000 BTC in March 2026 to retire convertible debt, though its mining revenue, which the model excludes, is a real offset. Among pure treasuries, Metaplanet carries the deepest cost-basis stress at about 40 percent underwater, a flywheel problem rather than a solvency problem.

Structurally, no. Luna collapsed because the asset backing its system was its own printed token in a reflexive loop. Strategy's obligations are backed by Bitcoin it does not issue, its preferred stock is perpetual with no principal repayment, and its coins covered all senior claims 2.4 times at June 2026 prices. The legitimate parallel is narrower: both stories involve yield promises that depend on market confidence, which is why the dividend track record is the metric to watch.

Arithmetically, for decades. Funding the full $117 million monthly preferred bill by coin sales at $62,000 costs roughly 1,890 BTC a month, 0.22 percent of holdings, sustaining payments for over 30 years before exhaustion. The cost is not insolvency but narrative: every sold coin erodes the Bitcoin-per-share growth story the common stock is priced on.

A live precedent exists: MARA's sale of roughly 15,000 BTC for $1.1 billion in March 2026, one of the largest corporate liquidations ever, was absorbed without a market cascade. Sector-wide forced selling would require prices near the major claims break levels, roughly $25,000 to $52,000 depending on the company, and remains a tail scenario at current coverage ratios.

A 0-to-100 composite published by Decentralised News scoring digital asset treasury companies on four weighted dimensions: asset coverage stress (35 percent), obligation runway stress (30 percent), cost-basis stress (15 percent) and dilution dependency (20 percent), computed from filing-sourced board inputs under any user-defined Bitcoin price scenario. Readings above 60 indicate critical stress; the formulas and weights are fixed and published.

The trade-off is now explicit: treasury equities add a capital structure, dividends senior to you, dilution mechanics and an mNAV premium or discount, on top of Bitcoin exposure. With most of the sector's premium gone, the historical reason to prefer the wrapper, accretive issuance compounding coins per share, is impaired at many firms. Direct spot ownership removes every layer of the stress test. Not financial advice.

Decentralised News publishes research, not financial advice and not solvency opinions. Strategy figures are sourced from SEC Forms 8-K dated June 1 and June 8, 2026 and the Q1 FY2026 earnings release; Metaplanet figures from its Q1 2026 report; MARA, Twenty One, Semler, NAKA and sector mNAV figures from company disclosures, K33 Research data and press reporting as of June 12, 2026; cells marked est. are editable estimates, not verified filings data. Model outputs are scenario arithmetic under stated assumptions, all of which are published above and several of which are deliberately conservative. Equities and crypto assets can lose all value; preferred dividends can be suspended; past distribution records do not guarantee future payments. Some links are referral links that support our free tools at no cost to you. The DN Treasury Stress Index methodology, alongside the wider instrument suite documented in the editor's books Blockchain Applied and Tokenized Trillions, is open to challenge via the contact page.