The Bitcoin company that said never sell is selling

Strategy’s whole machine only worked while its stock traded above the value of the Bitcoin it holds. That just stopped being true.

TLDR

• For the first time, Strategy is worth less than the Bitcoin sitting on its own balance sheet. The premium that powered everything flipped into a discount.
• The company that turned “never sell Bitcoin” into a brand reported its first Bitcoin sale since 2022, to cover dividends it owes whether Bitcoin goes up or down.
• The flywheel was only ever accretive above 1.0x mNAV. Below it, issuing stock shrinks Bitcoin-per-share instead of growing it. Same move, opposite result.
• The open question is whether this is a once-a-cycle discount or a machine quietly running in reverse.

The footnote on “never”

A while back, writing about the Bitcoin crash, I wondered out loud what Strategy would do if the drawdown got ugly. It was not a rhetorical question. The largest corporate holder of Bitcoin on earth had spent years promising it would never sell, and I wanted to know what “never” actually meant once the math turned against it.

The answer showed up in a regulatory filing. It turns out “never” had a footnote.

What Strategy actually is

Strip away the ticker and Strategy is a software company that became a Bitcoin holding company and then admitted it. It started life as MicroStrategy, a business intelligence firm from the 1990s. In 2025 it dropped the “Micro” and rebranded to Strategy, which was the most honest thing it had done in a while. The name finally matched the business.

Today it holds roughly 847,000 Bitcoin, more than any other public company by a wide margin. The legacy software operation still exists, but it is a rounding error next to the treasury. Forbes pegged fourth-quarter 2025 software revenue at about $129 million, which is not nothing, but it is also not the thing moving the stock. The stock moves because MSTR trades as leveraged Bitcoin. Its beta runs around 3.5, so it amplifies Bitcoin in both directions. On the way up, that was the entire appeal.

The machine, and why it only ran one way

Here is the part worth understanding, because it is the whole story.

From 2020 through 2024, Strategy ran what looked like a money printer. Issue new stock at a premium to the value of its Bitcoin. Use the cash to buy more Bitcoin. Watch Bitcoin-per-share rise. Let the bigger premium justify the next raise. Repeat. Saylor gave it a name, “Bitcoin yield,” and the market gave it a valuation. At its peak in late 2024, MSTR traded at close to 3.9 times the value of the Bitcoin it held. By 2025 it was the largest equity issuer of any US public company, two years running, raising around $25 billion in that year alone.

The catch was never hidden. It just was not bolded.

That loop only works above 1.0x mNAV, the ratio of the company’s value to its Bitcoin. As long as each new share sells for more than the Bitcoin it buys, the people already holding get richer in Bitcoin terms. The moment shares sell for less than the Bitcoin they buy, the same machine runs backward, and every new issuance dilutes the holders still standing there.

For four years, nobody had to think about the second half of that sentence.

I believed most of this

The idea was not stupid. The access pitch is real: ordinary investors got a regulated, liquid way to hold leveraged Bitcoin inside a brokerage account. The capital engineering was clever. And the long-horizon thesis, that Bitcoin grinds higher over a decade, might still turn out right.

I have written before about concentration and leverage, about how diversification fails exactly when you need it and how leverage takes a good idea and removes its patience. Strategy is that lesson at the largest scale anyone has attempted. I am not here to dunk on it. I am here because the thing I was watching for finally happened.

Here is where it breaks

This month, the premium became a discount.

Strategy’s valuation fell below the value of its own Bitcoin for the first time, a line a lot of people had been watching. CEO Phong Le had said at the end of last year that the company might consider selling Bitcoin if its value-to-Bitcoin ratio dropped under 1. It dropped under 1.

The receipts are ugly. MSTR is trading in the low $80s, down from an all-time high near $197. The Bitcoin pile sits underwater, bought at an average of about $75,650 a coin against a Bitcoin price near $67,000. The accounting catches up fast at that point. Strategy reported an unrealized digital-asset loss of roughly $14.5 billion for the first quarter, and a net loss of around $12.5 billion.

Below parity, the engine reverses. As one trading desk put it plainly, every new share sold now shrinks Bitcoin-per-share instead of growing it. The flywheel did not slow down. It started spinning the other way.

The gospel meets the coupon

Here is the part I keep getting stuck on.

Strategy did not just buy Bitcoin with stock. It also built a stack of preferred shares, marketed as “Digital Credit,” that pay fixed dividends. STRC alone carries an 11.50% annual payout. Add the stack up and the company owes somewhere between about $1.3 and $1.5 billion a year in preferred dividends, by analyst estimates, and those payments come due no matter what Bitcoin does.

So picture the box Strategy is in. The stock is below its Bitcoin value, which makes raising new equity dilutive. The preferred shares trade below par, which makes new credit more expensive. And the dividend checks still have to clear. The cleanest way to cover them is to sell the one asset the entire religion is built on never selling.

Which is what happened. Strategy reported its first Bitcoin sale since 2022, and disclosed it could sell up to $1.25 billion worth to fund dividends and debt. Saylor’s framing is that this is capital structure optimization, not a retreat, and that funding a full year of dividends takes only about 18,500 to 19,000 coins, roughly 2.2% of the stack. One analysis of the same filing put it less gently, calling it severing limbs to survive. Both descriptions fit the same document. That gap, between the press-release verb and the balance-sheet verb, is the whole story.

Leverage removes patience

I said this in a piece about diversification and it applies even harder here. Leverage takes a good idea and removes its patience.

The Bitcoin thesis might be right over ten years. The dividends are due this quarter. A balance sheet financed by fixed obligations does not get to wait for the long-term logic to come true. It has to make payments on a schedule the long term does not care about. That is the difference between holding Bitcoin and engineering exposure to it. Holding lets you wait. Engineering puts a clock on the wall.

Who finds out the hard way

A few groups are learning what they actually own.

• Retail traders who bought MSTR as “Bitcoin with extra upside” near the top, now down more than 50% while Bitcoin itself is down closer to 45%.
• Preferred holders who think “Digital Credit” is safe yield, when the cash behind that yield may be coming from selling a volatile asset into a weak market.
• Everyone else in Bitcoin, who now has to sit with the reflexive risk that the largest corporate holder turning seller can feed the very weakness it is selling into.

The honest counterweight

This is not a funeral, and I am not going to pretend it is one.

To be fair to the bulls, the tape has not been cooperating with the doom. The same stretch this milestone landed in, MSTR popped on earnings, the company kept adding coins, and the analysts who trimmed their price targets held onto positive ratings.

The bull case is coherent. Strategy still has cash set aside, around $1.4 billion in reserve and tens of billions in remaining issuance capacity. The sales so far are small against the size of the hoard. There is even a roughly $2.2 billion deferred tax asset from those first-quarter losses. And the core bull point is fair: if Bitcoin rallies, the premium can come back and the flywheel can simply restart. Trefis goes further, arguing the discount is more optical illusion than genuine distress once you account for the full capital structure.

It is not all clean, though. The Rosen Law Firm opened a securities investigation in late June into Strategy’s disclosures around its Bitcoin holdings, alongside broader scrutiny of how and when the company has reported its treasury moves. None of that is a verdict. It is the sound of the story getting contested in real time.

What this actually means

So here is where I land, which is the unsatisfying place.

Either this is a generational discount on the best Bitcoin proxy ever built, and the people buying MSTR below its Bitcoin value look brilliant in two years. Or it is a leveraged bet quietly running in reverse, propped up by selling the asset it swore to hold, while everyone argues about the multiple.

The same balance sheet that looks like a trap from one angle looks like a discount you get to buy from the other, and Bitcoin gets the deciding vote on which.

I do not know which. What I do know is that the cleanest version of the story, the perpetual machine that only ever printed up, was always missing a sentence. The premium was the product. The Bitcoin was the collateral. And never sell was a plan that worked right until the dividends came due. Whatever the token looked like on the way up, what you find out on the way down is the part that never made the pitch deck.

Thank you for reading.

-APL

Sources: Futunn, Yahoo Finance, Forbes, Trefis, TradingKey, SpotedCrypto, Intellectia, BitcoinTreasuries

The Bitcoin company that said never sell is selling was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.