Why Strategy sold 32 Bitcoin: the math says it had to

Strategy sold 32 BTC for $2.5M to fund preferred dividends. The $1.712B annual bill against a $900M reserve made the move mechanical, not symbolic.

The Editors · 7 min read · Jun 5, 2026

A bitcoin sitting on top of a pile of gold nuggets

Strategy sold 32 Bitcoin between May 26 and May 31, 2026, for about $2.5 million. The 8-K hit on June 1. The first take everywhere was that 32 coins out of 843,706 is a rounding error of 0.0038%, and Saylor's "never sell" pledge looked broken in the way a ceremonial speed limit gets broken: technically, on paper, without consequence.

But the math demanded it.

Strategy owes its preferred shareholders $1.712 billion a year. Its US dollar reserve at the end of May was $900 million. The gap is real and recurring, paid in cash every month. When the equity machine slows, the thing that closes the gap is Bitcoin. The 32-coin sale is the first time the company has put that mechanism in a filing instead of on a podcast.

The rest of this piece is what that means: for Strategy, for the 173 other public companies that copied the playbook, and for the Bitcoin price they all sit on top of.

The financial engineering loop, in plain words

Strategy now operates as a Bitcoin holding vehicle wrapped in equity and preferred stock. The loop is straightforward.

Issue equity (common or preferred) at a premium to the underlying Bitcoin value.
Use the cash to buy more Bitcoin.
The new Bitcoin lifts the net asset value per share.
The lift is accretive because shares were issued above NAV.
Repeat.

The market calls the premium "mNAV": the ratio of market cap to the dollar value of the Bitcoin treasury. Strategy's mNAV peaked above 3.0 in the bull leg; as of last week it sits at about 1.22. Above 1.0, the loop compounds. Below 1.0, issuing equity destroys per-share Bitcoin value, and the company can't fund its dividends by selling stock.

The dividend obligation lives in the preferred stack. STRC, the Variable Rate Series A Perpetual Stretch Preferred, pays an 11.50% annualized rate on a $100 stated amount, distributed monthly, forever. Across all preferred series, the combined annual dividend obligation is $1.712 billion. That number does not pause when Bitcoin falls.

The two numbers that decide it

Item	Amount
Annual preferred dividend obligation	$1.712 billion
USD reserve, May 31 2026	$900 million
Bitcoin sold in May 2026	32 BTC (about $2.5M)
Bitcoin held	843,706

$900 million divided by a $1.712 billion annual run rate is roughly 6.3 months of dividend coverage from the reserve alone. That assumes zero new equity issuance, zero further asset sales, and a flat dividend rate. None of those are guaranteed.

When the mNAV premium runs hot, Strategy refills the reserve by issuing preferred stock at par. In Q4 2025 alone, the STRC at-the-market program brought in $157.6 million. When the premium narrows, the pace slows. The reserve drains. The Bitcoin gets sold.

Saylor's only public response was one sentence: "Our goal is to make STRC the best credit instrument in the world." Read that as the company telling preferred holders their coupon ranks ahead of the Bitcoin pile. The 32-coin sale was the first proof.

Why the premium is narrowing right now

Bitcoin is down roughly 20% since May 15. Spot Bitcoin ETFs have seen 13 consecutive days of net outflows totalling $4.4 billion. The macro backdrop is risk-off: rising yields, equity weakness, and a sentiment shock from this same Strategy sale that triggered $1.25 billion in long futures liquidations in 48 hours.

When Bitcoin falls, two things compress the mNAV premium at once.

The denominator (Bitcoin value) drops, pulling the ratio mechanically toward 1.0.
The numerator (market cap) drops faster, because investors price in the debt load and the dividend bill.

That asymmetry is the trap. Strategy's market cap reflects both the Bitcoin balance and the $1.712 billion in annual fixed claims that sit senior to common equity. As the premium narrows, those claims become a larger share of enterprise value. The equity market re-prices the common share lower, which narrows the premium further. The loop runs in reverse.

The structural model published on bitcointreasuries.net puts the breaking point at roughly $50,000 BTC, the level where Strategy's mNAV approaches 1.0 and the equity playbook stops working. Below $26,290, common equity gets wiped under the preferred stack.

What this means for the 173 copycats

Strategy is the biggest, but it's not alone. BitcoinMiningStock.io tracks 174 public companies holding Bitcoin as a treasury asset. After Strategy, the largest by BTC holdings are:

Twenty One Capital (XXI): 43,514 BTC
Metaplanet (3350.T): 40,177 BTC
MARA Holdings: 35,303 BTC
Bullish (BLSH): 24,300 BTC

Most of them launched after Strategy's mNAV crossed 2.0 and proved the loop. Most don't have Strategy's scale, its full preferred stack, or its reserve. What they share is dependence on a high mNAV premium to keep issuing accretive equity.

When the leader sells Bitcoin to fund coupons, three things change for everyone running the same model.

The narrative cost goes up. "Bitcoin treasury company" used to imply "we hold and never sell." Now it implies "we sell when the math says we have to."
Equity holders re-price the risk. The premium investors paid for the structure includes a discount for forced selling. That discount widens.
The collective overhang grows. If 174 companies hit their own mNAV-1.0 line at similar Bitcoin prices, the supply they're forced to release lands in the same market window.

CryptoSlate reported that Metaplanet has held up structurally where Strategy investors are exposed. That's a capital-structure story. Companies without large preferred obligations don't have a monthly dollar bill that has to be paid in cash.

The honest counter-argument

Three things could disprove the thesis or push it out.

Bitcoin recovers. A move back above $80,000 widens the mNAV premium, the at-the-market preferred issuance restarts, the reserve refills. The 32-coin sale becomes a footnote.

Strategy finds non-Bitcoin liquidity. The company has issued convertible notes before and can issue more, though at terms set by a market that now knows the dividend bill. The cost goes up. The runway extends.

The dividend itself flexes. STRC's rate is variable. Strategy can lower it, at the cost of crashing the preferred's market price and gutting the claim of being a "best credit instrument."

None of these fix the structural point. They delay it.

What to watch from here

STRC at-the-market pace. Weekly filings show how fast Strategy is refilling the reserve through preferred issuance. If the pace slows, Bitcoin sales scale.
Monthly 8-K disclosures on BTC activity. The next month or two will tell whether 32 was the floor or the first step.
Mid-tier copycats. The first cracks in the model will show up in the smaller treasury vehicles, not the leader. Watch for dividend rate cuts or strategy shifts at companies in the 5,000-25,000 BTC range.
Bitcoin near $50,000. Watch it because the published model says that is where Strategy's mNAV breaks 1.0.

Saylor was right that 32 BTC is "inconsequential" on its own. The 32 coins matter because they prove the door is open. Once a company sells Bitcoin to pay its dividend, the next questions are practical: when, how many, and how it compounds when 173 other companies are running the same machine.