Due Diligence Capital Notes

By the close of February 2026, Strategy Inc. held 720,737 Bitcoin — roughly 3.6% of the 21 million coins that will ever exist. In a single week, the company added another 3,015 BTC, paying an average of $67,700 per coin, funded entirely by selling its own shares into the open market. Here is the number worth pausing on: the portfolio's blended average acquisition cost is $75,985 per BTC. The most recent purchases came in roughly $8,300 per coin below that figure. By any conventional accounting measure, the portfolio is sitting in an unrealized loss position.

And yet the capital markets kept funding the machine. Preferred shareholders kept subscribing. ATM proceeds kept arriving. Over $37 billion in additional issuance capacity sits unused across Strategy's five share classes. Far from pulling back, the company actually raised the dividend rate on one of those preferred instruments to keep investor appetite alive. So the question that nags at me is this: what exactly do the people financing this operation believe they're buying?

The ATM Flywheel: Selling Stock to Buy Bitcoin

To understand Strategy's financial engine, you need to grasp two mechanisms working in tandem — and why, when combined, they create something more powerful than either alone.

An ATM offering (at-the-market offering) is not the dramatic secondary offering you might picture, where a company floods the market with millions of new shares overnight and sends the stock into a tailspin. Instead, the company appoints a broker-dealer and authorizes it to sell small tranches of freshly issued shares day by day, at whatever the prevailing market price happens to be. The dilution is gradual, the market impact is managed, and the proceeds flow continuously. For Strategy, those proceeds flow in one direction: toward Bitcoin.

The second mechanism is preferred stock — a class of security that sits in the capital structure between debt and common equity. Preferred stockholders receive fixed or variable dividends before common shareholders get anything, and in a liquidation scenario they stand ahead of common equity holders. In exchange, they typically give up voting rights and the ability to participate in extraordinary upside. For investors who want yield with some Bitcoin-adjacent exposure but don't want the full volatility of MSTR common shares, preferred stock is a natural fit.

Put these two mechanisms together and you get a self-reinforcing cycle: Strategy raises capital from equity and preferred investors, converts it into Bitcoin, Bitcoin's long-run price appreciation raises the NAV (net asset value — the total value of BTC holdings minus liabilities, divided by the share count) of MSTR common stock, that elevated NAV supports a higher stock price which generates more dollars per ATM share sold, and the cycle repeats. The engine is designed to accumulate Bitcoin at scale, not to optimize for any single quarter's accounting outcome.

Breaking Down the Five-Instrument Capital Stack

What distinguishes Strategy's current setup from earlier iterations of this strategy is the sheer breadth of the capital structure. The company now runs five simultaneous ATM programs across different security types, each targeting a different investor base:

MSTR common stock — The flagship instrument. Between February 23 and March 1, 2026, Strategy sold 1,730,563 shares via ATM, generating $229.9 million in net proceeds. That's a single week of activity. Remaining ATM capacity on MSTR common stands at $7.61 billion.
STRK — A convertible preferred stock (meaning holders can, under defined conditions, convert their preferred shares into MSTR common equity). STRK pays a fixed quarterly dividend of $2.00 per share and carries the largest remaining ATM capacity of all five instruments at $20.33 billion — a number that signals how aggressively Strategy intends to deploy this particular instrument.
STRD — Another fixed preferred instrument, paying $2.50 per share quarterly, with $4.01 billion in remaining ATM capacity.
STRF — Also paying $2.50 per share quarterly, with $1.62 billion remaining. STRF, STRE, and STRD dividends declared for the March 31, 2026 payment date each carry the same $2.50 per share quarterly rate.
STRC — The Variable Rate Series A Perpetual Stretch Preferred Stock. "Perpetual" means it has no maturity date and never has to be redeemed. "Variable rate" means the dividend can be adjusted over time — and adjusted it was: effective March 1, 2026, Strategy raised the STRC annual dividend rate from 11.25% to 11.50%, translating to $0.9583 per share per month. As the company's filing states: "Strategy increased the regular dividend rate per annum on its Variable Rate Series A Perpetual Stretch Preferred Stock effective for monthly periods commencing on or after March 1, 2026 from 11.25% to 11.50%." The STRC ATM sold 71,590 shares during the same week for $7.1 million in net proceeds, with $3.54 billion in capacity still available.

Combined, the five programs have more than $37.1 billion in remaining authorized capacity. That figure represents pre-registered authority to raise capital without returning to shareholders for another vote — a strategic optionality that few companies of any size possess.

The Numbers Behind a Single Week

The Form 8-K filed March 2, 2026 lays out the arithmetic with notable precision. Strategy spent $204.1 million to acquire 3,015 BTC at an average price of $67,700 per coin. The funding source, quoted directly from the filing: "The bitcoin purchases were made using proceeds from the sale of shares under the ATM." Total net ATM proceeds across MSTR common and STRC preferred during the period were $237.1 million — a slight surplus over the BTC spend, leaving a small cash buffer.

Zoom out to the full portfolio and the numbers become genuinely staggering. Strategy's 720,737 BTC were acquired at a cumulative cost of $54.77 billion, producing that blended average of $75,985 per coin. To put the weekly accumulation in context: approximately 3,150 new Bitcoin are mined globally every week at current block reward levels. Strategy's single-week purchase of 3,015 coins amounts to nearly one full week of global new supply absorbed by a single corporate buyer.

The STRC rate increase deserves its own moment of attention. When a company raises a preferred dividend, it almost always signals that the previous rate wasn't attracting enough capital fast enough — that marginal investors demanded more yield before committing. Moving from 11.25% to 11.50% is a small increment, but it is directionally meaningful. Strategy is willing to increase its cost of capital to keep the preferred issuance pipeline active. That's a rational trade-off as long as the Bitcoin being purchased with that capital appreciates faster than the preferred dividend rate — but it is a commitment that grows over time.

Strategy also maintains a real-time disclosure dashboard at www.strategy.com, officially cited in the 8-K as a Regulation FD disclosure channel — meaning it qualifies as a public forum where material information reaches all investors simultaneously rather than flowing first to institutions. BTC holdings, security prices, and key performance metrics are published there continuously.

What Could Break This Thesis

Strategy's flywheel is elegant when Bitcoin is in a sustained uptrend. In a different environment, several specific scenarios would stress or break the model.

The cost basis becomes a trap. The portfolio's blended acquisition cost is $75,985 per BTC. A sustained decline below that level would place the entire $54.77 billion in deployed capital in an unrealized loss. More critically, a prolonged Bitcoin downturn would compress MSTR's stock price, reducing the NAV premium that makes ATM issuances accretive. If MSTR trades at or below its Bitcoin-adjusted NAV, selling new shares to buy BTC at market price no longer creates value for existing holders — it destroys it.

Dilution compounds against common shareholders. Strategy sold 1.73 million MSTR shares in a single week. Across $37+ billion in remaining ATM capacity, the implied forward dilution is substantial. Common shareholders are making an implicit bet that BTC per share rises faster than the share count — a wager that requires Bitcoin to outpace the dilution rate consistently over years. If that math breaks down, common equity absorbs the entire cost while preferred holders continue collecting fixed dividends.

Preferred dividend obligations become structural pressure. Five classes of perpetual preferred stock, paying between 8% and 11.50% annually, must be serviced regardless of Bitcoin's performance. These are obligations, not discretionary distributions. If access to capital markets narrows during a broad equity downturn, Strategy would need to service preferred dividends from existing cash or BTC sales rather than new issuances — fundamentally changing the unit economics of the model.

Tax treatment of dividends faces reclassification risk. Strategy expects the preferred dividends payable March 31, 2026 to qualify as non-taxable return of capital for U.S. federal income tax purposes — meaning investors defer tax on those payments by reducing their cost basis in the preferred shares rather than recognizing ordinary income immediately. This is an expectation based on current accounting positions, not an IRS ruling. A recharacterization to ordinary income would reduce the after-tax yield on STRC, STRD, and STRK, potentially dampening demand for future preferred issuances precisely when the company needs that demand most.

The Experiment at Scale

There is a certain cold logic to what Strategy is doing once you strip away the jargon. The company has concluded that Bitcoin will appreciate over long time horizons, that its stock trades at a premium to NAV because the market values its ongoing accumulation, and that the right move is to convert as much of that premium as possible into Bitcoin before the window closes. The current cost basis is almost irrelevant to this logic — what matters is the BTC per diluted share figure over a decade, not the unrealized P&L in any given quarter.

The STRC rate increase is a small tell about where we are in that arc. Investor appetite for Strategy's preferred instruments is real, but it isn't infinite or unconditional — it requires ongoing yield adjustments to stay active. Whether Strategy can deploy all $37+ billion in remaining capacity at rates that still make Bitcoin accumulation accretive to common shareholders is the central execution question for the next several years. At 720,737 BTC and counting, the experiment is already the largest corporate Bitcoin accumulation program in history. Whether it becomes a template or a cautionary tale will depend almost entirely on what Bitcoin does from here — and on whether Strategy's financial engineering remains one step ahead of its own cost structure.