Due Diligence Capital Notes

Most companies raise capital when they have a specific project to fund — a factory, an acquisition, a product launch. Strategy Inc. has no factory. Its "project" is buying more of a single, fixed-supply digital asset, and on March 23, 2026, it filed a prospectus supplement authorizing itself to sell up to another $21 billion worth of its own shares to do exactly that. To put that number in context: $21 billion is larger than the entire market capitalization of many S&P 500 constituents. Strategy is not a bank, not a commodity producer, and not a tech conglomerate. It is, functionally, a Bitcoin treasury vehicle running one of the most aggressive capital-raising machines in modern finance.

The question worth sitting with is not whether this is audacious — it obviously is. The question is whether the math still works, and what it means to be a new buyer of the common stock at today's price.

What Is an ATM Offering, and Why Does Strategy Use It?

An At-The-Market (ATM) offering is a mechanism that lets a public company sell newly issued shares directly into the open market, through broker-dealers, at whatever the current market price happens to be. Unlike a traditional secondary offering — where the company announces a fixed block of shares, negotiates a discount with underwriters, and then sells everything in one shot — an ATM lets the company drip shares into the market continuously, day by day, at prevailing prices. The company doesn't have to take a public "underwriting discount," and because the sales happen gradually, any individual day's selling pressure is relatively small.

Strategy has made ATM offerings the central engine of its capital strategy. The logic is clever: if the stock trades at a significant premium to the underlying value of its Bitcoin per share (what I described in an earlier post as the NAV multiple — the ratio of the stock price to the Bitcoin value attributable to each share), then selling new shares at that inflated price and immediately using the cash to buy Bitcoin is actually good for existing shareholders on a per-share basis. Each new share issued above NAV means the Bitcoin pile grows faster than the share count does. This is what practitioners call accretive dilution — a phrase that sounds like an oxymoron but is precisely accurate when the premium is large enough.

The critical variable, obviously, is maintaining that premium. And that is where $21 billion in fresh authorization gets interesting.

Breaking Down the Capital Structure

Before evaluating whether the new offering is accretive or destructive, you need to understand where common shareholders actually sit in Strategy's capital stack. "Capital stack" refers to the layered hierarchy of who gets paid first in any liquidation or restructuring — and in Strategy's case, the common shareholder is at the very bottom of a deep pile.

Consolidated senior debt: ~$8.25 billion as of December 31, 2025. Strategy has six series of convertible senior notes — bonds that, if the stock price is high enough at maturity, convert into Class A shares. These notes mature between 2028 and 2032 and could generate roughly 26.4 million additional shares upon conversion. But crucially, they are senior to equity: in a liquidation, debtholders get paid before any shareholders see a dollar.
Five series of perpetual preferred stock. Strategy has layered several preferred stock series on top of common equity. These include 50,246,513 shares of Variable Rate Stretch, 14,024,221 shares of 10% Stride, 14,020,744 shares of 8% Strike (STRK), 12,839,689 shares of 10% Strife, and 7,750,000 shares of 10% Stream. Perpetual preferred stock is equity that never matures, sits senior to common stock in the capital hierarchy, and must receive its dividend before common shareholders receive anything. It is not debt legally, but economically it behaves like expensive, permanent debt.
Class A common stock: 325,954,147 shares outstanding as of March 19, 2026. This is the layer that bears all residual risk. After the debt is paid, after the preferred dividends are covered, what's left belongs to you.

The new $21 billion ATM is authorized to sell additional Class A shares. The filing estimates a maximum post-offering diluted share count of approximately 526,752,018 — roughly 62% more shares than currently outstanding. That's not a rounding error; it's a structural reshaping of the equity base.

The Numbers Inside the 424B5

The prospectus supplement filed March 23, 2026 on SEC EDGAR is worth reading in full — it is unusually candid for a corporate filing. Here are the mechanics I found most important.

The prior program raised $9.61 billion. Under the original ATM program, Strategy sold 57,766,889 shares for total proceeds of $9,614,125,940.65. That's an average sale price of roughly $166 per share. As of March 19, 2026, $6.24 billion of capacity remained unused under the prior program. The new $21 billion authorization supplements that — so the total available firepower entering spring 2026 was closer to $27 billion when you combine the residual and the new capacity.

The stock was trading at $135.66 on March 20, 2026, down meaningfully from an all-time split-adjusted high of $543.00. For comparison, Strategy's split-adjusted stock price was $12.36 on August 10, 2020, the day it announced its Bitcoin strategy. That is an extraordinary run. But $135.66 represents a stock that has given back considerable ground from the peak — and the filing is explicit about the premium to fundamentals.

The prospectus discloses that the as-adjusted net tangible book value per share (the estimated accounting value of the company's net assets divided by the projected share count after the offering) would be approximately $36.88 per share. If you bought the common stock at $135.66 on the day of the filing, you were paying roughly $98.78 more per share than the book value of the net assets backing you. The filing calls this "immediate dilution to purchasers" — and states the number plainly, which I respect.

How does Strategy justify issuing shares at such a steep premium to book? Through this language in the 424B5 filing:

"Our equity market capitalization as of December 31, 2025 is in excess of our stockholders equity calculated in accordance with U.S. GAAP, and in excess of valuations that might traditionally be expected based on our operating performance, cash flows and net assets. Investors may therefore be unable to assess the value our class A common stock or evaluate the risks of an investment in us using traditional or commonly used enterprise valuation methods."

That is the company telling you — in its own prospectus — that standard valuation frameworks do not apply. Whether that is a warning or a feature depends entirely on your view of Bitcoin's trajectory.

As for the use of proceeds, the company states plainly:

"We intend to use the net proceeds from this offering for general corporate purposes, including the acquisition of bitcoin, paying cash dividends on our outstanding preferred stock, and for working capital."

Notice the order: Bitcoin first, preferred dividends second, working capital third. Common shareholders are funding everyone else's claims before they see any benefit.

The Accretion Calculation in Practice

Let me spell out exactly when this works and when it doesn't.

When it works: The stock is trading at a NAV multiple significantly above 1.0. Strategy sells shares at, say, $135. The Bitcoin per share implied by the company's holdings is, say, $60. Strategy takes the $135 in cash and buys Bitcoin. The Bitcoin per share attributable to each existing share increases, because the cash deployed per new share issued is more than the Bitcoin value already backing each existing share. Shareholders are diluted in count but enriched in Bitcoin per share. This is the bull case, and it is mathematically rigorous as long as the premium persists.

When it breaks: The NAV multiple collapses — either because Bitcoin falls sharply, or because the market stops believing in Strategy's accumulation story, or both. If Bitcoin drops to $40,000 and the stock follows the underlying value down, new ATM shares are sold at or near NAV. At that point, each new share issued simply maintains rather than grows the Bitcoin-per-share ratio. The "accretive" engine stalls. And the $8.25 billion in senior debt doesn't move: it stays fixed while the asset base shrinks.

What Could Break This Thesis

I want to be direct about the failure modes here, because they are serious.

Bitcoin price volatility. The filing itself notes that Bitcoin "has traded below $65,000 per bitcoin and above $120,000 per bitcoin in the 12 months preceding the date of this" filing. A nearly 2x range in 12 months is extraordinary. Strategy's entire value creation engine runs on Bitcoin appreciating faster than shares are diluted. A sustained bear market — not a brief correction, but a multi-year drawdown like 2022 — would impair the deployed proceeds, collapse the NAV premium, and leave the ATM machine idle precisely when it can no longer be accretive.

Severe and compounding dilution. The maximum post-offering diluted share count of 526,752,018 is roughly 61% higher than current shares outstanding. Stack on top of that the potential conversions from six convertible note series (~26.4 million additional shares) and the conversion rights embedded in several preferred series, and the dilution path for common shareholders becomes genuinely difficult to model. Each new issuance at a declining NAV premium reduces the per-share benefit.

Junior position in liquidation. I want to say this clearly: if Strategy ever needed to wind down — whether due to a Bitcoin collapse, a regulatory prohibition on Bitcoin treasury strategies, or a debt covenant violation — common shareholders sit below $8.25 billion in debt and below five series of perpetual preferred stock. The prospectus states explicitly that common stockholders could receive nothing in such a scenario. This is not a hypothetical buried in fine print; it is the contractual reality of the capital structure.

No minimum, no commitment. The ATM has no floor: there is no minimum amount Strategy must raise, no specific Bitcoin acquisition schedule, and broad management discretion over how proceeds are deployed between Bitcoin, preferred dividends, and working capital. Long-term believers in the Bitcoin accumulation thesis cannot point to a contractual guarantee that the $21 billion will actually go into Bitcoin on any particular timeline.

Conclusion

Here is what I keep coming back to: the $21 billion ATM offering is not a sign of desperation, and it is not a reckless gamble. It is the logical extension of a strategy that has been internally consistent since August 2020. Strategy's bet is that Bitcoin's long-run appreciation will outpace the dilution of issuing shares at a premium — and that the premium itself will persist because enough investors want leveraged Bitcoin exposure through a public company with index inclusion, options markets, and preferred-stock infrastructure.

The risk, though, is asymmetric in a way that is easy to underestimate. When Bitcoin is rising, every mechanism in the capital structure works together: the ATM is accretive, the NAV premium expands, convertible bondholders are happy, and preferred dividends are covered. When Bitcoin falls hard, those same mechanisms compound the damage — the premium evaporates, the ATM becomes neutral or dilutive, and the senior claimants (debtholders, then preferred shareholders) absorb the first dollars of recovery while common equity absorbs the first dollars of loss. The $21 billion authorization extending into this environment is a statement of conviction. Whether that conviction is rewarded depends almost entirely on one variable that Strategy cannot control at all.