Bitcoin Per Share Is Three Different Numbers. Only One Survives a Flat Price.

A company can grow its bitcoin per share while quietly handing value to its preferred holders. Both things happen in the same quarter, on the same balance sheet, and the press release will happily print whichever number looks better.

That's the trap in the treasury-company trade right now. Bitcoin per share sounds like one number. It's three, and they diverge fast once a company starts issuing convertibles, warrants, and perpetual preferred.

The real bottleneck isn't the bitcoin. It's the claims stacked on top of it.

The three-number problem

Start with the easy one. Raw BPS divides total bitcoin holdings by common shares outstanding. Clean, boring, and it's what most people quote on social media because it takes four seconds to compute. A company holding 500,000 BTC with 250 million shares outstanding shows 0.002 BTC per share. Done.

The problem is that raw BPS lives in the present tense. It counts shares that exist today and pretends the warrants, convertible notes, and convertible preferred sitting in the capital structure don't. They do. They're just waiting.

So you get Bitcoin Per Assumed Diluted Share, or BPADS, which divides the same bitcoin pile by a fully diluted share count. Because the denominator only grows, BPADS is always less than or equal to raw BPS. The gap between the two is basically a measurement of how much future dilution the company has already sold.

That metric got real traction starting in 2024, right around the point treasury companies stopped having simple balance sheets. The timing wasn't an accident.

But BPADS has its own blind spot, and it's bigger than people admit. Full dilution assumes every warrant gets exercised and every convertible note converts. Some won't. A convertible struck well above the current share price may sit there for years collecting dust. Perpetual preferred can stay outstanding indefinitely if holders prefer the dividend to the common equity. BPADS is a scenario, not a forecast. Treating it as one is a category error.

Then there's Net BPS, the one that actually respects the capital stack. The math: take total bitcoin holdings, subtract the bitcoin-equivalent value of senior claims, divide by shares. Those senior claims are dollar-denominated, which is exactly why this gets interesting. A company with $1 billion in debt and preferred at a $100,000 bitcoin price needs 10,000 BTC to cover it. At $50,000, that same obligation eats 20,000 BTC. Same company. Same debt. Very different answer for the common shareholder.

You also have to decide what to do with cash. Apply it against the senior claims first, and you preserve more bitcoin for the common. That's the methodology behind Strategy's mNAV calculation. Or deduct the full claims from bitcoin and report cash separately as cash per share. Both are defensible. What isn't defensible is doing one and describing it as the other.

Where the accretion math breaks

Here's where I'll be blunt. Most of the accretion claims floating around crypto Twitter are noise, because they compare Net BPS at two different dates and call the difference management skill.

Net BPS is price-sensitive by construction. Senior claims are denominated in dollars, bitcoin isn't, so a rising bitcoin price mechanically lifts Net BPS even if the financing decisions in between were terrible for common holders. Falling prices hide genuinely good deals the same way. Comparing a March number to a September number tells you almost nothing about capital allocation. It tells you what bitcoin did.

Meanwhile, raw BPS is even worse as an accretion test, and it fails in a direction that flatters management. Issue debt to buy bitcoin and raw BPS goes up no matter what the coupon is. Issue preferred at a steep discount to par with an expensive dividend and raw BPS goes up too. In both cases gross bitcoin per share rises while the residual value belonging to the common shareholder may be falling. That's not accretion. That's a magician's sleeve.

The only clean test is Net BPS before and after a transaction with bitcoin's price held constant. That isolates what actually changed: holdings, claims, cash, share count. If Net BPS goes up at a flat price, common shareholders got richer in bitcoin terms. If it goes down, no amount of gross-per-share growth saves the story.

So why does half the market still quote raw BPS? Because it's the number that always goes up. And here's a take that'll annoy some treasury desks: a company that reports only raw BPS in its shareholder letter has made a disclosure choice, not a mistake. It's the most flattering of the three figures, and it requires the least explanation.

Flip side, and I'll say this too, treating BPADS as the headline is also a choice. It's the scariest of the three numbers, and scary sells attention just as reliably as flattering sells shares. Fully diluted is a stress test, not a base case.

What the common shareholder should actually ask

Net BPS is the closest thing this sector has to a bitcoin-denominated liquidation value per common share. Multiply it by the bitcoin price and you get a dollar figure you can stack against the share price to see the premium or discount. That comparison is the whole game. It's also the one thing raw BPS and BPADS can never give you, because neither one knows what the senior claims are.

Does that mean Net BPS is perfect? No. Real recoveries depend on covenant terms, other assets, liquidation costs, and taxes on anything sold. But it's directionally right, and directionally right beats confidently wrong.

The scaling roadmap for this sector just got more interesting, and not for the reason most people think. It's not about how much bitcoin the biggest treasury company can accumulate. It's about whether the market learns to price the claims underneath. Right now it mostly doesn't. That's a mispricing, and mispricings get resolved.

Nobody cares about infrastructure until it breaks. Treasury balance sheets are infrastructure now, and the crack shows up the first time a company reports gross bitcoin per share growth in the same quarter its Net BPS falls at a flat price. When that happens, don't take the press release's word for it.

Ask for Net BPS at a fixed bitcoin price. If the company can't or won't give it to you, you've learned something more useful than any number it would've printed.