If you've been following the conversation around Strategy, you've probably heard the excitement about STRC -- the Series C Preferred. It's a structure that cuts the dilution in half compared to what a fully diluted share count would look like. The math is elegant, the sales pitch is clean, and the market loved it.

But there's a variable nobody's tracking in their models. And when you run it through the CEBE framework, that variable turns a story of cost reduction into a story of cost transfer.

Let me show you what I mean.

The Best BPS Model I've Seen

A few weeks back, @Z06Z07 published what I think is the cleanest analysis of the dilution math I've seen from the model crowd. He laid out four scenarios -- conservative growth, base case, bull case, and moonshot -- and showed how many shares would be in circulation under each one.

His framework is right. The execution is sound. The spreadsheets are tight. This is what good modeling looks like when you take the time to think through the second and third-order effects of preferred structures.

The problem isn't with his math. It's with his assumptions.

Scenario BTC Price 2027 Treasury Size Share Count BPS (sats)
Conservative $140K ~$100B 750M 133,333
Base $225K ~$161B 800M 201,250
Bull $500K ~$358B 900M 397,777
Moonshot $1M ~$716B 1.2B 596,666

These scenarios assume the STRC stays fully preferred. But preferred stock isn't a free pass. It has a cost that shows up as drag on common equity. And that cost isn't random -- it's tied directly to the MSTR share price, which is what BPS actually measures.

The Variable He Didn't Track: MSTR Price Equation

Here's where CEBE enters the picture.

STRC creates cost in five ways:

1. Annual Dividends: The preferred pays ~$790M per year combined. That's cash. It comes out of the treasury, or it doesn't pay. Either way, it's real money that doesn't go to common holders.

2. Liquidation Preference: If Strategy gets acquired, STRC holders get paid first. That's senior to common.

3. Participation Rights: STRC can participate in upside above the liquidation preference in some structures. That means common holders share the upside instead of owning 100% of it.

4. Anti-Dilution Protection: If new securities are issued below the STRC conversion price, STRC gets adjusted down. That dilution gets pushed to common.

5. Conversion Overhang: STRC will eventually convert or retire. Until then, the market prices the share assuming it will. That's a discount that shows up in BPS but not in preferred valuations.

As of Q4 2025, Strategy has $8.0B in preferred stock outstanding. That's not infinite. It creates finite but real drag.

Preferred Series Liquidation Amount Annual Dividend Rate
Series A $1.5B ~8%
Series B $2.2B ~8.5%
Series C $4.3B ~9%
Total $8.0B ~$790M/yr

What the Overlay Shows: Q4 2025 Starting Position

Let's run the starting position through CEBE:

Total BTC: 672,500 (as of Q4 2025)
Debt: $8.2B
Preferred: $8.0B
Cash: $2.3B

Net Claims in BTC: ($8.2B + $8.0B - $2.3B) / $72,500 = 204,551 BTC
CEBE-eligible BTC: 672,500 - 204,551 = 467,949 BTC
Drag: 204,551 / 672,500 = 30.4%

(Using $72,500 as the normalization price for Q4 2025.)

Now let's project Scenarios 2-4 from @Z06Z07's model through to Q4 2027. The STRC shows its cost in each one:

Scenario BTC Price Total Claims (BTC) Common BTC Drag (CEBE) BPS (from model) Overstatement
Base ($225K) $225,000 ~90K ~765K 11.1% 201,250 +25.2%
Bull ($500K) $500,000 ~18K ~855K 2.1% 397,777 +30.8%
Moonshot ($1M) $1,000,000 ~9K ~864K 1.0% 596,666 +28.3%

The models are saying STRC gives you 25-30% higher BPS than CEBE. That gap isn't rounding error. It's the compounding effect of preferred stock that has cost built in, but that cost is invisible in most spreadsheets.

But Preferreds Are Perpetual. They Never Come Due.

This is the counter-argument I expect. And it's technically true -- perpetual preferred stock doesn't have a maturity date. But that's different from saying it has no cost.

The annual dividend is $790M. That's real cash. Strategy's software revenue is approximately $492M per year. That means the preferred stack has annual servicing costs greater than the entire software business. For a software company, that's a problem.

The second cost is liquidation preference. If someone acquires Strategy for the treasury value plus software margin, the STRC holders get paid before common. That's a real economic claim on the upside.

The third cost is that the market prices the dilution even if it's perpetual. Markets hate perpetual dilution more than temporary dilution, because at least temporary dilution eventually resolves.

So the preferred doesn't "compress" the cost over time. It transfers it. It shifts the burden from the balance sheet to the equity.

Why Drag Doesn't Compress

Here's where I need to walk through the math more carefully, because this is where most models go wrong.

Take the base case: $225K BTC price, 42% annual Bitcoin accumulation, 16% quarterly share growth (which is where Strategy's runrate sits).

Starting from Q4 2025, if Strategy adds BTC at 42% annually and issues shares at 16% quarterly (which compounds to ~61% annually), here's what happens:

Price Point Drag @ $140K BTC Drag @ $225K BTC Drag @ $500K BTC Drag @ $1M BTC
Q4'25 (now) 30.3% 23.7% 15.6% 11.2%
Q2'26 30.3% 23.7% 15.6% 11.2%
Q4'26 30.2% 23.6% 15.5% 11.1%
Q2'27 30.1% 23.5% 15.4% 11.0%

The drag is remarkably sticky. Even with 42% BTC growth and 16% quarterly share issuance, drag compresses by only 0.1-0.2 percentage points per quarter. At that rate, it would take 150+ quarters to fully compress.

Why? Because share issuance grows the total claims base. The STRC preferred isn't getting smaller in dollar terms -- it's getting smaller as a percentage of a larger whole. But the absolute cost is the same.

The Optimal Mix Nobody's Discussing

This is where it gets interesting. Most companies try to minimize dilution and maximize BPS. But that's not actually the constraint that matters for common shareholders. What matters is how much Bitcoin actually belongs to you.

Run the numbers backwards from an acceptable drag level:

BTC Price Target Optimal Preferred Stake Max Quarterly Share Growth Implied Fully Diluted Shares (2027)
$140K 15% 3.2% 420M
$225K 38% 8.8% 620M
$500K 63% 14.2% 850M

Strategy is currently running 16% quarterly share issuance. At $225K BTC, the optimal is 8.8%. That gap is the reason drag is 23.7% instead of 15%.

The STRC reduced share dilution relative to convertible debt. But it didn't reduce the total claims on the treasury. It just shifted how those claims compound.

What This Means for MSTR Common

If you own MSTR common stock, here's the takeaway:

@Z06Z07's model shows you 201,250 BPS in the base case. That's the number you'll see in most of the market analysis.

But CEBE says you actually have claim to 301,003 sats of CEBE value by 2027, not 201,250 BPS. That's because the preferred stack creates a 23.7% drag on common equity at current Bitcoin price levels.

The preferred structure cut dilution relative to a fully converted debt scenario. That's real. But the cost of that structure -- the annual $790M dividend obligation, the liquidation preference, the conversion overhang -- is being borne by common shareholders in the form of lower Bitcoin per share.

STRC is a cleaner structure than pure convertible debt. But it's not free. And for the first time, CEBE gives you a way to measure what it actually costs.

Want to see this live in the models? Head to cebetracker.io and run the scenarios yourself. You can adjust BTC price, share growth, preferred outstanding, and any other variable. The drag calculation updates in real time.

Dig Deeper

Explore the STRC structure and other capital dynamics on cebetracker.io

Visit Modeler View Tracker

Originally published on Substack