\n\t\n\n\"We hold Bitcoin forever. We never sell.\" — a mantra that sounds wonderful as long as the creditors are polite and the premium is above one. The rules of the game don't ask how charismatic you are once the needle creeps toward zero.
\n
Picture a vending machine by the exit of your brokerage app: you feed in a dollar, you take out a dollar twenty. Of course you get in line. Not because you're stupid, but because you're alive — and because any sane person would stand in it, until retirement, without a break. In your head it's called \"investing.\" From the outside it's \"stood by a machine that printed me money, until it started printing it for someone else.\" For six years running, that line was called \"buy MicroStrategy stock.\"
\n\n\n\n*The brass perpetuum mobile as it was sold to the room: a flawless wheel feeding on its own premium to the asset's value. Nobody in frame is looking at the meter yet — and it is already counting the way back to one.*\n\nThis scheme existed earlier too — the company formed its Bitcoin treasury back in August 2020. But in 2024–2026 it became a structural problem, not just a risky bet. Two events changed the operating system underneath it. In January 2024 the SEC approved spot Bitcoin ETFs — and the \"unique wrapper\" trading at a 3.9x premium instantly became an ordinary wrapper next to a dozen cheaper ones. In May 2026 the mNAV touched 0.94x for the first time — the stock cheaper than the Bitcoin standing behind it. This isn't a warning signal on the horizon. It already became reality on the market while you read this text. Why read it now rather than a year from now? Because if the floor is already cracked, \"wait a little longer\" isn't a strategy — it's the title of a documentary about those who waited.
\n\nThe mechanics were elegant to the point of indecency. MSTR stock cost more than the Bitcoin standing behind it — at the peak, nearly $3.9 of market price for every $1 of Bitcoin in the vault (November 2024). Almost four times over. The company took that overpriced share, sold it on the market at full price, and used the proceeds to buy more Bitcoin — and the premium from this didn't fall, it grew, because the market watched the pile rise and applauded. More Bitcoin per share → higher premium → cheaper capital → more Bitcoin. The wheel spun itself and accelerated with every turn. A perpetual-motion machine. The first in modern financial history that, instead of violating the laws of thermodynamics, violated only the investor's common sense — and that, it turned out, is a far softer constraint.
\n\nFor three centuries engineers submitted blueprints for perpetual-motion machines to patent offices — wheels that push themselves, pumps that lift water by their own weight. In 1775 the French Academy of Sciences simply stopped considering them: by official resolution it declared it would no longer accept applications for perpetuum mobile, because not one in all of history had passed a single test — the measurement of friction. A perpetual-motion machine always spins exactly as long as nobody has put a meter to it: in physics the meter is called friction, in finance it's the premium to the asset's value, and both relentlessly drag the wheel back toward one. MicroStrategy built the most beautiful wheel of the decade. And then someone walked up and looked at the reading.
\n\nThis text isn't about Michael Saylor being stupid. He did everything right within the rules of the game: to assemble a machine that legally turns $1 into $1.20 and runs for six years is not stupidity, it's the craft of an illusionist of the highest order. It's just that the rules of the game don't ask how charismatic you are once the needle approaches one. The text is about something else: about what happens to any such wheel when the market finally learns arithmetic. The market always learns arithmetic. It just sometimes lets you read a couple of books on \"the philosophy of money\" first and feel smarter than the meter.
\n\n\n\nLet's give the machine an honest name, because it'll appear in every section from here. Call it the premium→parity spiral: a mechanism in which an asset trades above its intrinsic contents, that premium feeds itself on the way up — and the very same premium kills itself on the way down, because everything that spun the wheel one way spins it back with equal force.
\n\nGoing up, the spiral looks like genius. Premium above one → issuing shares adds Bitcoin per share → the pile grows → the investor sees growth and pays an even higher premium → capital gets cheaper → more Bitcoin. Each turn makes the next one easier. In corporate filings this is beautifully called \"accretive.\" It sounds like an accounting term and works like a slide: the faster you go, the faster you accelerate. The slide up is the most pleasant ride. But nature doesn't build one-way slides.
\n\nNow the same scheme going down, and here there are no pretty words left. The premium slides toward one → issuing shares no longer adds Bitcoin per share but takes it away (sell a cheap share — dilute existing holders) → the model, to do no harm, must stop issuing → without issuance there are no new purchases → the very demand that held up the premium disappears → the premium falls further still. The same inertia. The opposite sign. There was no \"exit on a flat stretch\" button in any prospectus — it simply keeps going down until it hits the ground.
\n\n| Loop link | Spiral up (premium > 1) | Spiral down (premium → 1) |
|---|---|---|
| Share issuance | Adds Bitcoin per share — holders get richer | Takes Bitcoin per share — holders get diluted |
| Management response | Print more: every turn is accretive | Stop printing, or you harm your own |
| Company demand for Bitcoin | Grows with each issuance | Vanishes along with the issuance |
| Narrative | \"Eternal growth, Bitcoin standard\" | \"Why is the share below its own Bitcoin?\" |
| Next day's premium | Rises — faith feeds faith | Falls — fear feeds fear |
| What it's called out loud | The founder's genius | The same word, whispered and in past tense |
And the strangest part: in this construction there's no point of rest in the middle. The wheel is stable only at the very top (a high premium feeds itself) or at the very bottom (parity). The entire middle is a slope. It's the same middle where you convince yourself \"well, it dropped a bit, but the fundamentals are strong\" — and \"dropped a bit\" here isn't a stage, it's a slide down the cobblestones toward one. Parity isn't an accident or a management failure. Parity is the floor the spiral structurally strives for the moment the premium is removed. The question was never \"will it get there,\" but \"when will someone approach the meter first.\"
\n\nIn the filings these are two different \"cycles.\" In life — the same slide: going up you scream \"more!\", going down \"stop!\". Spoiler: there was no \"stop\" button in any prospectus. In its place was a line in fine print: \"past performance does not guarantee future performance.\" But who reads the fine print while the wheel spins up?
\n\nThe meter was put to the machine not by a regulator, nor by a visionary short-seller in a sweater. It was put there by the most boring financial product in the world — the exchange-traded fund. It didn't take the stage in a black T-shirt. It didn't promise you'd \"become a co-owner of humanity's digital gold.\" It gave no interviews about \"an upgrade to the planet's money.\" It just quietly clicked Bitcoins through the cash register — and that's why it took all the money.
\n\nBefore January 2024, buying Bitcoin into a brokerage account without touching exchanges and wallets was inconvenient. MicroStrategy sold exactly this convenience: \"want Bitcoin in a normal portfolio — buy our stock, we're the wrapper.\" For the wrapper you paid a premium. And then the SEC approved spot Bitcoin ETFs, and a dozen funds began selling the same wrapper — without debt, without preferred shares, without a charismatic founder on podcasts, without leverage that cuts both ways. Just Bitcoin in your account, one to one, for a fee of a few hundredths of a percent.
\n\nThe ETF doesn't promise you dinner with the messiah and doesn't walk onstage with a thesis about a \"new monetary standard.\" It just honestly promises: \"you'll pay us 0.2% a year, and we won't try to become a meme on Twitter.\" And the market chooses the boring one. Spot Bitcoin ETFs gathered over $100 billion in assets — one of the fastest launches of a new fund class in the industry's history (per The Block and Bloomberg). BlackRock's iShares Bitcoin Trust alone (ticker IBIT) grew to roughly $66 billion under management, and cumulative net inflows since launch passed $65 billion. That whole wall of money went where Bitcoin costs exactly what Bitcoin costs — without a markup for the show.
\n\nThe ETF is like the invention of the mirror that ends the trade of the court painter who, for a separate fee, painted your portrait but flattered the chin too. Why pay a master for a flattering image when an honest mirror hangs in every shop window? The premium for the \"wrapper\" didn't fall dramatically. It simply stopped having a reason to exist — and the market, unlike a man in love with himself, notices very fast when the reason has gone. The spot ETF is a black hole of liquidity: everything within a certain orbit is inevitably pulled to the center, where Bitcoin costs exactly what it costs. No surcharge for the founder's passion.
\n\n\n\n*The painter painted the portrait with a flattered chin — for a 3.89x premium. Then the janitor wheeled in a rack of mirrors at 1.00x. The painter froze with his brush. The market chose the boring mirror.*\n\nLet's drop to elementary-school level, because that's exactly where all the magic lives — and all the deception. The scheme mathematically requires one single thing: that the premium be greater than one. Not \"preferably.\" It requires it, the way an internal-combustion engine requires that something burn in the cylinder.
\n\nLet's count on our fingers. Suppose MSTR shares together cost, say, $1.5 for every $1 of Bitcoin in the vault (a 1.5x premium). The company issues a new share, sells it at market, and uses the proceeds to buy Bitcoin. Since it was paid one and a half times more for the share than the Bitcoin bought with that money costs, every old shareholder, after the operation, owns a larger slice of Bitcoin than before. Issuing new shares makes the old ones richer. This is the trick the room always gasps at — and then writes threads \"no one could have predicted.\" There is exactly one thing to predict here: the \">\" sign in the schoolbook inequality \"premium > 1.\"
\n\nNow remove the premium. Shares cost exactly $1 per $1 of Bitcoin (parity, 1.0x). Issuing a share brings in exactly as much money as the Bitcoin bought with it costs. The old shareholder gained nothing: their slice was diluted by the new share by exactly as much as Bitcoin was added. Zero. And if the premium has sagged below one — every issuance now robs the old holders in favor of the new. Below parity the model begins spinning in reverse and threshing its own passengers. Your \"I'm averaging down, and the average price is falling\" is the same arithmetic, just without press releases and a Bloomberg name.
\n\n| Premium (mNAV) | What issuance does | Model state | The unvarnished bottom line |
|---|---|---|---|
| ~3.9x (peak, Nov 2024) | $1 of stock → ~$3.9 of Bitcoin to holders | Acceleration | Alchemy works, the room gives a standing ovation |
| ~1.9x (interim) | $1 of stock → ~$1.9 of Bitcoin | Inertia | Still accretive, but you can hear the wheel |
| ~1.24x (May 2026) | $1 of stock → ~$1.24 of Bitcoin | Braking | The magic has nearly worn down to arithmetic |
| 1.0x (parity) | $1 of stock → exactly $1 of Bitcoin | Stop | No more point in printing shares |
| ~0.94x (a moment, May 2026) | $1 of stock → less than $1 of Bitcoin | Reverse | Every issuance robs its own — the model eats its passengers |
This is why the company so dislikes being asked to say this arithmetic out loud. All the rhetoric is about the \"Bitcoin standard,\" about \"buying forever,\" about a vision for decades. Not a word about the fact that the whole construction rests on a single schoolbook inequality: premium > 1. The premium is the oxygen in the pressure chamber: above one, everyone walks around with rosy cheeks. At parity the air is cut: you can still joke, but not for long. It's like an illusionist who talks for an hour about the magic of the universe so the audience won't notice the little mirror under the table. The little mirror is called mNAV. And it just showed 0.94.
\n\n\n\nNow to the real sizes, because abstraction is merciful and numbers are not. As of May 2026, Strategy holds 843,738 Bitcoin, bought for roughly $63.9 billion — the largest corporate Bitcoin treasury on the planet, next to which every other company looks like a child who brought pocket money to the casino. The machine worked. The pile was gathered. This isn't a bankruptcy simulation — it's a living mountain of money.
\n\n843,738 coins is like a credit card with an enormous limit that you've sworn never to take out of the safe. Technically you have money. Practically — no. The whole mountain makes sense as collateral only under one condition: as long as you don't have to sell it. The mantra \"we never sell\" isn't Zen philosophy but engineering necessity: the moment the world's largest Bitcoin holder starts selling, it crashes the price of what it's selling with its own sale — i.e., saws off the branch the whole mountain hangs from. So 843,000 coins isn't liquidity. It's a decoration of strength that stops being strength the second it has to be converted to cash.
\n\nAnd cash will someday be needed. The preferred shares that financed the buying demand dividends — about $640 million a year in hard money (BeInCrypto's estimate based on Strategy's offering prospectuses), regardless of Bitcoin's mood. These $640 million a year aren't an abstraction. They're people in suits whose phone rings exactly once a year — to remind you it's time to collect the cash. They don't listen to podcasts about \"never selling.\" They read the payment schedule. Bitcoin pays no dividends — it just lies there and glows.
\n\nOn top of that, the company faces the risk of being kicked out of stock indices: by BeInCrypto's estimate, exclusion could pull $2.8 to $8.8 billion of passive money out. A passive fund doesn't argue with your vision. Its Excel sheet just got updated, and it sold you without taking off its headphones. With no emotion toward the narrative whatsoever.
\n\n\n\nAnd the quietest trap — the convertible bonds. Part of the treasury was built on debt that can be repaid either in money or in shares — provided the share rises to a certain price. The issues line up for 2028, 2029, and 2030. The conversion price on one series is around $672 per share (exact: $672.40). In May 2026 the share traded around $159. \"Painfully below\" is minus 75%, not a poetic metaphor. If it stays there until maturity, the debt will have to be repaid not in paper but in hard cash. Convertible bonds are like an army calmly camped under the city walls. Year after year. The townsfolk joke about it — until the grain runs out. The creditor on a convertible bond is the most polite of all. He stays silent for years, doesn't call, doesn't remind. He just one day comes due and politely asks whether it'll be in paper, or whether you'll have to open the safe after all.
\n\n\n\n*The founder poses at the summit. On the branch that holds it all sits the janitor with a handsaw and a look of \"I just get paid to clean.\" Each creditor in the doorway has his watch set to his own maturity date. They're in no hurry — time works for them, not for Bitcoin.*\n\nThe thing we kept calling a wheel has a precise name in market theory — and it was given by a man who made billions on it. George Soros called it reflexivity: market prices don't just reflect reality, they change it, and the changed reality changes prices — and the loop can spin itself, detached from any \"fair\" level. What's important and usually left out of the quote: Soros described reflexivity not as an engine of growth but as a mechanism of boom and bust. A loop that reinforces itself upward, by its very nature reinforces itself downward. He got rich precisely on that — betting against loops that had already begun to reverse.
\n\nApply it to our machine. Up: high share price → cheap capital → more Bitcoin → even higher price. A reflexive boom, faith feeding itself. Reflexivity is a broken pipeline: every new price \"release\" automatically ships into production a change in reality that, in the next cycle, breaks the system even harder. And when the pipeline reverses: the share price falls → the premium compresses → issuance stops → the company's demand vanishes → the narrative cracks → the share falls more.
\n\nThe next time you write in a chat \"it's just flying, no point overthinking it\" — that is your private version of reflexivity. The only difference is that Soros made money on that phrase, and you'll dump your portfolio into the red on it. Whoever said \"it works by itself\" on the way up must find the courage to finish the sentence on the way down: yes, it still works by itself. It's just that now \"it\" is the fall.
\n\nReality is the worst creditor: it too never calls, it just one day sits down at your desk and asks for the bill for all the turns of the wheel you happily booked as an asset while it spun the right way. And here there's a black joke no screenwriter would dare invent: a machine built on pure reflexivity will most likely be finished off by the very mechanic invented and described by the man who personally got rich on it. As a private lesson: a loop you mounted on the way up cannot be stopped on the way down by willpower.
\n\nNot every high-stakes bet is a swindle. There's a whole category of situations where concentrated risk on a single asset turns out, in hindsight, to be a good decision: those who bought Amazon in 2001 during the dot-com crash, those who held Apple after it nearly went bankrupt in 1997. There's also a substantive argument in Saylor's defense you can't wave away in one line. If Bitcoin really reaches $500,000, the mNAV returns to the 2x–3x zone organically — and all this analysis about \"the meter\" will look like midway panic. A Strategy trading below parity in 2026 may, in 2029, look like an implausibly cheap entry point if even half the forecasts about institutional mass adoption come true. Saylor doesn't hold Bitcoin because he doesn't understand the risk — he holds it because he believes the 10x scenario is more likely than zero. That's not an irrational position. In some Monte Carlo simulations it even wins.
\n\nBut here's what this argument doesn't explain: who has the right to this bet, and who doesn't. Saylor personally made billions on MSTR stock — and if the bet loses, he stays a billionaire with a smaller account. A retail investor who bought MSTR in November 2024 at $500 with an mNAV of 3.9x cannot afford either the patience or the temporary loss of −58% y/y. A rational bet on 10x looks the same in both directions only if you can survive the road. Preferred shareholders, convertible creditors, and passive index funds don't bet on Saylor's long-term convictions — they read the prospectus with the maturity date. So \"Saylor may be right\" isn't a refutation of the analysis. It's just a clarification: right he'll be — for himself. For the rest of the line, the question isn't whether he's right — it's whether they have the time and capital to wait for it to materialize.
\n\nThe machine may live a long time yet: Bitcoin can rise and temporarily flood all the cracks, the premium can squeeze back up on a new narrative, management can restructure the debt. The premium→parity spiral isn't a death sentence with a date — it's a lens through which you can see what to watch. So instead of a forecast — an honest instrument. Here are the readings by which anyone can read the meter themselves — without asking the inventor's permission:
\n\nNone of these readings require insider access or a gift of prophecy. They're all on the meter bolted to the base of the machine, right where no one riding it wants to see it. If the machine has no meter — then you'll be the meter. And the show will be paid for with you.
\n\n\n\n*The meter bolted to the base. The needle creeps toward one. In the cracked glass the slowing wheel is reflected. Beside it a sticker, \"Ignore,\" worn smooth with fingerprints. One finger finally lifted off the eyes.*\n\nWhen the machine spun, the money didn't appear from nothing — it was redistributed. Any flywheel with reflexive pricing has its distribution matrix. Here it is, unadorned:
\n\n| Who | What they got / paid | Note |
|---|---|---|
| Early MSTR shareholders (pre-2023) | Got: real share-price growth; every issuance at 2–3x increased their Bitcoin per share | The flywheel is a winner if you got in and out before the premium compressed |
| Late retail buyers (Nov 2024 — May 2026) | Paid: bought the wrapper at 3.9x — 1.9x and now hold a position near parity or below; −58% y/y despite Bitcoin's own rise | Paid a premium for a narrative the ETF no longer supported |
| Institutional convertible-bond holders | Got: coupons and conversion rights; priority over shareholders if repaid in cash | Their money financed the Bitcoin buying; in the losing scenario they collect first |
| Spot-ETF holders (IBIT et al.) | Got: clean Bitcoin exposure for ~0.2% fee — no leverage, no premium to NAV | Took the \"institutional\" buyers who once gave MSTR its premium |
The flywheel isn't a fraud in the criminal sense — it's legal, public, and documented in detail in the prospectuses. But the distributive mechanics are unambiguous: every dollar of premium paid by a late retail buyer for a share above NAV directly financed the issuance and purchase of new Bitcoin for the benefit of those who held earlier. The scheme requires neither malice nor deception — it only needs a line of people willing to pay a premium for the narrative. The line existed. It still stands today.
\n\nThe French Academy in 1775 didn't declare perpetual-motion machines forbidden — it declared them boring. It stopped spending sessions on them, because it knew in advance: however beautiful the blueprint, somewhere in it friction hides, and the wheel will stop. Not \"may stop.\" Will stop. The only question is on which turn someone thinks to raise the measuring device.
\n\nMicroStrategy built the grandest financial perpetuum mobile of its time — a scheme that for six years honestly and legally turned $1 of stock into $1.20, $1.50, at the peak nearly $3.9 of Bitcoin, and a whole generation of investors believed that this time friction had been abolished. Friction turned out to be the premium. The meter — the spot ETF that commoditized the wrapper. The hand that raised the device — the boring market that, in the end, always learns arithmetic, even if it needs a few extra years of euphoria. The ~3.9x premium rode down to ~1.24x and for a moment touched 0.94x. The wheel still turns by inertia. But inertia is not an engine.
\n\nSo the next time you're shown a machine that supposedly makes money from nothing — a fund with \"impossible\" returns, a token that \"only goes up,\" a business whose model rests on the paper about it being worth more than everything behind it — don't look in the room for the fool. If you buy paper above what stands behind it, hoping to exit first — the fool is in the mirror. Ask the one boring question that crumbles perpetual-motion machines into scrap metal: where's the meter here — and why is everyone in the room so diligently not looking at it?
\n\n\n\t\n\nA perpetual-motion machine doesn't lie about physics — it stays silent about the meter. Every time, someone in grey overalls walks up late at night with a flashlight, reads the gauge, and sees that the needle has long since been creeping toward one. The alchemy that turns a dollar of stock into a dollar twenty of Bitcoin works exactly as long as the market hasn't learned arithmetic. The market always learns arithmetic. It just sometimes lets you read a couple of books on \"the philosophy of money\" first and feel smarter than the meter.
\n