The $186.8 Million Gap in American Bitcoin’s Filing Is the Real Story

On June 30, American Bitcoin’s balance sheet carried 3,090 pledged Bitcoin at $184.9 million. The same filing attached a liability of $371.7 million to that same pledge. That spread—$186.8 million—is not a typo, and it is not the kind of accounting noise you can dismiss with a footnote. Decoding the signal hidden in the noise, it is the most important number in the entire Q2 report. The filing is built around a hardware deal with Bitmain: 11,298 mining machines, $49.4 million in value, and an unusual payment mechanism. American Bitcoin did not pay cash. It pledged Bitcoin. Today, 38.6% of its treasury sits in Bitmain’s hands, and the two companies are locked in a 24-month window that will end in an either/or settlement. Either American Bitcoin pays cash to get its coins back, or it lets the coins go and treats them as the cost of the machines. You know American Bitcoin by its founder’s last name. Eric Trump is co-founder and Chief Strategy Officer; Donald Trump Jr. is involved. The company is an 80%-owned subsidiary of Hut 8, one of North America’s most established publicly listed mining operators. That combination—Trump-branded, Hut 8-operated, publicly disclosed—makes this one of the most closely watched balance sheets in mining. The Q2 filing, which CryptoSlate reviewed, is the first hard look inside the machine. The headline numbers tell a familiar bear-market story. Bitcoin is roughly 50% below its October 2025 peak. American Bitcoin recorded a GAAP loss of $57.2 million, including a $71.2 million digital asset impairment. It sold stock through an at-the-market offering to raise $33.6 million, diluting shareholders by about 3%. On paper, this looks like a miner bleeding in a downturn. But the filing also shows something that cuts against that narrative: total Bitcoin holdings grew 14% quarter-over-quarter, to 8,002 BTC. Per-share sats—the MicroStrategy-style metric that matters to treasury-focused investors—rose 11%. In the middle of a brutal drawdown, American Bitcoin is accumulating. The question is at what cost, and who is really holding the risk. The structure of the Bitmain deal is where the truth pools. When I say “pledge,” I do not mean a loan in the traditional sense. American Bitcoin delivered 3,090 BTC to Bitmain to secure delivery of the miners. The contract gives American Bitcoin a redemption right: within a roughly 24-month window, it can pay a fixed cash amount and get the Bitcoin back. If it does not, the Bitcoin is applied to the purchase price at a contractually agreed “floor valuation.” That floor, based on the liability number, implies a value of roughly $120,000 per BTC. Here is the trap. Bitcoin is trading near $60,000. The same 3,090 BTC are worth $184.9 million on the open market. But the liability on the books is $371.7 million—the value of those coins at the contract’s floor price. That is not a mistake. That is a deeply out-of-the-money redemption strike. Any rational treasurer looks at that gap and concludes: we will never pay $371.7 million to get back coins worth $184.9 million. Unless Bitcoin rallies dramatically before the window closes, American Bitcoin will let the pledge settle. This is the part most coverage misses. The market sees a miner pledging 38.6% of its treasury to buy machines and screams “forced selling.” But look at the direction of the incentive. American Bitcoin is not selling at the market price. It is handing Bitmain coins that will be credited at a $120,000 floor. If the contract values each BTC at $120k for settlement, then American Bitcoin is paying for the $49.4 million miner fleet with a fraction of the pledged BTC—perhaps 400 coins, not 3,090. In that scenario, the remaining pledged BTC should, in economic terms, come back. But under the accounting construct, the entire package is reported as a $371.7 million liability. That is the optionality hidden in the books. The 24-month window is not a countdown to liquidation. It is a call option on Bitcoin. If BTC rips past the strike, American Bitcoin pays cash, keeps its coins, and the $371.7 million liability disappears. If BTC stays below, it hands over Bitcoin that Bitmain values at $120k each, which means the physical settlement is far less painful than the current mark suggests. The accounting is misleading precisely because the economics are non-linear. The $186.8 million gap between the market value of the collateral and the stated liability is the expected value of that optionality. It is a convex position built into a miner’s balance sheet. In my years auditing crypto treasuries, I have seen many companies avoid selling Bitcoin by borrowing against it. Most of those structures have a liquidation price—a line in the sand that forces a sale. American Bitcoin’s structure has no forced liquidation. It has a very high strike price, a long maturity, and a counterparty that already holds the collateral. That is not a weak position in a bear market; it is a strategically leveraged one, provided you believe Bitcoin has any chance of recovering within two years. Now follow the money flow beneath the headline numbers. American Bitcoin raised $33.6 million through an ATM program. That is small for a company with 8,002 BTC, but the dilution was only 3%, meaning the company effectively bought its own per-share Bitcoin density at a very low cost. It used that cash to fund operations, pay down expenses, or hold dry powder for redemption decisions. The treasury grew 14% even with the Bitmain pledge, because the company kept mining and kept accumulating with its free cash flow. The loss is largely non-cash write-downs on the BTC it already holds. The impairment charge is an accounting artifact of falling prices, not a cash burn. Compare this to the rest of the sector. MARA Holdings still holds tens of thousands of BTC and is the largest public miner, but its per-share sats growth has been diluted by repeated raises. Riot Platforms has strong power infrastructure but a smaller treasury relative to market cap. Bitdeer builds its own miners, giving it a cost advantage, but it does not carry the same political brand access. American Bitcoin is not the biggest, not the cheapest, and not the most technically independent. Its edge is structural: an 80% parent that knows how to operate mines, a name that opens doors in Washington, and a balance sheet designed to survive a prolonged bear market. Let me give you the contrarian angle, because I think the market is asking the wrong question. The question is not “will American Bitcoin lose its Bitcoin?” The question is “what is the implied strike price of the deal, and where does Bitcoin need to be for the company to redeem?” Based on the liability of $371.7 million against 3,090 BTC, the implied strike is about $120,000. If Bitcoin is below that at maturity, the rational decision is to let Bitmain keep the coins. That is not capitulation. That is a pre-planned exit price. The company effectively sold Bitcoin at $120,000 via a collateralized equipment purchase, while also getting the miners to keep generating revenue in the meantime. If the market is right today, American Bitcoin has locked in a future sale at roughly double the current spot price. If the market is wrong, and Bitcoin recovers, the company can redeem in cash and keep its coins. That is not the behavior of a distressed miner. That is the behavior of a treasury desk that has built a covered call around its mining capex. The second contrarian insight is about the per-share sats metric. In a bear market, analysts anchor on GAAP losses and ignore the balance-sheet density. A 14% increase in BTC holdings with only 3% dilution means per-share sats grew 11% in one quarter. If you value American Bitcoin as a treasury company, it is compounding its Bitcoin per share faster than most of its peers, even while the market prices it as a failing miner. The filing is a signal that the company is no longer just a miner. It is a Bitcoin accumulation vehicle with a mining hedge attached. Bubbles burst, but architecture remains. The architecture here is a 24-month window that forces a decision, and that decision will define the company’s next cycle. There are real risks, and I will not bury them. Bitmain is a single point of failure for equipment delivery, and China-U.S. export controls remain an unpredictable variable. The Trump brand cuts both ways: it provides media access and potential political capital, but it also invites scrutiny that other miners do not face. A politically exposed person in a crypto company is a gift to hostile regulators and a target for congressional investigations. The contract’s floor valuation is not public, so the $120k implied strike is an inference from the disclosed liability. If the actual floor is lower, the economics shift. And if Bitcoin stays depressed through 2027 and 2028, the redemption windows will expire one by one, and the market will see the promised coins hit Bitmain’s balance sheet. The pain is avoidable only if price cooperates. Still, I keep coming back to the gap. A $186.8 million difference between the market value of pledged assets and the reported liability is not a footnote. It is the market being told, in the most technical language possible, that this balance sheet contains a bet. The company is long volatility, long Bitcoin, and long optionality. The quarterly loss is the fee for holding that option. The redemption window is the expiry date. And the 11,298 miners are the income-generating collateral that funds the wait. So ignore the panic about 3,090 BTC sitting in Bitmain’s custody. That Bitcoin is not lost. It is collateral for a structured trade that most analysts have not modeled. The real disclosure, hidden in the gap between fair value and liability, is that American Bitcoin has effectively written a forward sale at $120,000 per coin—and kept the right to buy that forward back if Bitcoin exceeds expectations. That is not a miner’s playbook. That is an options desk wearing a miner’s helmet. Three years from now, we will look back at this filing and know exactly when the decision was made. The clue is not in the loss line, the impairment charge, or the ATM raise. It is in the spread between what the coins are worth and what the contract says they are worth. Follow that gap, and you will see the shape of the next cycle before the price chart confirms it. Where liquidity flows, truth eventually pools—and right now, the liquidity is flowing toward a bet that Bitcoin will eventually be worth more than the strike price printed in a liability line. The last page of the Q2 report is not a summary. It is a setup. American Bitcoin is not betting on the next block reward; it is betting on the next bull market. And it has built a balance sheet that will reward that bet if it hits, and penalize it only if Bitcoin stays dead for two more years. Look at the redemption windows. Watch the Bitcoin price at each one. The company will not tell you what it is doing. The accounting already has.