The headline number is $2.8 billion. I spent an afternoon trying to make it reconcile with the estate, and it does not.
Mt. Gox's court-appointed trustee, Nobuaki Kobayashi, has moved the repayment deadline — previously October 31, 2025 — out to 2027, citing creditors who have not completed the procedures required to receive their coins. That is the fact. Everything layered on top of it in the coverage is inference wearing a reporter's jacket. The estate once held roughly 200,000 BTC, of which about 140,000 remained for distribution when the case entered civil rehabilitation. At any price between $60,000 and $110,000, $2.8 billion converts to somewhere between 25,000 and 47,000 coins. That leaves 90,000 to 115,000 BTC unaccounted for, and nobody has published the reconciliation.
The ledger never lies, only the narrative obscures.
Some groundwork, because the coverage assumes you already know it. Mt. Gox was the largest bitcoin exchange on earth until February 2014, when it halted withdrawals and collapsed into Japanese bankruptcy proceedings. In 2018 the case converted to civil rehabilitation — a restructuring track whose purpose is to return assets in kind rather than liquidate them at bankruptcy-era prices. The trustee is court-appointed, supervised by the Tokyo District Court, and answerable to a creditors' meeting. The 2021 rehabilitation plan passed with overwhelming creditor support. Actual distribution began in 2024 through an early lump-sum payment window. The prior public deadline was October 31, 2025.
Which means the $2.8 billion figure has exactly two plausible readings. Either it represents what remains after the 2024 early-payment tranche — a residual balance — or it counts a single asset class, presumably BTC, excluding BCH and cash. Those readings carry opposite market meanings: a residual means most of the overhang has already been absorbed; a single-asset reading means the overhang is still substantially intact. The reporting does not distinguish. That omission is the entire information content of the story, and it was dropped.
Run the arithmetic both ways. The network issues roughly 450 BTC per day after the halving. At the pessimistic $60,000 mark, $2.8 billion is about 47,000 coins — 104 days of global new issuance. At $110,000 it is about 25,000 coins — 56 days. Even the harsh reading is one to three and a half months of miner output, and against spot ETF flows that routinely clear thousands of coins a day in either direction, that is a rounding error in the marginal price mechanism.
| Reading of $2.8B | Implied BTC | Days of new issuance | |------------------|-------------|----------------------| | Residual after early payments | 25,000–47,000 | 56–104 | | Full estate | ~140,000 | ~311 (inconsistent with $2.8B) |
The bottleneck is not on-chain. It never was. In a civil rehabilitation, a creditor must file a claim, open an account at a designated exchange — historically Kraken, Bitstamp, BitGo — clear that venue's KYC, and confirm a receiving address. Throughput on that path is set by manual review and per-jurisdiction identity standards, not by block space. The trustee is not waiting on confirmations; he is waiting on humans. That distinction matters because it makes the delay structurally inevitable rather than malicious. It also means the assets are safe and movable, which quietly rules out the worst-case hidden story: there is no missing-coin headline waiting.
Distribution is in kind. The 2017–2018 pattern of the trustee selling on the open market is gone; creditors receive coins. So the sell pressure is dispersed across thousands of independent decisions over months rather than concentrated in a single session. Structurally that prevents a flash crash. It also stretches the uncertainty indefinitely, which is a worse outcome for anyone trying to price the risk.
One omission deserves its own paragraph. BCH was distributed alongside BTC. BCH order books are far thinner than bitcoin's, so the same dollar notional of BCH supply moves its price several times harder. The coverage says nothing about BCH. That is not a minor gap; it is the part of the event where the actual percentage damage lives.
The other blind spot is who the creditors actually are. For years, specialist funds bought Mt. Gox claims at a discount to face value. Those holders have a lower nominal cost basis than the original depositors but a hard exit horizon, and their willingness to sell is materially higher than the retail narrative assumes. Twelve years is also a generational turnover: original users have lost keys, changed emails, died. Calling the recipient base "long-term holders" is a comfortable assumption with no evidentiary support. Based on my 2022 Terra work — where Anchor withdrawal patterns preceded the headline by weeks — I would trust the composition of the claimant list over any characterization of it.
The tradable signal is not the announcement. It is the UTXO. In 2024, transfers out of trustee-marked addresses to exchange deposit addresses preceded the press coverage. An algorithm does not sleep, nor does it feel fear — and a wallet balance does not issue clarifications. When I built the ETF flow dashboard last year, the only discipline that mattered was reconciling every flow against a known denominator. Here, nobody has published the denominator, which is precisely why the headline is unpriceable.
Here is where the consensus breaks. The popular reading is: delay equals bullish, because selling pressure is postponed. Correlation is a suggestion; causality is a truth. What is actually observable is a pattern of deadline creep — 2024, then October 2025, now 2027 — where each extension manufactures the precedent and the psychological permission for the next. The trustee's incentive is procedural legitimacy, not speed. As long as a defensible KYC backlog exists, accelerating carries downside for him and no upside. The formal extension also functions as a liability shield, protecting the trustee from claims of failure to distribute on schedule. His interests and the creditors' interests are adjacent, not identical.
The corollary is that "70 to 80 percent priced in" is an unfalsifiable claim. The only falsifiable version of the thesis is wallet movement, and that is observable.
Watch the marked addresses through the extension window. If coins begin moving to exchange deposit addresses while the 2027 date sits on the calendar, the date is decorative and the overhang is live. If the wallets stay cold, the residual reading holds and the story is administrative. Either way, the next signal will not arrive as a press release. Trust the hash, not the headline.