Block height 961,632 is not a halving. It is not a difficulty adjustment. It is, according to a small coalition of Bitcoin node operators, the moment when software supporting BIP-110 begins rejecting blocks that contain transactions carrying more than eighty bytes of OP_RETURN payload. Developer Kevin Loaec has framed the stakes in the starkest terms available: users who attempt to sell the resulting fork token may watch their real Bitcoin migrate to a stranger's wallet via a transaction replayed on the main chain.
Here is the number that should anchor the entire conversation: 2.6 percent.
That is the share of miners currently signaling support for BIP-110. Not 95 percent, which is the BIP9 activation threshold. Not even 55 percent, the informal economic majority level that dictates policy enforcement viability. Two point six percent. The ledger does not lie, only the narrative does. And the narrative assembling around this would-be fork โ one of stolen BTC, chaotic chain splits, and ordinary users devastated by a technicality โ needs to be audited against the raw data before anyone moves a satoshi. Certified eyes, unfiltered truth in the blockchain.
The warning is technically real. The mechanism is technically real. But the probability distribution has been omitted from the headline.
Context: What BIP-110 Actually Changes
BIP-110 is not a consensus overhaul. It is not a new proof mechanism. It is not a scalability breakthrough. It is, at its core, a standards policy adjustment: a proposal to constrain the amount of non-payment data that can be embedded in a Bitcoin transaction through OP_RETURN outputs.
The history carries weight. OP_RETURN outputs were effectively disabled in Bitcoin's early years because they created provably unspendable UTXOs โ database entries that would bloat the chain forever with no possibility of cleanup. In 2014, developers began permitting small amounts of metadata onto the chain through a standardness rule change. The eighty-byte limit became the de facto standard for data-bearing transactions. Services like Veriblock and OmniLayer built on this allowance. The question of whether Bitcoin should carry arbitrary data โ images, text, timestamps, token metadata โ has simmered in protocol discussions for over a decade.
The current BIP-110 proposal seeks to enforce a stricter interpretation with meaningful teeth. Software supporting the proposal is expected to reject blocks containing OP_RETURN outputs that exceed the threshold, beginning at block height 961,632. This is not a user-activated soft fork in the SegWit tradition. It is not a miner-activated soft fork under BIP9's signaling regimen. It is a policy divergence executed through software versioning: nodes that adopt the rule and nodes that do not will, at a specific height, disagree about which blocks are valid.
That disagreement produces two chains. And the two chains share everything โ the complete UTXO set, the entire transaction history, every signature, every address balance. This is the inheritance rule of chain splits, and it is the root cause of the risk Kevin Loaec has spent his warning budget on.
A fork without replay protection means a transaction signed on one chain is perfectly valid on the other. The transaction format carries no chain identifier. The cryptographic signatures do not know which network they belong to. The only distinction between the two chains is which set of miners chooses to include the transaction in a block. Everything else is identical. That sameness is the vulnerability.
Notably, mapping current block production to height 961,632 suggests a target window in late 2025 or early 2026 โ not the imminent horizon that panic-driven headlines imply. The discrepancy between the alarm's urgency and the protocol's actual timeline is itself a data point.
The Replay Mechanism: A Step-by-Step Autopsy
The casual summary โ "sell fork coin, lose real BTC" โ obscures the mechanical elegance of the attack. Walk the exploit scenario with the precision it deserves.
Step one: The fork materializes at block height 961,632. The minority chain, sustained by the 2.6 percent hashrate coalition, continues producing blocks. Every UTXO that exists on the main chain at that instant exists on the fork chain. The fork chain inherits the full ledger state, including all private key ownership.
Step two: A user notices they hold "free" fork coins. An exchange that has chosen to credit accounts with the fork token, or a wallet that has imported the fork chain's state, presents these coins as an available balance. The user decides to sell. Human psychology completes the rest of the attack before any code executes.
Step three: The user signs a transaction on the fork chain, transferring their fork coins to a buyer or a trading venue. The transaction carries a cryptographic signature generated by the user's private key. It spends a UTXO that exists on both chains. For the user, this is a sale of speculative tokens. For the protocol, this is a signed authorization to move a specific input.
Step four: The buyer โ or any observer who sees the transaction in the fork chain's mempool โ takes the exact bytes of that transaction and broadcasts them to the main Bitcoin network. The transaction validates on the main chain. It spends the identical UTXO, transferring the user's real BTC to the buyer's address. The user's actual Bitcoin is gone. There is no recourse. The transaction is immutable. The signature was self-generated. The chain does not care about intent.
This is not a theoretical edge case. It is the canonical replay attack, documented in every serious analysis of unprotected chain splits. In my own forensic work tracing the 2022 DeFi collapse, I built causal graphs showing how a single unguarded interaction could trigger a cascade of liquidations across interconnected protocols. The mechanism here is simpler and more direct: one transaction, two chains, zero protections.
Kevin Loaec's advice to hold assets motionless is not timidity. It is the mathematically complete set of safe actions. During an unprotected fork, the only transaction that cannot be replayed is the transaction that is never signed. The unspent UTXO is a cryptographic strongbox. The moment it becomes an input to a signed transaction, it becomes a liability on every chain that shares the same state.
The Hashrate Survival Analysis: What 2.6% Actually Means
Interrogate the survival probability of the fork chain. The numbers are brutal.
Bitcoin's global hashrate at this juncture sits in the hundreds of exahashes per second. The 2.6 percent signaling figure translates to a theoretical allocation of roughly six to eight exahashes per second for the fork chain โ assuming every signaling miner follows through. That level of hashrate produces a block, on average, every five to six hours. Not every ten minutes. Every five to six hours.
The implications compound relentlessly.
Confirmation times become absurd. A merchant awaiting three confirmations on the fork chain waits the better part of a day. Users abandon the chain not out of ideological disagreement but out of simple unusability. The chain becomes a museum artifact rather than a payments network.
The fork chain is trivially vulnerable to 51% attacks. The main network's hashrate is roughly thirty-eight times larger. Any hostile miner can redirect a slice of their power to the fork chain and reorganize it at will. Double-spend attacks against any venue that lists the fork token become not merely possible but probabilistically inevitable.
The difficulty adjustment mechanism becomes a death spiral. Bitcoin adjusts difficulty every 2016 blocks. On the fork chain, at five to six hours per block, that adjustment arrives after four hundred to five hundred days โ if the chain survives that long. Until the adjustment arrives, the fork chain carries main-chain difficulty with a fraction of the hashrate. Block times stretch into tens of hours. The chain becomes, for all practical purposes, non-functional.
I have observed this dynamic before. When I built the causal graph mapping the Terra collapse in 2022, I watched a system survive only as long as its participants maintained belief โ and belief eroded in direct proportion to visible decay of fundamentals. The BIP-110 fork chain does not need an active attack to die. It needs only to be ignored. The protocols of the economy will handle the rest.
Historical Precedent: The BCH Contrast
The 2017 Bitcoin Cash fork is the default reference point, and the comparison deserves scrutiny. BCH split from BTC over a substantive issue: block size limits. The BCH faction argued for eight-megabyte blocks and backed the argument with a meaningful share of hashrate, prominent ecosystem players, and a genuine user base. Critically, they implemented replay protection before the split.
BIP-110's would-be fork has none of those characteristics. The issue โ OP_RETURN data limits โ is a policy dispute over the permitted volume of metadata on the blockchain. It does not change Bitcoin's economic model. It does not alter the security budget or the issuance schedule. It does not address scalability, privacy, or programmability. It is a governance argument about a byte limit, executed through software versioning.
The BCH fork, despite its substantive backing and replay protection, remains a cautionary tale even for its proponents. BCH's market capitalization is a fraction of its 2017 peak. BSV, which split from BCH over further block size increases, is nearly forgotten. The pattern is unambiguous: even forks with genuine support and real technical differences struggle to establish durable value. A fork with 2.6 percent hashrate, zero replay protection, and a dispute over metadata payloads has no credible path to viability.
What the BCH precedent does validate is the main chain's resilience. Bitcoin absorbed the BCH split, and within weeks the price was setting new highs on the way to the late-2017 bull run. The attention effect โ media coverage, exchange listings, public debate โ brought new buyers into the ecosystem. The fork narrative, contrary to fear-driven expectations, became a marketing event for Bitcoin itself.
The critical difference: 2017 had at least two viable sides. The BIP-110 "side" is a rounding error. The economic incentive for miners to maintain a chain producing blocks every six hours for a token trading at a discount from BTC is nonexistent. Mining is a business. Businesses do not subsidize ideological statements.
Token Economics of a Zombie Chain
Consider the fork chain's token model, because it fails every viability test.
Supply: The fork chain inherits the entire BTC supply โ every historical coin, every future issuance. But this is not an asset with independent value. It is a claim on a chain with no demand side.
Value: The cost of producing a fork chain block is identical to the cost of producing a main chain block โ the same ASICs, the same electricity, the same overhead โ for a fraction of the expected revenue. With block times stretched to six hours, transaction fees on the fork chain approach zero. The block subsidy becomes the only revenue, denominated in a token that has no pricing benchmark and no liquidity.
Incentives: The core question โ why would a miner maintain a chain with 2.6 percent hashrate โ has three possible answers. Ideological conviction: the belief that Bitcoin should not carry arbitrary data. Strategic positioning: accruing goodwill with a potential future faction. Short-term speculation: mining the fork chain early in hopes that exchange listings or media attention create temporary demand. All three answers converge on the same outcome: the absence of sustainable revenue.
Liquidity: Even if a secondary exchange lists the fork token, the order book will be thin, the spreads will be punishing, and the settlement risk โ replay attacks on any asset that touches the chain โ will discourage institutional participation. The fork coin is not an asset. It is a liability wearing an asset's clothing.
The real economic exposure is behavioral. Users who attempt to capture a "free" fork token are spending real BTC risk to acquire a token with no fundamental demand. The expected value calculation is catastrophic: a high probability of losing principal against a low probability of acquiring tokens that trade at a fraction of a cent. Auditing the dream to find the debt โ the free money narrative conceals the actual balance sheet of risk.
Governance Autopsy: The Process Failure
The 2.6 percent miner signaling figure deserves a governance decryption. Bitcoin's BIP9 mechanism historically required 95 percent miner signaling for a soft fork to activate within a defined window. The rarely discussed but politically significant economic majority threshold sits around 55 percent. BIP-110's support clears neither bar by an order of magnitude.
The implementation detail matters. The BIP-110 software appears to skip conventional activation gates entirely. Nodes adopting the rule simply begin rejecting non-compliant blocks at the predetermined height. This is not a negotiated soft fork. It is a unilateral declaration โ and Bitcoin's consensus layer, as always, renders the final judgment.
Bitcoin's governance has survived worse disputes. The blocksize war of 2015-2017 produced factions, personal attacks, and existential predictions. SegWit2x collapsed under its own weight. The pattern that emerges is adversarial collaboration: intense conflict, messy compromise, and a main chain that absorbs shocks and continues producing blocks every ten minutes.
The governance lesson is that process legitimacy matters more than proposal merit. A proposal that commands 2.6 percent support has not earned legitimacy through the BIP process. It has attempted to bypass the process through mechanical enforcement. The code remembers what the market forgets โ the code will remember this attempt as a failed governance coup, not a successful reform.
There is a deeper structural point. Bitcoin's governance does not require unanimity. It requires enough alignment that the cost of dissent exceeds the benefit. At 2.6 percent, the cost of dissent is nearly zero, which is precisely why the dissent is happening. The BIP-110 faction has calculated that the expected value of a fork โ even a doomed one โ exceeds the expected value of continued negotiation. That calculation is almost certainly wrong, but it explains the behavior.
The Contrarian Reading: The Real Attack Surface
Here is the counter-intuitive conclusion that the warnings obscure. The headline framing โ "Real BTC Theft" โ is technically accurate and statistically misleading. The attack surface is not the fork. The attack surface is the user's own curiosity and greed.
Reconsider the numbers. If the fork materializes at all โ and 2.6 percent support makes that genuinely uncertain โ the only users at risk are those who move assets during the fork window, attempt to sell fork coins, or import wallets into fork-aware clients. Every one of those actions is voluntary. The attacker does not break into wallets. The attacker does not exploit a zero-day vulnerability. The attacker exploits the gap between user expectation and protocol reality.
The victims, if any, will be drawn disproportionately from the least sophisticated participants. Sophisticated traders will route around the fork with the precision of a treasury operation. Institutional holders will receive custodial warnings days in advance. The "free money" seekers โ the users who see an unearned token in their wallet and immediately attempt to sell it โ will be the ones signing transactions into the replay trap.
This is not a systemic event. It is not a market structure failure. It is a behavioral tax on impulsivity during a period of manufactured uncertainty. The narrative amplification becomes part of the attack surface. The headlines are the weapon. The fork itself is nearly irrelevant.
But the contrarian analysis cuts in both directions. The blanket "do nothing" advice, while correct for self-custody users, does not apply uniformly. Patterns emerge where amateurs see chaos: users holding BTC on mainstream exchanges are insulated by the exchange's compliance infrastructure, which will almost certainly implement replay protection or suspend withdrawals. Users with self-custody wallets face the full risk surface. The risk distribution is deeply uneven, and blanket warnings obscure this structural detail.
There is a second contrarian point. The media ecosystem has a perverse incentive to amplify this story. "Bitcoin fork" generates clicks. "Miner split" generates panic. "Real BTC theft" generates outrage. The truth โ a 2.6 percent faction attempting a doomed policy fork โ generates neither. The information asymmetry between what the headlines imply and what the data shows is the actual story worth covering.
Takeaway: The Signals That Matter
The blocks around height 961,632 will arrive regardless of commentary. The signals to watch are concrete: miner pool statements from Foundry USA, AntPool, and F2Pool; exchange announcements regarding BTC withdrawal status; and the actual block production rate on any minority chain, if one emerges.
My professional read, based on a decade of on-chain forensics and specifically my work mapping institutional flows during the post-ETF era: this fork does not sustain itself. 2.6 percent signaling support does not keep a chain alive. The economic mathematics are unforgiving, and the governance history is unambiguous.
But the window between now and the decisive height is precisely when users make the mistakes that warnings describe. The safest position is also the simplest: hold. Do not claim fork coins. Do not sell fork coins. Do not move BTC to "secure" it from a threat that does not exist until the moment you act. The unspent UTXO is the only cryptographically safe UTXO.
From certification to conviction: mapping the flow of this entire event leads to one destination โ patient inaction. That is the entire trade. The ledger will settle the rest, as it always does.