Bitcoin wallets revisit a noisy RBF signal

Bitcoin Optech’s June 19 newsletter highlights a Bitcoin-Dev discussion about a small but revealing wallet behavior: whether wallets should stop explicitly signaling opt-in replace-by-fee, or RBF, in the transactions they create. RBF lets an unconfirmed Bitcoin transaction be replaced by another transaction paying higher fees. For years, wallets signaled that behavior through a particular use of the nSequence field. According to the discussion summarized by Optech, that signal no longer has the same policy meaning now that full replacement is the default behavior in Bitcoin Core.

The technical point is narrow, but it matters. Under BIP125, a transaction signals opt-in RBF when at least one of its inputs sets nSequence below MAX-1. Bitcoin Core nodes using default policy can now replace transactions regardless of those values. In other words, the old signal no longer decides whether replacement is possible in the way it once did. What it can still do is reveal something about the wallet that produced the transaction. In Bitcoin, privacy often depends on reducing exactly these small, repeated formatting differences.

The initial proposal from rkrux was to make the Bitcoin Core wallet stop signaling by default. A related pull request was opened on May 28 and closed on June 16 after feedback. The discussion quickly moved beyond a simple choice between signaling and not signaling. Murch and Electrum contributor SomberNight pointed out that MAX-2 is already the dominant value in observed transactions, based on the data cited by Optech. If Bitcoin Core moved alone to a non-signaling MAX-1 value, its transactions could stand out more, not less.

That makes the issue a coordination problem for wallet developers. Once an old network-policy convention stops doing useful policy work, it can become a fingerprint. The better outcome may be convergence on the most common behavior, rather than a technically tidy change made by one wallet in isolation. The user-facing result, if wallets coordinate, would be almost invisible: transactions would keep working, fee replacement would remain available under current relay policy, and wallets would leak fewer unnecessary clues about their origin.

The broader lesson is that Bitcoin privacy is not only about major cryptographic upgrades. It also depends on routine implementation choices: sequence values, default flags, transaction structure, and whether independent wallets behave alike. This discussion is not a consensus change, and it does not launch a new feature. It is a maintenance conversation about making Bitcoin transactions less distinguishable at the edges, which is often where real wallet privacy is won or lost.