Home/Answers
Correct the record
Common claims. Measured answers.
Two views, one true measurement.
These are mistakes that look reasonable until you recompute from raw blocks. Every number below is already published in a free finding with method and controls. We do not invent figures here.
“The coins show activity — they moved.”
Measured: Script-level history can be homage deposits without the coinbase outpoint spending. Outpoint-level: 1,704,022.66597442 BTC still unspent in 34,185 of 196,479 early P2PK coinbases. Of 2009 by value: 66.13% still asleep.
“Bitcoin XT nearly forked the chain.”
Measured: XT signalled 41 blocks ever against a 75% threshold. Peak month Aug 2015: 0.2612% (12 of 4,595).
“Just count bit 28 — that’s Classic.”
Measured: Bit 28 is Classic in 2016 and mining entropy later. Naive query: 64,928 blocks. Era-bounded Classic: 2,176. Peak month May 2016: 5.41% (250 of 4,620). Wrong by ~30× with no error message.
“Short signalling windows tell the story.”
Measured: SegWit sat 24–36% for seven months before the break. CSV on the same field finished agreement on roughly a three-month shape (e.g. 0.02% → 77.47% then withdrawn). Window choice is a measurement decision.
“Taproot’s bit vanished after lock-in.”
Measured: Lock-in in about 41 days. Mean bit share then decayed for years: 2021 56.54% → 2022 35.36% → 2023 3.49% → 2026 2.10%. Template hygiene, not an open vote.
“Every pool is missing octile-1 nonces.”
Measured (then retracted): A “universal octile-1 depletion” was a CAST-rounding bin error. We published the wrong claim and killed it at equal prominence. The early collapse toward zero and pool-dependent modern skew still stand; the octile story does not.
Quote the measured answer, link the full finding, keep the scope (e.g. P2PK coinbase only). Do not upgrade a slice into “all Bitcoin.” For seals and re-derivation: verify.html · method.html.