Regime: choppy, unchanged
On the last closed daily bar (3 October) Bitcoin closed about 2.77% above its 20-day average, and only +0.38% over seven days. That is inside the band the regime grader treats as trendless, so the standing choppy read (issued 2 October, running to 9 October) is unchanged and nothing new was appended. An older rotation read is also still open until 6 October.
New calls
The structural scorer also produced a fresh NEAR-versus-Bitcoin relative read, but no call was issued. The structural lane is at its cap of six standing slots, and it is noted here only as a coverage gap.
- Solana (SOL), 7 days. Leans to reach +4.05% from issuance at some point within the week. The engine puts this at roughly 67%, against a touch base rate of about 59% for a move that size. Our weekly coverage read and price momentum both lean the same way, with momentum the stronger of the two.
- Ethereum (ETH), 7 days. Leans to reach +3.02% within the week, at roughly 60% against a base rate of about 56%. The gap is thin. Momentum carries most of it, our weekly coverage read leans only slightly up, and the most recent one-day coverage window tagged Ethereum bearish. Treat this one as a marginal read.
None of these numbers is calibrated yet. A touch call's probability sits near 60% largely because the touch base rate itself is near 60%. That is a matter of what is being measured, not evidence of skill. The complement of a touch call is "does not reach the level", not "moves the other way".
Receipts
Two 7-day calls matured on 4 October 02:25 UTC and were resolved at 03:49 UTC. Both were misses:
| Call | Issued | Resolved | Outcome |
|---|---|---|---|
| Uniswap (UNI) leans higher, 7 days | 27 Sep 02:25 UTC | 4 Oct 03:49 UTC | Miss (-7.65%) |
| Solana (SOL) leans higher, 7 days | 27 Sep 02:25 UTC | 4 Oct 03:49 UTC | Miss (-1.26%) |
Both were graded on direction alone, the rule in force when they were issued. Both moved the wrong way, so no grading rule would have scored them as hits.
Two 1-day calls (Bitcoin, XRP) reached expiry at 03:48 UTC but had not been resolved when this was written. They are reported next run.
Accuracy
Under the current grading rule (touch of the stated level), the public claim pool is still below its floor: 0 of the 25 resolved 7-day calls needed, and 2 of the 25 needed on the 1-day horizon. So there is no Brier Skill Score to report yet, and we do not quote one from a sample of two. Under earlier rules the record was negative: our probabilities scored worse than simply quoting the base rate. Those rules are not comparable to the current one, and we have not yet shown that our probabilities beat the base rate.
