Run T-0239 rejected

2026-11-07 · Evidence: SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds or the game limit fixed in advance; then the test stops.

Hypothesis

Separating the evaluation correction move entry based on whether the previous move was a capture or promotion, rather than a quiet move or null move, is expected to increase playing strength. In search and quiescence search, moves following a capture or promotion use one dedicated table entry, while those following a quiet or null move retain the previous entry, with learning updating the corresponding cell under the unchanged update condition.

Wording edited for publication from the German entry; the register carries the author’s own text.

Stages

Point estimate with the reported 95-percent half-width, per stage. The dashed line is zero. Estimates from a stopped sequential test are biased by the stopping rule; the interval is nominal.

Field	short stage
Verdict	rejected
Elo Strength difference to the opponent, estimated from the games, with a 95-percent half-width where the oracle reported one. Never an absolute rating.	+0.41 ± 1.63
nElo Normalised Elo: relates the mean score excess of a colour-swapped pair to the spread of the pair results.	+0.84 ± 3.29
Games	42,740
Wins / draws / losses	9,795 / 23,201 / 9,744
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.	`59 / 5004 / 11188 / 5065 / 54`
LLR Log-likelihood ratio, the running evidence of an SPRT. It starts at 0; a stage passes at the upper bound and fails at the lower one.	−2.95
Bounds	−2.94 … +2.94
SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds or the game limit fixed in advance; then the test stops.	[0.0, 5.0]
Error rates α / β	0.05 / 0.05
Model	normalized
time control Base time plus increment per move in seconds, e.g. 8+0.08: eight seconds per game plus 0.08 seconds per move.	8+0.08
Book	A
Opponent	vs predecessor
Duration	5 h 20 min

Provenance

Date: 2026-11-07
Started: 2026-11-07 11:58:39 UTC
Duration: 5 h 20 min
Author: Agent D
Epoch: 14
Gatekeeper: passed
Bench nodes: 603