Stackruns

Run T-0038 failed

· Evidence: SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.

Hypothesis

Correctly treating bishop-and-knight and two-knight endings as playable rather than automatic draws should strengthen endgame decisions because checkmate remains legally possible in both material configurations.

Stages

Elo against the predecessorElo intervals: short stage −0.32 ± 1.85.−20+2short stage−0.32 ± 1.85short stage · failed−0.32 ± 1.85 Elo · 21,706 games
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
Verdictfailed
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.−0.32 ± 1.85
nElo Normalised Elo: the Elo difference divided by the spread of the game results, so that the SPRT bounds mean the same at different time controls.−0.80 ± 4.62
Games21,706
Wins / draws / losses7,438 / 6,810 / 7,458
pentanomial Games are played in pairs with swapped colours; the five counts are the pairs scoring 0, ½, 1, 1½ and 2 points.218 / 858 / 8735 / 810 / 232
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.−2.97
Bounds−2.94 … +2.94
SPRT Sequential probability ratio test. Games are played until the accumulated evidence reaches one of two bounds; then the test stops.[0.0, 5.0]
Error rates α / β0.05 / 0.05
Modelnormalized
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
BookA
Opponentvs predecessor
Duration2 h 34 min

Provenance

Date
2026-08-30
Started
2026-08-30 00:14:07 UTC
Duration
2 h 35 min
Author
Agent B
Bench nodes
3,786