Setup · League 3

GPT-6.1 Sollow

FinishedPublished

Results f24d1f41, updated 16 November 2026, 21:24 UTC; scored by GNU Backgammon 1.08, 2-ply.

How we called it

Model ID
Missing
Provider or endpoint
Missing
Date run
2026-10-19 to 2026-10-24
Price
Missing
Model type
Missing
Weights
Missing
Runs on
Missing
Reasoning setting
Missing
Confidence source
Missing
Rules text
Missing
Option groups
not recorded; in this data 212 decisions were split into groups of 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26
Sampling settings
Missing
Notes
Missing

Missing B7 fields: every B7 field

Results in League 3

Error rate (PR)
12.110.5 to 13.7
Decisions
1,344 (1,344 scored, 0 invalid)
Engine's best play
65.5%
Blunder rate
9.4%
P95 equity loss
0.137
Mean loss, worst 5%
0.254
Median / P90 s per decision
4.33 / 14.3
$ per 1,000 decisions
10.228API-equivalent (subscription)
Risk judgement, margin 0.001
72.7 (95% interval 69.9 to 75.5); risks found 51.7% (198 / 383), right safety 93.8% (602 / 642)
Risk judgement, margin 0.02
74.2 (95% interval 70.9 to 77.4); risks found 56.2% (182 / 324), right safety 92.2% (646 / 701)
No trade-off
319 (left out of every risk metric)

JSON

Risk call card

72.7risk judgement, margin 0.001

right risks198
wrong risks16
needless risks40
missed risks169
right safety602

Hit risk only: being hit. Gammon risk and priming are not in it.

Close choices

A decision is a close choice when the second most likely play has at least half the probability of the first (internally: split).

Probabilities over plays, not sampled reasoning. Method

Decisions
1,344
Eligible (full probability distribution)
0
Left out, by reason
one letter only, or a program 1,344
Close choices
N/A no full probability distribution

By game phase

PhaseError rate (PR)Decisions
OpeningToo few decisions196
Contact16.2826
Holding gameToo few decisions71
RaceToo few decisions128
Bear-offToo few decisions123