chat model, one letter · League 3

Claude Opus 5.5low

FinishedPublished

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

How we called it

Model ID
claude-opus-5-5
Provider or endpoint
Claude subscription
Date run
2026-10-20 to 2026-10-24
Price
$4.0 / $20.0 per million tokens in / out; API-equivalent (subscription); price date not recorded
Model type
chat model, one letter
Weights
closed
Runs on
subscription
Reasoning setting
effort low, the lowest Opus 5.5 accepts (adaptive thinking only)
Confidence source
none
Rules text
long
Option groups
up to 26 options per request (letters A-Z); more legal plays are split into equal groups, a winner per group, then among the winners; in this data 178 decisions were split into groups of 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26
Sampling settings
not sent (API default)
Notes
Missing
Earlier name
Claude Opus 5.5 (off) (renamed on 28 Oct 2026 to the reasoning effort its requests sent)

Missing B7 fields: price.date

Results in League 3

Error rate (PR)
15.413.1 to 18.0
Decisions
1,411 (1,411 scored, 0 invalid)
Engine's best play
61.1%
Blunder rate
13.7%
P95 equity loss
0.158
Mean loss, worst 5%
0.279
Median / P90 s per decision
3.92 / 10.6
$ per 1,000 decisions
29.062API-equivalent (subscription)
Risk judgement, margin 0.001
66.9 (95% interval 64.7 to 69.2); risks found 40.1% (187 / 466), right safety 93.8% (541 / 577)
Risk judgement, margin 0.02
67.1 (95% interval 64.5 to 69.5); risks found 41.9% (168 / 401), right safety 92.4% (593 / 642)
No trade-off
368 (left out of every risk metric)

JSON

Risk call card

66.9risk judgement, margin 0.001

right risks187
wrong risks19
needless risks36
missed risks260
right safety541

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
1,411
Eligible (full probability distribution)
0
Left out, by reason
one letter only, or a program 1,411
Close choices
N/A no full probability distribution

By game phase

PhaseError rate (PR)Decisions
OpeningToo few decisions199
Contact20.6856
Holding gameToo few decisions73
RaceToo few decisions161
Bear-offToo few decisions122