registrycompare

Role then format against Signature echo

Both scored on the same cases against claude-fable-5, same judge. The intervals tell you whether the gap is real or noise.

suite reasoning-chainmodel claude-fable-5n=168 public cases each

role-then-format

v2 by @ilya
91.7
86.5 to 95.0 · holdout 90.3

signature-echo

v2 by @dara
92.9
87.7 to 96.0 · holdout 92.4
Intervals overlap. The 1.2 point gap is not significant at this sample size.
Metricrole-then-formatsignature-echoBetter
Public score91.792.9right
Holdout score90.392.4right
Interval width8.58.3right
Cost per run$1.850$1.850same
Median latency2.4s2.8sleft
Sample size168154left

Reading this verdict

Overlap
Two entries are only meaningfully apart when their confidence intervals do not overlap. A higher number with an overlapping interval is a coin flip, not a win.
Holdout
The public score can be gamed by tuning to visible cases. A gap between public and holdout is the tell.
Reproduce
evalness compare \ role-then-format signature-echo \ --model claude-fable-5