rung rung-seed   knob superseded_weight   probes 26 (13 current / 13 history)
baseline at the shipped default 1.0: 21 / 26 right

   value    right  fixed  broken      cur     hist   verdict
     1.0   21 / 26      0       0   12 / 13    9 / 13   baseline
     0.9   21 / 26      1       1   13 / 13    8 / 13   BREAKS 1: p07
    0.75   19 / 26      1       3   13 / 13    6 / 13   BREAKS 3: p03 p07 p08
     0.5   18 / 26      1       4   13 / 13    5 / 13   BREAKS 4: p03 p04 p07 p08
    0.25   18 / 26      1       4   13 / 13    5 / 13   BREAKS 4: p03 p04 p07 p08
     0.1   18 / 26      1       4   13 / 13    5 / 13   BREAKS 4: p03 p04 p07 p08
     0.0   18 / 26      1       4   13 / 13    5 / 13   BREAKS 4: p03 p04 p07 p08

HEADROOM, observed, from the rows (ADR-RS decision 22b):
  improvement — probes not right at the default: 5 / 26
  regression  — probes right at the default:     21 / 26

🔴 NO single global value clears the 0-broken bar. Every value that fixes a probe breaks another, which is the pre-registered answer NO — and it is a result, not a failed run.

per-probe rows (182) -> work/regression/2026-09-12-priors-and-tables/evidence/sweep-rung-seed-superseded_weight.jsonl
