docs/statistics.md explains the Wilson interval but doesn't walk a concrete number, which is the part a
non-statistician needs.
Add a worked example: take a pass rate like 46/50, show the Wilson interval bounds, and explain in plain
language what the interval means for reading a HOLDS/REGRESSED result. Reference the implementation in
tjbench/stats.py (wilson_interval).
Acceptance criteria: one worked example in docs/statistics.md a reader can follow without a stats
background.
docs/statistics.mdexplains the Wilson interval but doesn't walk a concrete number, which is the part anon-statistician needs.
Add a worked example: take a pass rate like 46/50, show the Wilson interval bounds, and explain in plain
language what the interval means for reading a HOLDS/REGRESSED result. Reference the implementation in
tjbench/stats.py(wilson_interval).Acceptance criteria: one worked example in
docs/statistics.mda reader can follow without a statsbackground.