Skip to content

metrics: refresh the big-case board (270 case(s), 780 scored stakes read(s)) - #2283

Merged
fedcourtsai-dev[bot] merged 1 commit into
mainfrom
metrics/big-cases
Oct 7, 2026
Merged

fedcourtsai-dev[bot] merged 1 commit into
mainfrom
metrics/big-cases

Conversation

@fedcourtsai-dev

Copy link
Copy Markdown
Contributor

Deterministic re-roll of metrics/big-cases.{json,md} from the committed ledger by fedcourts big-cases: 270 case(s) ranked over 780 scored stakes read(s) of 808; the rest carry no score and are excluded from every mean rather than imputed.

Not a performance surface. A stakes read is neither scored nor ranked: it resolves against nothing, so no figure in this diff is a forecast, an accuracy or a skill number, and the mean says nothing about how likely a case is to be granted. The board reads the ledger directly and applies neither the forward-claim nor the leakage exclusion — a wider population than the scored boards. The artifact's own provenance block and metrics/README.md carry the reading rules.

The board is a byte-stable pure function of data/, so this PR exists only because the ledger moved; a day that changed nothing opens nothing. Auto-merge is armed on the same required checks a reviewed PR faces.

Run: https://github.com/ModelMirrorAI/fedcourtsai/actions/runs/37573245645

@fedcourtsai-dev
fedcourtsai-dev Bot enabled auto-merge (squash) October 7, 2026 04:49
@fedcourtsai-dev
fedcourtsai-dev Bot merged commit bb249ab into main Oct 7, 2026
7 checks passed
@fedcourtsai-dev
fedcourtsai-dev Bot deleted the metrics/big-cases branch October 7, 2026 04:55
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants