Accuracy record

Every estimate is retained with its inputs. Where a realized fill of comparable size is subsequently observable on the same venue, the estimate is scored against it, and the distribution of those errors is published — not a single vanity accuracy percentage, but the error distribution by size bucket, by venue and by market condition.

The design constraint#

Accuracy scoring must be automatic and adversarially reproducible from published data. If Crifine is the only party able to compute its own accuracy, the score is worthless. Every input to the score is therefore published on the evidence pages the score is derived from.

Status#

The accuracy record is roadmap. It requires a longer series than currently exists and a matched set of observable fills. It ships before any mechanism that would collateralize an estimate — you cannot bond a claim whose error distribution nobody has measured yet. The public statement of this commitment, including why the $CRIF accuracy bond waits on it, is at the accuracy record.