Matched route screens do not establish adapter causality

A time-matched screen showing a lower automated failure-candidate rate for one model route establishes only a difference between the observed route cohorts. It does not establish that an adapter caused the difference or that detector candidates equal confirmed failures. To attribute improvement to adapters, replay identical causal checkpoints through the base model and each adapter under matched conditions. Separately adjudicate detector candidates against the original audio or context before estimating failure prevalence.