Calibration Is Not Classification Accuracy
A model can be accurate and badly calibrated, or well calibrated and useless. What each property means, how to measure both, and why per-class thresholds are a decision rather than a result.
Learn / level
Foundations
2 published pieces at this level.
Equal Counts Do Not Prove Two Stores Agree
Why a matching row count is the weakest possible integrity check, and what a ladder of comparisons on identity, content and provenance actually rules out.