assayed

Wilson score confidence intervals for measured success rates. No dependencies.

When you measure how often something succeeds (requests that complete, payments that settle, tests that pass), the raw rate hides how much you actually know. The Wilson score interval gives an honest range that stays inside 0 to 1 and behaves well for small samples and for rates near 0 or 1.

Install

pip install assayed

Use

from assayed import wilson_interval

low, high = wilson_interval(481, 500)          # 95 percent by default
low, high = wilson_interval(8, 10, confidence=0.99)

wilson_interval(successes, trials, confidence=0.95) returns (low, high). It raises ValueError if trials is below 1, if successes is outside 0 to trials, or if confidence is not strictly between 0 and 1.

A TypeScript and JavaScript version with the same function is published on npm as assayed.

Reference

E. B. Wilson (1927), "Probable inference, the law of succession, and statistical inference", Journal of the American Statistical Association 22(158), 209 to 212.

Licence

MIT. Copyright (c) 2026 Everyday AI Ltd.