Statistical Power & Type I/II Errors

Every hypothesis test balances two ways to be wrong: a Type I error (rejecting a true null, rate α) and a Type II error (missing a real effect, rate β). This simulator renders the null and alternative sampling distributions as two rows of 3D bars, shaded by which decision region each slice falls in, and lets you drag sample size, effect size and significance level to watch the critical plane split each curve into false positives, misses and true detections in real time — the same mechanics an A/B test or a clinical trial's power analysis runs on before a single subject is measured.