How it works: We simulate running multiple independent statistical tests where the null hypothesis is true every time — there is no real effect.
Each test still has a α = 5% chance of a false positive. With n tests, the probability of at least one false positive is:
P = 1 − (1 − α)n