>>12178417I'd replace normal distribution with central limit theorem here; the normal distribution is important because it pops up so much, but that's really only because of the central limit theorem, and by itself is just one of the many other probability distributions; basically all statistical tests with p values (that aren't rank-tests/non-parametric) rely on the central limit theorem, aka all of science more or less relies on it; and the most commonly used test isn't even normal distribution, its the t-distribution (although yes, its the normal distribution with an extra parameter)