In a non-p-hacked study with an adequate sample size, you should only expect around a quarter of p-values to be .01 < p < .05.

Hi, is this statement about "a quarter" an empirical estimate derived from data?