I wish! The words empirical distribution are music to my ears. No, usually the rule people use is this: "If A1 and A2 show a statistically significant differen

I wish! The words empirical distribution are music to my ears. No, usually the rule people use is this: "If A1 and A2 show a statistically significant difference, then do not reject the null hypothesis regardless of A1/B or A2/B."