@andybaxter@ArthurCDent@mark_ledwich You run a test with identical treatments for both groups. You get a p value of 0.04. there isn't actually a 96% chance that the null hypothesis is false. There's just a 4% chance you'd get the result you got, or one more extreme.