And a honest n=1 is better than a double blind n=1000 but outcome-interested study. It’s so easy to make data tell the story you want.