Maybe there should be official guidelines for stats communication that require a certain level of legalise-like specify. No more "there is no evidence for.." of the sci world.