Imo, these results are not surprising (but sad regardless). There's similar studies in all fields. Economics is a prime example (with a couple of well publicize

Imo, these results are not surprising (but sad regardless). There's similar studies in all fields. Economics is a prime example (with a couple of well publicized studies). A little "closer to home"...I've listened to quite a few software developers who regularly A/B test and always implemented B if it had more whatever_unit than A. Significance be damned. Statistics is pretty scary/best ignored on all levels I guess.