i bet i review more experiments in month than most academics do in in a year
-
-
hmm idk we do quite a bit of this on my team only bring a change to experiment review if theres a statsig positive change, otherwise we kill it in the dead of night
-
but ranking probably works somewhat different than product growth
End of conversation
New conversation -
-
-
isn't bimodality expected here? there's selection, |z-values | < 1.96 would usually not be considered publishable
-
exactly
- Show replies
New conversation -
-
-
Same with trading. If I ever tweak an algo to look good, but in a way that's obviously fake, i'll just end up losing money :) It's why Taleb is based.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
highly memeable graph shape
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Or as my wife puts it: We need to be right and they need to publish.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
That is an alarming imagine right there.
- End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
