split icon indicating copy to clipboard operation
split copied to clipboard
verified publisher icon splitrb

Winning alternative have more probability of losing (?)

Open caepenna opened this issue 7 years ago • 0 comments

One of my experiments have one alternative winning by 7% but it has lower probability of winning (41%). Does it make any sense? I can't figure out the math behind it.

There is still no sufficient confidence, and I think it will normalize with more participants, but I was intrigued by this and wanted to understand it better. Does anyone have any idea why this is going on?

screen shot 2018-08-08 at 17 36 23

caepenna avatar Aug 08 '18 20:08 caepenna