Quote (Snyft2 @ 11 Aug 2018 20:22)
thats the beauty of statistics, we dont need to know :blush: ! the point is this: if some thing or change applies to a sample that is large enough, it will have a large enough chance to apply to the entire population from which that sample was taken :blush: ! and in general statistics you can assume that even a sample that is 30 or higher is valid, while steam has a sample of more than 1000 times that :wacko: !
so this is how you do it: if you take steam players as a sample of the poe players population, and harbinger had 25k more players than bestiary in that sample, even if you assume steam players are only 1/10 of the poe population, the chance that it applies to the entire population is so huge that it exceeds the level of 5-6 sigma, which i think was the level where something is considered as scientific truth, i mean if you make a hypothesis which is worthy enough and you prove it with 5 sigma, you get a nobel prize :wacko: basically you can say with 99.999...% of accuracy that poe had less players in abyss than in garbinger. now that is probably due to the 10 act expansion that came out back then, but the reason isnt needed at start, just the numbers!
or it just means that between harbinger and bestiary more people decided to use the stand alone client...