statistically abnormal player lucifer

Other Reports Discussion forum.
User avatar
adminless
Site Admin
Site Admin
Posts: 5939
Joined: Thu Nov 03, 2016 19:05
in-game nick: not available
Location: Spain

statistically abnormal player lucifer

Post by adminless »

hey hello guys

hehe long time without activity at this sub-forum, well that's good I guess. so this time reporting that now that accuracy game to game standard deviation has been added to the stats site this player, Lucifer, coming from a average Russian connection has returned data statistically abnormal. at the moment he just has a 15.7% accuracy on average (which is nothing special) but he does it so with a ±13.0 standard deviation meaning that he basically goes from close to 0% games to games where he doubles his average accuracy into values close to the top players. such a standard deviation, as can be seen from the accuracy rank, is extraordinary uncommon (usual accuracy standard deviation is around 6/7 and typically between 9 and 3) and therefore is something that must be checked as soon as possible as although this does not imply that he is in fact cheating under any circumstance it definitively evidences that there's something weird about this guy that needs to be checked.

it could just be that he goes from using only rockets to using only rails from game to game or that his connection breaks badly from day to day or simply the difficulty variance of the different rooms he plays from day to day, whatever, all that can be said so far is only that there's solid data (gathered over more than 130 games and 10 hours of playing time) that evidence a abnormality that must be checked on this guy.
User avatar
adminless
Site Admin
Site Admin
Posts: 5939
Joined: Thu Nov 03, 2016 19:05
in-game nick: not available
Location: Spain

Re: statistically abnormal player lucifer

Post by adminless »

ok guys never mind, found the issue now while digging it more thoroughly, turns out the source of this variance is the network that this guy is coming from, looks like there has been more than a couple of different guys playing as Lucifer (with first cap) and even from the very same location which explain the reason of the abnormal variance value (as it couldn't be other way, it was measuring different players). I had unceremoniously missed to check that on the initial investigation so issue cleared now.
User avatar
adminless
Site Admin
Site Admin
Posts: 5939
Joined: Thu Nov 03, 2016 19:05
in-game nick: not available
Location: Spain

Re: statistically abnormal player lucifer

Post by adminless »

btw as this was spotted from now onward the "lucifer" nick (and their squats) will be disallowed until someone register it, sorry about that guys but this is something that shouldn't be happening. unfortunately this is what typically happens when there's various players at the same home (family), people share their cfg/game files with their friends or people play from public places (ex. a cyber coffee shop) then to prevent this mix up of games the "lucifer" nick has just been disallowed from now onward. if you were its legit owner you can just request it to get it registered by answering this thread.