You need to be logged in to your Sky Poker account above to post discussions and comments.
You might need to refresh your page afterwards.
what does sky poker do to stop bots?
Doubleme
Member Posts: 1,484
in Poker Chat
so I have seen a few players online who I am suspicious that these are bots. I wont mention any specific player names because that is unfair to mention names here in the forums on such suspicions.
Generally what makes me suspicious of a bot is when I have noticed them consistently not just once or twice or a few times but consistently do plays that just look absurd and terrible in such a way that no human player would do yet have win rates that are statistically significant.
I am assuming of course that the plays they make which look terrible are actually not. for reasons that I cannot grasp, I am talking about plays such as calling off half their stack when 25 blinds deep heads up with j7 off suit.
I am doing a statistics degree I know how to calculate statistical significance if a player is regularly doing weird plays that make no sense to me not just occasionally but regularly and achieving win rates I cannot with statistical significance then I have to logically conclude they are playing on a higher level then me. I do not claim to be the best player out their I just take what I can.
So I don't mean to be arrogant and just be like well if they think on a higher level then me then has to be a bot because that is ridiculous I just get suspicious when it looks totally absurd. yet consistent winner.
I just wanted to know that sky takes proper steps to detect and eliminate bots.
Generally what makes me suspicious of a bot is when I have noticed them consistently not just once or twice or a few times but consistently do plays that just look absurd and terrible in such a way that no human player would do yet have win rates that are statistically significant.
I am assuming of course that the plays they make which look terrible are actually not. for reasons that I cannot grasp, I am talking about plays such as calling off half their stack when 25 blinds deep heads up with j7 off suit.
I am doing a statistics degree I know how to calculate statistical significance if a player is regularly doing weird plays that make no sense to me not just occasionally but regularly and achieving win rates I cannot with statistical significance then I have to logically conclude they are playing on a higher level then me. I do not claim to be the best player out their I just take what I can.
So I don't mean to be arrogant and just be like well if they think on a higher level then me then has to be a bot because that is ridiculous I just get suspicious when it looks totally absurd. yet consistent winner.
I just wanted to know that sky takes proper steps to detect and eliminate bots.
Comments
then if the z value is greater than 3 I judge that to be statistically significant.
typically the convention in statistics is to take 95% confidence interval which for a one sided test would be 1.645 z value
or for a two sided test 1.96 z value.
or we could take a z value that would give us 99% confidence which would be a z value of 2.3
but I just go with a z value of 3 which would be 99.865% if we want to go 99.99% we would go 3.71 z value.
It is possible for a player to be making strategical errors and/or be prone to tilt or punt and still achieve "statistically significant win rates". The plays will be terrible, they won't actually be good for reasons beyond your comprehension (as you suggest). You can be 'bad' at poker and still win long term as long as you are not quite as bad as the majority of the players you are playing against regularly. It's all relative.
I said I was concerned about the potential of bots and mentioned I had looked at the statistics myself i did not say I have proof of bots if I did I would not post here it would contact sky and discuss that with them.
I get one person questioning whether I even know A level statisics when I am doing a degree in the topic then when I show I do I get told its utterly irrelevant why do people want to derail me?
All I am asking is wether sky poker have proper protcols in place to catch bots, because quite frankly I do not want to get destroyed by them at a later point.
I have been a consistent profitable player on this site for years, nothing to boast about at all, but im not just some idiot fish who got outplayed and then cried must be bots and oo look they won over a sample of 27 games must be significant.
I am aware you get dullards popping up claiming to have a PHD in Mathematics and can proof the site is rigged who then go on to make statements that show they did not even attain a GCSE in the subject, but did I even claim to have a degree in statistics? no I said I was currently doing a degree in the subject.
If I wanted to make up random shite I would claim to have won the fields medal for mathematics and to have won £7million lifetime earnings from poker and secretly be one of the worlds best but cant give away my identity it all done on other sites and because i lost with AK vs AJ 3 times in a row this definitively proves sky poker is rigged trust me I am a world class poker player and mathematical genius.
Did I do any of that ****? no so stop acting like i did.
Fwiw I don't think "bots" is the right word.
I assumed you're talking about programmes designed to make optimal decisions so the player doesn't have to.
The post seems to be about players who make sub-optimal plays yet are profitable in the long-run.
I would suggest that 27 games is a very small sample to draw a conclusion like this from. Some of the best players on this site would lose money over 27 games - likewise even the worst players may be profitable over the same period.
https://www.primedope.com/tournament-variance-calculator/
Here is a simulator so you can see how a "bad" player can make money over a small sample.
a fish can be profitable over a few hundred games. I think either I did not write clearly or you did not read clearly but I have never intended to imply 27 games is a meaningful sample. I have also used that simulator before with a little bit of revision i could make a random walk in python to do something similiar to that anyway.
well i could program a random walk before would need to revise to do it again.
I think people have issue with your assumptions.
"I am assuming of course that the plays they make which look terrible are actually not. for reasons that I cannot grasp, I am talking about plays such as calling off half their stack when 25 blinds deep heads up with j7 off suit."
Why are you deciding that this must be a profitable play and also why are you deciding it must be a bot?
Apologies - was my fault - I didn't read it clearly.
But it wouldn't be a bot issue.
You're doubting the credibility of the random card generator OR the player is using technology to manipulate the software.
* I mean OR in the mathematical sense
You haven't really provided anything to discuss, so it just comes across as a bad beat post disguised as feigned concern over bots.
Were there bounties involved in all cases?
Was there ICM involved in any of the cases?
Did the J7o call a 25bb open jam or had they raised/3-bet? (doesn't make it a good play but sometimes there are people who feel they are pot committed even if they are not and just want to try and get lucky to finish it off)
While there's no doubt more than a few questionable plays made by various people I've yet to encounter anything I would consider to be 'bot' like play.
"There are three kinds of lies: lies, damned lies, and statistics."
It's a phrase describing the persuasive power of numbers, particularly the use of statistics to bolster weak arguments.
Oh and they're just the kind of player anyone who wants to make a bit of money playing poker NEEDS at their table. Just hold on tight though because as you've seen, it can be a bit of a wild ride.
I think I may have got a bit silly with this one not as one of my silly but maybe worrying about nothing.
firstly their win rate could be a lot higher and they had quite a few things they were doing well as well as some absurd plays so maybe its a human and not a bot the absurd plays are leaks that stop them winning a higher % of the time?
further I guess Maybe I have to rethink statistical significance for myself 3 standard deviations above random chance is something I take as evidence of statistical significance and that is probably correct because I am highly unlikely to get results like that by pure chance if I should not be winning.
I wont ever mention names on forums if I am going to be talking about suspicions of foul play or someone been a bad player. I have been accused of been a bot myself in chat before which I kind of liked in a weird way because I don't rank myself as an awesome player just a profitable one and someone calling me a bot was like saying your amazing dude lol. I replied "yes I am borg" because I thought that was funny. but How would I feel if people made forum posts citing me as a bad player and laughing at me? actually I know exactly how i would feel because it happened at or wrongly accused me of cheating? I may mention player names in a one on one chat or on a table with only six players and maybe I shouldn't or maybe that's okay but I kind of think its wrong to do on the forums for everyone to see.
However considering that I sharkscope 30/40 players a day I give myself ample opportunity to find players who can be 3 standard deviations out from their expected value.
I have two big fears in poker and sometimes I think I overestimate these fears.
1. bots artificial intelligence killing the games and making it impossible for humans to win
2. government legislation changing and making it impossible to make any decent money from the game.
sometimes I can be irrational about these fears but these remain an ever concern for me in terms of poker.