If Stars finds something and pulls the "here are three fiddy as a refund" we can't settle for that. Asking for refunds of actual money lost to these accounts plus rake paid while playing with them plus getting intel on operator to press charges should be something we fight for together.

If I can help with SPSS let me know, my DB is not large enough to provide stats.

Hi Willy, I appreciate what you did but such query should be only used to make a preliminary list of players to check, its no sort of proof.

Slightly altered query.
1. AGG% =42-46 look at my OP
2. BB 3bet / SB 3bet isnt that great to use, I will keep it to myself why.
3. Squeeze / 3Bet ratio = over 0.85
4. Check-Raise (HU pots) flop = over 13
5. Flop Fold to CB vs 1, SRP = less than 46

The way its done, you make a preliminary list of players based on few factors. The group of course has those factors/stats in common for them, after all thats how you picked them. Then you have to test that group of players using uncorrelated stats, for example if you were preselecting them based on aggresion %, and checkR on flop,turn,river you should test them further on their cbetting strategy IP and OP. Only if they pass the additional tests based on as many stats as possible you can conclude that they are part of the ring.

I have checked around 10 new accounts from your list and those are all false positives. And for that reason I am not reposting the list anymore.

For now I'd be happy if these accs either just get banned or at least a reasonable explanation could be provided that they are not botting/colluding. It seems to me that there is not yet enough hard evidence that could prove beyond doubt (in the sense that it would be enough for Stars to ban them) that they do bot/use some kind of a software/collude but hopefully Stars can get more insights from their DB and other stuff they can look at.

here is few most obivois spots abouth their stats, and this is that kind of thing no1 else of good/winning regular does if he is not suspect already. sorry for my bad english, just trying to prove few of most obivous points on those bots situation who are still playing much normal tables, there is much more samekind of spots in their stats too.

https://www.youtube.com/watch?v=O38Kj6KZlt4

Yes I didn't intend it to be any sort of proof so shouldn't have posted it.

1. AGG% for Fedor is 38 ..this is some notecaddy compiled stat but rest of the players have it around 42-46 range
2. I'll take off BB/SB 3bet ratio
3. I'll use 0.85
4. Wouldn't lowering this stat cause more false positives -
don't they all have this higher than 13 on HU pots ?

I created new stat check-raise flop / check-raise turn ratio.

Takes some time to process all the hands. Will pm the results tomorrow to avoid false positives.

Can we crreate a new thread only with FACTS?
THere is a lot of spam here more than a BBV.

This is my wife's account. You won't find me or her playing at the same time as I work very long hours and she plays poker throughout the day. She has multiple sclerosis so I am not allowed to relay what my wife is saying? She's down about $80k since she joined so how on earth can she be cheating. My own poker accounts are my own business and I play a lot higher than my wife.

I mainly play on winner or William hill poker sites or Ipoker network.

It's a sad day if my wife losing $80k is some kind of bot!! So why should I step back? It's not a crime for a husband and wife to have different accounts and at no time have we ever played each other.

I play live 95% of the time as I'm not comfortable with online poker.

Anyhow good luck all at the felt & I hope pokerstars can address all the issues raised in this thread. At the peak you have over 200,000 players so at least they are aware that a small number of PLO regulars could be BOTS. At least give them a chance to investigate what looks like collusion.

I guess you didn't notice my ipoker PLO bot ring thread from over 6 months ago...

http://forumserver.twoplustwo.com/38...94/index4.html

Dat All-in- EV disparity...

(That was iPOKER I couldn't find any disparity on Stars)

Well done OP and others.
I added an informative quote by the OP to the OP.

The total bot count stands on 24 on PokerStars: 16 accounts are gone and 8 are up and running. I post a spreadsheet with stats for those still playing together with 2 banned for reference. The second spreadsheet is the difference from average for each of them. I have included every single postflop stat from RPTR apart from Cbet and FvCB, those are a combination of others and therefore useless.




Grethe and some others are operating on old data sets. I have set up a Gmail account schweindigs@gmail.com, please send me all the usefull stuff and instruct me if I should keep it private or not. I will share everything on Google drive on my account and make it public for everyone to copy but not to edit.

Schwein you've done an excellent job man.

Btw it's obvious but if some 200 regs are running insanely below EV it could be a sign that the suspected players are sharing hole cards.

Anyone received answers from Pstars regarding this matter?

+1

Hello Simon,

Thank you for your report. Your email was escalated to me as an expert in bot detection and as a member of the PokerStars Game Integrity Team.

Please rest assured that we do not tolerate automated players (bots) on PokerStars. We have an extensive arsenal of detection tools in order to ensure that each player is a human being.

We are currently conducting an investigation of this issue. This may take several days to complete and we appreciate your patience during this period.

If you have any further information to provide regarding this matter, we will take it into consideration.

PokerStars will advise you of the outcome of the investigation as soon as possible.

Regards,

PokerStars Game Integrity Team

But Schwein wrote before that :

Using the spreadsheets from post #335 I have calculated the squared distance from average stats for all 21 stats from RPTR. For additional reference I have added a totally legit Russian grinder N@T@L@T just to show how much different he is from all those botters.



And another chart showing the distance (difference) between account pairs.



First 12 columns are botter pairs, where RockfellaJ and BOOTTLEGGER are already banned.

Next 6 are legit accounts pairs where I compared legit grinders Nakamator, Shuller_A1t, N@T@L@T, Bagrovui with each other.

Last column is alleged botter SuSaaNiN compared to legit grinder N@T@L@T.

Small distance means accounts are super similar, large distance means they are very different.

Oh sorry didn't see that , if i understand correctly Schwein created an Alias at his DB for all the suspected accounts ? If they indeed run alot below EV in a large sample i don't think it's irrelevant.

great work schwein

Missed this earlier

This is Reg tables without HU




And 200 zoom without HU, couldn't find the post to quote that someone asked for

it would be great to see how is the smallest squared distance from avg bot stats for player who we can be sure that he is not a part of this bot ring

But it is impossible to do without some automatic method for collecting stats for many players

I am afraid it is because its just my database, without any additional data mined hands and mostly from zoom tables where those suspected accounts rarely played on same table. I made a suggestion how to check properly if they are sharing holecards and shimmy started working on that.

yeah i'm waiting for more hands to arrive, hopefully sometime this week.

this is off topic, but how exactly does russian ptr track zoom? you can only observe like 1% of hands dealt, so how the **** are they grabbing every hand? maybe stars should look into that

they just hacked stars client to let them observe any of zoom tables , not only hands with big pots which we can observe

The bright side is that their services is helping to indentify bots

Dunno if this is possible, but we should suspect that bots never "misclicks" like humans right?

If we have several 100k hands on some of the suspects and they NEVER checked back nuts or minbet nuts on river as a misclick or called river with nuts or coldcalled a 3/4b with a total crap hand that would deff add to the suspecion no?

fwiw



$-wize they run a bit below, evbb-wize a bit above.

Not sure this would tell us anything, it's likely there is human input involved IMO.