NL Bots on Full Tilt

[DWarrior]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]

WTF, I'm rusty on my statistics, but are you going to tell me that a .5% difference in VPIP over 100,000 hands is 3 SD's? [censored], post your math.

[/ QUOTE ]

(14%)*(1-14%)/sqrt(100000) = .038% = 1SD

Of course it's higher than that because all 100,000 decisions are not independent identically distributed.

[/ QUOTE ]

I think the test statistic formula is .005/sqrt(.14*.86/100000) = 4.6

Assuming independence, and I'm just using a textbook formula for population proportions.

[/ QUOTE ]
We just went over this last week at work..
(sigma) = squareroot( (sumation((x-change of x)squared) divided by n-1

[/ QUOTE ]

I'm not sure what this refers to.

But can't we also use the Goodness of Fit test for this?

We have 4 suspected bots in the OP's screenshot:

VPIP1: 14376/105366 = 0.136438699
VPIP2: 15840/112514 = 0.14078248
VPIP3: 11683/82577 = 0.141480073
VPIP4: 5721/41414 = 0.138141691

The total VPIP is 47620/341871 = 0.139292306

So if they are indeed botting, the expected hands played of the 4 bots should be:
14676.67
15672.33
11502.34
5768.65

respectively

Throwing it into my calculator for GOF, using 3 degrees of freedom, we get a p-value of .0108, which means there's just over 1% chance that this happened by chance if we assume the 4 players were using a similar strategy.

[BluffTHIS!]

[ QUOTE ]
[ QUOTE ]
1. they play eactly the same strategy and play as a team on many postflop hands, which is within t&c. you're allowed to have 2 players to a hand.

[/ QUOTE ]

Post flop?? You're way to experienced to think that anyone is going to believe that.

[/ QUOTE ]


I brought the post-flop point up before and he didn't respond to it unless I missed it in the avalance of posts. The level of complexity expands by orders of magnitude with each added street of betting. This is the achilles heel of their "defense" in explaining the stats.

[nation]

i think it's completely possibly to avoid tilt and play exaclty the same when your friends are next to you and have a stake in your results. remember, if one guy plays differently and loses, his friends could get pissed because its partially their money. playing together and sharing results are very conducive to a strict system.

[nlnut]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
nlnut care to explain what this thread was about?

[/ QUOTE ]

The thread I created 2 days after my account was suspended...what inside the thread are you questioning? Please be specific.

[/ QUOTE ]

I'm guessing he's wondering why you deleted your initial post in that thread.

[/ QUOTE ]we decided to just let FTP do the investigation...not 2+2.

[rurollin]

[ QUOTE ]
[ QUOTE ]

3. brandonjoseph is one of the guys, i just cleared it up with chuck and brandon. if you read his posts, he was just defending his friend. he has little experience with online message board etiquette and what types of conversation are accepted and encouraged. i was the same way when i started posting.

[/ QUOTE ]

So BrandonJoseph is a big liar then? He is one of the guys that had his account locked? I assume he is one4thethumb?

[/ QUOTE ]

Oneforthethumb is he...? [img]/images/graemlins/cool.gif[/img]

[dp13368]

Dwarrior, this just takes into account VPIP. I think the more relevent information here is the astounding similarities between every single stat (vpip being one of the least complex).

[mikechops]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
lol at people saying that the numbers aren't close enough to be conclusive... they are nearly identical for every stat on every street.

as for the arguers of the 14.24 and 13.67 stats, this could easily be from the RNG dealing one bot AA-TT some small percentage more than the other over the sample. 100k is large but not large enough to smooth out all randomness in the game. These numbers are so close it's not even worth aguing about it...

[/ QUOTE ]

The VPIP #'s are well over 3 standard deviations apart. That isn't close or anything like it. Of course that doesn't proove anything and there could be other explanations, but I think a subtle difference in PF strategy is more likely than a bug in the RNG.

[/ QUOTE ]

WTF, I'm rusty on my statistics, but are you going to tell me that a .5% difference in VPIP over 100,000 hands is 3 SD's? [censored], post your math.

[/ QUOTE ]

I did earlier, but binomial distribution SD = sqrt(n*p*(1-p)) = ~ 110 hands or 0.1%. (n=100k, p=0.14)

If we take the difference of two such distributions we need to multiply the SD by sqrt(2) or around 0.15%. So a difference of 0.58% is actually close to 4 SDs apart.

[SukitTrebek]

Chuck,

Nice to see you again. I'll start off with this question -- Why do you suck at poker? With very very extensive history between us, why did you let me keep reraising you for weeks without taking a stand?

Also, did Full Tilt completely exonerate your accounts or did they place you on some kind of monitoring or probation?

A picture of your play in pokertracker and pictures of the other three players can make this misunderstanding go away pretty quickly.

-Trebek

[fjbourne]

[ QUOTE ]
I wanted to address a couple things that have come up a lot in this thread:

1) People are wondering how a computer can click the buttons, enter data, etc, and saying that maybe its a program telling people what to do, with a human doing the data entry. Over 5 years ago I was using a program called WinRunner to automate testing of websites. This program would see a window, automatically 'learn' where all the buttons and data entry fields were, and allow me to program input and mouse clicks in specific time intervals, etc.
This is absolutely the type of program that is being used. You don't even need to be able to 'see' the buttons. I could program it to just 'click' in a certain space on my screen. So if you set up your tables in the same spots, then there you go, buttons are always in the same place. I could program it to wait for other input to show up on the screen (like the cards) and react certain ways to whatever it saw. This isn't even new technology. No humans are needed for this.

2) You people that are saying the stats are 'similar' and maybe these guys all just play the same style, or that they are all in the same room using the same set of general rules to play ARE JUST CRAZY! There is no way that one individual person could record stats that EXACT over 2 100K hand sessions if they tried. Like, we aren't just talking about vpip and pfr. EVERY stat in pokertracker was identical. Every single one. I'm far from an expert, but my background is engineering and I've done a ton of statistical analysis. That is just too large a sample size for the numbers to be that exact.

PLUS you have to factor in things like tilt that come into play in poker. Are you telling me that even two guy sitting next to each other reading from the same 'playbook' of how to play, could stick to that plan so perfectly that things like tilt would just never take over?

I don't like to rush to judgement on anything, but the stats ALONE are just unbelievable enough to me for 3 people to have so exactly, and then when you factor other things in like they getting up from tables, never playing together, names linked to pittsburg, etc, the evidence seems overwhelming to me.

Just think about this: If you wrote down all these rules of how to play pots, and you and a friend tried to execute them over 100K hands, do you really think you would stick to the script SO perfectly with no other human emotion involved that you could come out with the EXACT same stats? I've played poker before, so I just don't believe that is possible.

Mark

[/ QUOTE ]

A+++++++

These exact thoughts were going through my head

[nation]

One big thing you guys continue to miss is that all players had to send in ID's and utility bills to FTP in the beginning of the investigation.

All accounts were fully exonerated.

So keep that in mind, because all of this discussion is to ease your minds that these accounts don't have bots playing on them. They've already been cleared 100% by FTP, this is for your benefit and whether you believe them or not has no bearing on the fate of their accounts.