NL Bots on Full Tilt

[vetman81]

[ QUOTE ]

He does have a lot of posts in the software portion of this site... and has a good knowledge of programming from the looks of it.

[/ QUOTE ]


Nation earlier:
[ QUOTE ]
Chuck is a good guy, and to be honest, doesn't even know how to configure his virus protection; I had to do it for him.

[/ QUOTE ]

A few of nlnuts' posts would suggest otherwise:

one

two

three

[$crooge]

[ QUOTE ]
[ QUOTE ]

Ok, I know I'm a low-post person (so, according to a poster early on in this thread, I have no validity) but there also seems to be evidence that the bot owner created a new account on 2p2 to try to divert attention away from himself. Why would an innocent person do that?

[/ QUOTE ]

That was a friend who plays with me. He created the account once I told him of the thread. I decided not too post until I talked to nation who advised me too come in here and defend myself.

[/ QUOTE ]

He seems to still be posting. Do you think he is helping your case?

[MyTurn2Raise]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
chuck, i want to know why none of you guys try to improve your game. ie. why do you all play the same style, you're playing break even, if you're all sitting together playing shouldn't there be some serious discussion on how to play hands better?

[/ QUOTE ]



We play for rakeback and low variance. The system we use provides both of these. it is a low winrate, but it's enough for us.

[/ QUOTE ]

yeah, but you guys play so many hands. you should be improving if you play that many hands. only a robot or completely unmotivated person plays the exact same break even system without making their game better than/different than the person sitting next to them. do you guys not have brains? are you only capable of playing a set system of hands a set way?

[/ QUOTE ]

not exactly true

last summer, I shortstacked over 200k hands of fullring on OnGame for just above breakeven at the table because the bonuses were so good and I could use economies of scale

I needed the money then and was very happy with only slight tweaks to the system.

[KotOD]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
chuck, i want to know why none of you guys try to improve your game. ie. why do you all play the same style, you're playing break even, if you're all sitting together playing shouldn't there be some serious discussion on how to play hands better?

[/ QUOTE ]



We play for rakeback and low variance. The system we use provides both of these. it is a low winrate, but it's enough for us.

[/ QUOTE ]

yeah, but you guys play so many hands. you should be improving if you play that many hands. only a robot or completely unmotivated person plays the exact same break even system without making their game better than/different than the person sitting next to them. do you guys not have brains? are you only capable of playing a set system of hands a set way?

[/ QUOTE ]

I take it you've never heard of Johnstown prior to this.

[ackbleh]

The first thing that crossed my mind after reading the OP was 'these might be bots, but probably not. This certainly isn't conclusive proof'. After reading a good portion of the replies to this post, I'm absolutely astounded that so many people are so quick to agree with the OP's assertions.

Look, there are a lot of people out there that do a lot of things that seem odd or unusual to you and me. I would find it odd to have 6 or 7 accounts in the names of friends, running 3 or 4 accounts at a time on multiple side by side computers, so that I could 20 table 1/2 NL at party... but I know it's been done, as early as a couple years ago. And quite frankly, if someone were doing this, I expect you'd see exactly what you are seeing -- accounts that do not play against each other, show identical type stats, tend not to chat, and are slow to take actions simply because they have 3 or 4 mice (and screens) in front of them.

When the suspected accounts did adjust their play to your attempted exploitation...and fairly quickly, really... that's evidence that supports the presence of a real person playing the accounts. Instead of fairly considering all explainations, you only considered the less likely alternative that the bot programmer was sitting there intently watching all of the games and then instituted some type of manual override, but only in pots that you are in. Seriously, do you think people program bots and then constantly watch them play? Then why do they spend all that effort to program their bots if they're just going to sit there and watch them play anyways? It's far more likely that you have a serious multitabler situation, where either one person is playing on multiple accounts or one person has trained others to play the same basic playstyle with hard and fast preflop betting rules.

And you know what, it could be that this is a bot and the programmer was sitting there watching a new version or something and had a well developed override function, I suppose. But that's hardly the only explaination, and not even close to the most likely explaination in my opinion.

Similar playstyle between different non-competing accounts is no proof of bottery, and I actually think Full Tilt has shown intelligence here in understanding that. Where they may have shortcomings is in not having appropriate tools to identify interaction characteristics of bots. Because the ugly truth is that some people play very rigid, rule-based poker that is very difficult to distinguish from the play of a programmed bot. The more reliable methods of determining if an account is controlled by a human or a bot involve analysing the method of interface with the application... things like mouse movement and clicks, window focus shifts, etc. These are things where a human being is going to behave by necessity with some natural randomness where mimicing this randomness will be fairly difficult, and often overlooked, by bot programmers.

OP, please don't take this post as a big ball of hate, cause it's not...you're entitled to your opinion obviously... I am just surprised at the lack of dissenting opinions being presented in the forum of a supposedly informed community. Maybe these accounts are controlled by a bot. Probably not, in my opinion. But the information presented here is certainly not a case of proof beyond reasonable doubt, and certainly not enough to close accounts and confiscate funds.

One potential benefit of this thread is that it could bring attention to what sites really need to do to detect bots -- active programming that analyzes the nature and location of button clicks, mouse movement, and general interfacing with the software. Full Tilt pretty clearly does not have this type of technology. I highly doubt anyone else does either, very specifically including Stars... at least not an active basis. It's possible a site may be able to access data on an account after suspicion has already been raised... I doubt most if any sites have this right now, but they should... but I'm fairly confident that none are doing active detection work on all accounts as they play. The sites simply don't have any motivation to get it done.

[nlnut]

[ QUOTE ]
Chuck,

Forget my previous post.... - what I would like please is a screenshot of your PT stats for all the players linked together to show that they don't have identical stats....



Vava

[/ QUOTE ]

This would be impossible...we do have the same stats because we play the same strategy. Not against FTP T&C

[fjbourne]

[ QUOTE ]


[ QUOTE ]

He does have a lot of posts in the software portion of this site... and has a good knowledge of programming from the looks of it.

[/ QUOTE ]


Nation earlier:
[ QUOTE ]
Chuck is a good guy, and to be honest, doesn't even know how to configure his virus protection; I had to do it for him.

[/ QUOTE ]

A few of nlnuts' posts would suggest otherwise:

one

two

three

[/ QUOTE ]

hmmm

[jcg2005]

nlnut care to explain what this thread was about?

[IRuleYouHard]

[ 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

[DWarrior]

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

[/ QUOTE ]

Could it have anything to do with the 1.5 month lock he experienced recently?