NL Bots on Full Tilt - Page 101 - Internet Poker

darkcore · 05-10-2007, 09:13 PM

Quote:

Also, regarding the timeouts.

They would occasionally timeout in places most people wouldn't, like in a hu pot. The other player would donk the river or something, and it would think and think, then timeout. (And often immediately get up from the table)

I can't remember once them timing out in the more traditional manner where there is a five way pot, they have nothing, got sidetracked on a different table, and forgot to hit check/fold.

IMHO, FT should be able to run through their thousands of hands and find that they timed out way way less than other players who play 6-8 tables.

This piece of evidence is meaningless. Why would a bot not act?

why would a bot not act? that is a very good question.

a friend of mine coded a diablo 2 bot back in the days. basically it would do meph-runs for him and collect all the nifty golden items. overall it worked well, but roughly once in every hundred runs it would just "freeze" before mephi and got killed. we tried for a couple of weeks (on and off with testing obv) to find out why. so if you know the answer, let me know...

damn, i am glad i do not play at full tilt...

1p0kerboy · 05-10-2007, 09:21 PM

DWarrior is making a whole lotta sense. I think we're just going to keep peddling the peddle while the wheel is off the ground here.

But there may be questions left to be answered here. IF the suspected bots are indeed innocent and are just operating a 'poker sweatshop', do we want to find a way to not allow this in the future? Or is it okay?

Hellmouth · 05-10-2007, 09:23 PM

Quote:

Also, regarding the timeouts.

They would occasionally timeout in places most people wouldn't, like in a hu pot. The other player would donk the river or something, and it would think and think, then timeout. (And often immediately get up from the table)

I can't remember once them timing out in the more traditional manner where there is a five way pot, they have nothing, got sidetracked on a different table, and forgot to hit check/fold.

IMHO, FT should be able to run through their thousands of hands and find that they timed out way way less than other players who play 6-8 tables.

This piece of evidence is meaningless. Why would a bot not act?

seriously, if anything its more proof they arent a bot.

Well, *obviously* the players take over sometimes.

Doesn't this seem to contradict the argument that no one could play with stats that close without botting? If they are periodically taking over they must still be able to carry out the basic algorithm without changing the stats too much. Is that even possible?

Greg

Prospace · 05-10-2007, 09:24 PM

Hi have verified through my sources in Pittsburgh that it is not bots. There is 6-8 players in a room. They are paid 8$ per hour or at least thats what it was one year ago.

Our House · 05-10-2007, 09:35 PM

Quote:

Also, regarding the timeouts.

They would occasionally timeout in places most people wouldn't, like in a hu pot. The other player would donk the river or something, and it would think and think, then timeout. (And often immediately get up from the table)

I can't remember once them timing out in the more traditional manner where there is a five way pot, they have nothing, got sidetracked on a different table, and forgot to hit check/fold.

IMHO, FT should be able to run through their thousands of hands and find that they timed out way way less than other players who play 6-8 tables.

This piece of evidence is meaningless. Why would a bot not act?

seriously, if anything its more proof they arent a bot.

I guess you guys missed the part of the theory that talked about how humans sometimes took over during bot play.

See, this is precisely the reasoning that keeps repeating over and over and over in this thread. You guys bring up extremely shaky evidence and evidence that contradicts your argument. When that gets disproved, you guys use the "well, sometimes" excuse to lay that to rest. Then you bring it up 3 pages later.

I think this thread should be locked now, obviously everybody else is locked into their opinions by now.

DWarrior,

You're not catching the point of my posts. I don't have an opinion either way. I'm just as curious as the next guy to find out what is going on with these players/bots. If you scour through my posts, you will see that I've approached the situation from both sides on several different occasions.

The reason I'm asking the questions and/or making the statements that I am is to see what kind of answers are given. There's only one truth to this issue, and we may or may not have covered it by now. Hopefully at some point, it will come out.

Actually, if I were forced to render an opinion, it would be that there is a combination of botting/non-botting going on and at least some sort of shadiness that isn't allowed in FT's T&C.

cardcounter0 · 05-10-2007, 09:35 PM

Quote:

On a side note, if these players do, in fact, discuss late-street decisions together before making them, do you see why their stats would have a much higher likelihood of convergence?

No I don't see why. Could you please explain how introducing another random element helps the stats come closer together? It seems players A, B, and C all choose to diverge from the system an equal amount of times. Then thru discussion, players A, B, and C all arrive at an alternate course of action equally, even more exactly than a bot programmed to bet the river .28357962% of the time would do.

Maybe if on some streets, they could randomly decide to sometimes flip a coin to decide what to do, the stats would be even closer together than opposed to a set script telling them what to do everytime.

Let's test that.
I will program a computer to always raise the river. You can discuss with your friends what course of action to take sometimes, and throw in a random factor or two. Let's see which approach arrives at a more consistent raise on the river.

Players A, B, and C have the most identical stat of river bets out of all the stats. So *obviously* this would be the place that is the least likely to be scripted, right?

**_dave_** · 05-10-2007, 09:38 PM

Been following this thread closely since the start, at first thought OMG BOTZ!!! like many others, now I am pretty much convinced there is NO bots running here.

This is IMO what one could descrive as "wet-ware". One knows how to play, finds local poker n00bs, bankrolls them, teaches "the system", and oversees / coaches to make sure there is no deviation.

To Quote BSG, "There are many copies - And they have a plan..."

IIRC another 2p2 sleuth thought that this thread: http://forumserver.twoplustwo.com/showfl...rue#Post8511500 was the reason 1forthe thimb was dropped from the team, and 0_drunkenboxer recruited - timeline a little late, maybe nlnut took some time before decideing 1forthethumb couldn't be trusted with the monies anymore?

There is no way these are automated bots, despite the stats, which on initial inspection are ridiculously damning. I don't hink this idea has been mentioned before...

What if the 3 team players do have slightly different deviations from the 1.5bb/100 masterplan - but each computer has the login/pass saved for one of the team, and they swap around which account they are playing on?

After all, they are sharing profit/loss so no real reason to only play "your" account - and one of the computers has a better chair!

If they split their play evenly between seats/accounts, all stats converge, regardless of minor discrepancies, no?

Just a thought that may explain the identical stats...

On a more technical not: They TIME-OUT???? never a bot.

Quote:

they're lying through their teeth and they're botting/running a script that tells them what to do

There is a BIG difference between these alternates...

they're lying through their teeth Certainly one of them is a proven liar, this is obvious upon reading the thread.

and they're botting This would be against t&c, a bannable offence on all sites.

running a script that tells them what to do BUT THIS IS TOTALLY FINE!!! I can't find a definitive guide on FTP, but such programs are explicitly allowed according to the list at http://www.pokerstars.com/poker/room/prohibited/, and are not on the list of unfair advantage software on Party's site.

On a more comical note, we should be glad the guy masterminding a mega-tabling poker sweatshop is a 1.5bb/100 FR nit, not CTS running his "system" + work ethic on residents of the Hermosa beach house - now that could devastate the online poker economy faily quickly I would think

KotOD · 05-10-2007, 09:46 PM

Quote:

Those people were not posting it to a thread with 22k views, on a forum that they are unfamiliar with, and where many people that are looking at it think they are scum.

He's from Johnstown, it's a foregone conclusion.

**_dave_** · 05-10-2007, 09:49 PM

Quote:

hey guys. What if there were 3 players and 3 accounts. and on day 1 player a was paired with account a, b with b, and c with c. On day 2, a played on account b, b on c and c on a. day 3 player a played on account c, b on a and c on b. even if player a was a LAG and player b was a TAG, and player c was a LP, all 3 accounts would look the same after 100k hands. Yes or no.

LOL Nick B - you beat me to it - seems perfectly reasonable to me, obv

DWarrior · 05-10-2007, 09:53 PM

Quote:

On a side note, if these players do, in fact, discuss late-street decisions together before making them, do you see why their stats would have a much higher likelihood of convergence?

No I don't see why. Could you please explain how introducing another random element helps the stats come closer together? It seems players A, B, and C all choose to diverge from the system an equal amount of times. Then thru discussion, players A, B, and C all arrive at an alternate course of action equally, even more exactly than a bot programmed to bet the river .28357962% of the time would do.

Maybe if on some streets, they could randomly decide to sometimes flip a coin to decide what to do, the stats would be even closer together than opposed to a set script telling them what to do everytime.

Let's test that.
I will program a computer to always raise the river. You can discuss with your friends what course of action to take sometimes, and throw in a random factor or two. Let's see which approach arrives at a more consistent raise on the river.

Players A, B, and C have the most identical stat of river bets out of all the stats. So *obviously* this would be the place that is the least likely to be scripted, right?

But...my entire argument up to that point has been why it doesn't make sense that the stats would all converge on the late street due to botting...if we agree on the fact that they're not botting at least some of the time. Do you understand that part?

Now, I tried to give an alternative explanation as to why these stats *could* possibly converge. It seems to me that if they made at least some of the decision together, their stats would be more likely to converge than if they each played completely separately from one another. This is because the 3 (or more) of them would arrive at one strategy that would be consistent across their accounts. Does that make sense? Sure, it's still not likely to converge, but it's a lot more likely to converge than the 3 of them playing separately. And, like I said before, they do play without running a bot at least some of the time.

DWarrior · 05-10-2007, 09:54 PM

Quote:

hey guys. What if there were 3 players and 3 accounts. and on day 1 player a was paired with account a, b with b, and c with c. On day 2, a played on account b, b on c and c on a. day 3 player a played on account c, b on a and c on b. even if player a was a LAG and player b was a TAG, and player c was a LP, all 3 accounts would look the same after 100k hands. Yes or no.

LOL Nick B - you beat me to it - seems perfectly reasonable to me, obv

It makes sense for them to converge. However, I don't see why anybody would do that.

stonescar · 05-10-2007, 09:55 PM

Quote:

hey guys. What if there were 3 players and 3 accounts. and on day 1 player a was paired with account a, b with b, and c with c. On day 2, a played on account b, b on c and c on a. day 3 player a played on account c, b on a and c on b. even if player a was a LAG and player b was a TAG, and player c was a LP, all 3 accounts would look the same after 100k hands. Yes or no.

LOL Nick B - you beat me to it - seems perfectly reasonable to me, obv

So they played the exact same amount of hands every sequence? And none of them would adjust to the players they were up against? I don't think the solution to the stats issue is that simple...

EDIT:
And as Dwarrior said, why bother? Sounds like a lot of organization for no particular benefit.

KotOD · 05-10-2007, 09:56 PM

Quote:

that was meant for the turn and river...we didn't slow play flops obv by our stats...

No.

575/2497 = 0.230276332
216/961 = 0.224765869
538/2403 = 0.223886808
411/1836 = 0.223856209

DWarrior · 05-10-2007, 09:57 PM

Quote:

Also, regarding the timeouts.

They would occasionally timeout in places most people wouldn't, like in a hu pot. The other player would donk the river or something, and it would think and think, then timeout. (And often immediately get up from the table)

I can't remember once them timing out in the more traditional manner where there is a five way pot, they have nothing, got sidetracked on a different table, and forgot to hit check/fold.

IMHO, FT should be able to run through their thousands of hands and find that they timed out way way less than other players who play 6-8 tables.

This piece of evidence is meaningless. Why would a bot not act?

seriously, if anything its more proof they arent a bot.

I guess you guys missed the part of the theory that talked about how humans sometimes took over during bot play.

See, this is precisely the reasoning that keeps repeating over and over and over in this thread. You guys bring up extremely shaky evidence and evidence that contradicts your argument. When that gets disproved, you guys use the "well, sometimes" excuse to lay that to rest. Then you bring it up 3 pages later.

I think this thread should be locked now, obviously everybody else is locked into their opinions by now.

DWarrior,

You're not catching the point of my posts. I don't have an opinion either way. I'm just as curious as the next guy to find out what is going on with these players/bots. If you scour through my posts, you will see that I've approached the situation from both sides on several different occasions.

The reason I'm asking the questions and/or making the statements that I am is to see what kind of answers are given. There's only one truth to this issue, and we may or may not have covered it by now. Hopefully at some point, it will come out.

Actually, if I were forced to render an opinion, it would be that there is a combination of botting/non-botting going on and at least some sort of shadiness that isn't allowed in FT's T&C.

Here is how this time-out argument goes over and over:

Person 1: Timing out = bot ldo. THEY'RE BOTTING
Person 2: No, timing out is more likely to be human. Timing out = human ldo.
Person 3: Well, they obviously play some of the time. So they time out when it's a human. Timing out = human, but they're still botting.

EDIT: my point is, sure, scripts can time out, and so can humans. Therefore, I believe this piece of evidence is 100% useless and should be thrown out.

**Alobar** · 05-10-2007, 10:02 PM

Quote:

Also, regarding the timeouts.

They would occasionally timeout in places most people wouldn't, like in a hu pot. The other player would donk the river or something, and it would think and think, then timeout. (And often immediately get up from the table)

I can't remember once them timing out in the more traditional manner where there is a five way pot, they have nothing, got sidetracked on a different table, and forgot to hit check/fold.

IMHO, FT should be able to run through their thousands of hands and find that they timed out way way less than other players who play 6-8 tables.

This piece of evidence is meaningless. Why would a bot not act?

seriously, if anything its more proof they arent a bot.

I guess you guys missed the part of the theory that talked about how humans sometimes took over during bot play.

See, this is precisely the reasoning that keeps repeating over and over and over in this thread. You guys bring up extremely shaky evidence and evidence that contradicts your argument. When that gets disproved, you guys use the "well, sometimes" excuse to lay that to rest. Then you bring it up 3 pages later.

I think this thread should be locked now, obviously everybody else is locked into their opinions by now.

Nothing these guys say can ever prove them right, if they start announcing their strategy I really don't see how that will have any effect on any side of the argument.

When statistical evidence shows they aren't using the same strategy, it's dismissed because of changing conditions. When it matches, it's obvious proof (even though this is even more affected by changing table dynamics, etc, and could easily be the result of chance as well).

Then we have the guys who keep saying these guys are changing stories "a million times", when this couldn't be farther from the truth.

FTP did an investigation, but what does FTP know?

A mod says they didn't do it, obviously the mod sucks, lynch him.

Anyone whose opinions differ from your own is an idiot.

etc, etc, etc.

EDIT: oh yeah, and now the whole world will hear about from newspapers. That will certainly accomplish a lot and will shed a lot of light on this investigation.

excellent post