Rigged: The AA test

Boltneck is completely right, I couldn't have said it better.

Although all I came here to say is that if we are going to have this discussion yet again can we do it somewhere else? It's off topic to this thread and doesn't belong here.

lLol. Nick e-mailed you? Oh sure, I mean after those 12 quality posts you made 6 months ago EVERYONE has been awiating your return with baited breath. I personally have been hounding Nick to get you to come back. Too bad you won't be staying long though. We 30,000 poker pros will miss you sorely, even if you aren't an expert.

Thx LD

Attached Images

Rigged AA.JPG (26.7 KB, 191 views)

?????

What are you talking about? Did you read the story about the Absolute Scandal? It goes directly to the former CEO of the company Scott Tom!!! Either you don't know about this topic or you're spinning bro'.

This is NOT about individual's who were cheating. It never has been. It's about "rigging". Look up the definitions of cheating and rigging and you'll see there are overlaps, but in my opinion "rigging" more closely fits the Absolute Poker scandal, as it does with the Full Tilt Bot Scandal.

You're right that we can't prove all the implications of what has been uncovered so far, but the guys who uncovered these things certainly presented cogent arguments that there was something fraudulent going on. It really all boils down to how one interprets the facts as we know them.

Rigging, at least as everyone whines about it, is about a poker site giving you AA and someone else KK and making sure the KK wins. Other examples are you get AA hit a set of aces only to lose to a person playing j-10 suited and catching a str8 or a flush.

The absolute scandal had nothing to do with rigging in that sense. The only thing that happened in the absolute scandal was a player was able to see everyone elses hole cards. He had no control over what cards came and I dont believe in any way it was "rigged" unless you want to call that rigging which if you do then go ahead but that is not what the rigged discussion, or this thread, is about at all.

So either you dont know what your talkin about or your spinnin bro.

You know, I don't totally disagree with what you just said. My idea of "rigging" might not be exactly what this thread is about, and I appreciate your civility in pointing that out. All I'm saying is "rigging is in the eye of the beholder". If someone conspires to create a situation in which they might be able to win any given hand on command, I'd say that's rigging.

If a software team was to invite me into a brainstorming session on how to "rig" the software, I wouldn't suggest that KK wins over AA as you said. I would say, make the software such that at any given time the house could win a hand if it so chose. Period.

Let me give you an example. We all know that all hand histories are saved. We save them ourselves. We also know that there exists software that can replay those save HHs. Ultimatebet recently added a replay feature. So, if hands can be replayed, they can also be "re-used" in situations where the house wants Seat 3 to lose to Seat 6. Compared to some machine control code I've seen, that would be fairly routine for a high level programmer.

So, what else might be possible. When I worked in a software team, I can remember many a meeting where someone from Marketing or Manufacturing or Field Support would come to one of our meetings and when someone asked them if they had any issues, they would say, "would it be possible to........(fill in the blanks)....."

The answer from the development team was always the same, "anything's possible if that's what you want us to work on."

So, my idea of "rigging" is to make the software such that X can beat Y at any given point in time.

When you look at it that way, we aren't disagreeing all that much.

In all honesty, I'm just trying to avoid being naive about this whole thing.

what has this thread turned into, this was a good thread but it seems to be on a downswing. The whole point of this thread is to see how often aces are dealt since some people say that sites deal bigger hands to increase their rake. So this was a test to put that theory to bed.

OK so back to original post how does that prove online poker isnt rigged?

That you are only getting the correct percentage of monster hands?

People lose hands and cry its rigged, I think they are just crying.

As for multiple monsters on board yeah it seems weird. Sure I play more hands online alot more than live but I just havent seen so many mosters vs monsters (flushes, boats, quads) as I do online.

Then there are the free rolls.
I sit there in a free roll for 2 hours and get nothing but rags. Ill play multiple SnGs in that same time from beginning to end and go through all ranges of hands.

Live games I seem to make pretty steady progress, online is alot more of a roller coaster but I thinks thats due to online donkiness (people tend to take dumb gambles and they some times suck you out).

L never keep statistics myself but i do see aces kicked what seems to be alot bu then again no hand is going to be an all in winner all the time and i do love checking your stats out besides im not ver good at it and not even sure how you all get them anyways good luck all of you and take care

I think that this is a brilliant idea, and glancing down at just the first few posts is enough proof that online poker isn't rigged, over time you still win what you should and don't get rivered every single time, like everyone seems to believe happens on sites such as pokerstars or Full Tilt Poker.

However, they'll still be people commenting on this thread and saying that proofs nothing, etc etc... let them moan, they only moan because they are the players who lose money overtime playing poker, gl at the tables all...


no1yidmax

It makes me wonder sometimes with all the times that I have had high pairs. Some nights I receive them over and over, and other times I can't get high pairs to save my life.

It doesn't prove that on-line poker isn't rigged, but I think it demonstrates pretty well that if it is, then it's done in some other way than manipulating the shuffling algorithm. I am fairly convinced that this at least is not the case.

In fact, I'm amazed at how accurate the RNG at Poker Stars is. In over 300,000 hands we see that our contributers have been dealt AA EXACTLY the number of times one would expect. A simple test yes, but why should anyone think that any other hand should be over or under represented. If this ignorant assumption is wrong, what reason do we have to believe that any of the other equally paranoid assumptions would do any better.

The truth of the matter is that with the use hand tracking software as widespread as it is, any significant monkey business on the part poker sites would be too easy to expose.

I agree 100% and I think you all just proved that the RNG is not the issue. The RNG makes for a convenient scapegoat, but in my opinion ever since Reliable Software Technologies cracked the early RNG in 1999 such that they could predict every card in the deck in real time after seeing the first few cards, I think it's safe to assume that all the major players were forced to make the RNGs more robust. Plus, it seems like the shuffle algorithm has been independently audited at most poker sites.

But as of this date, I haven't seen any mention of the high level code (that sits between the RNG and the user interface) being audited at any site. That's where the rigging would take place.

So, while this thread does prove what the author set out to prove, I would argue that it doesn't prove whether or not there is any kind of rigging going on that isn't RNG specific.

See interesting info here re hand instances. Click on data set description in upper right hand corner.

UCI Machine Learning Repository: Poker Hand Data Set

OK, so if 86,000,000 shuffle combinations are not enough for you, read this link on why some random generators' codes may be broke by a geeky poker player. It's all way over my pointed head.

How We Learned to Cheat at Online Poker: A Study in Software Security

GERAT INFO guys keep up the good work.

Looks like you guys are putting a lot of effort in to this, hats off.

full tilt

14,748 hands

expected 66.7
dealt: 69

wow i run hotter than kathy liebert

I think you're the first FT contributer to have more dealt than expected. Thanks Comb, I'll update the chart when I get home.

Another 15000 hands compliments of Combuboom. It pushes Full Tilt a full 1% in the right direction. Whew! I was starting to worry.

Attached Images

Rigged AA.JPG (27.7 KB, 137 views)

More hands since I posted those other AA results:

Site: Pokerstars
Number of ring hands: 7962
Dealt AA: 30
Expected: 36
I am dealt AA: 1/265.4 hands
Won: 93.33%

Good stuff. Interesting statistics...

But, for arguements sake, lets say a site did want to "rig" the game in order to produce a larger rake profit in ring games.

Simply monitoring the number of times your dealt a specific hand (AA,KK,whatever) and then comparing that percentage to the actual statistical probability wouldnt show you anything. If I were the person(s) responsible for the "rigging", I wouldnt screw with something that would so easily be detectable. Think about it. Its like a magician doing tricks. While your looking here, hes doing it over there.

I would be far more interested in a more telling statistic. How many times does the actual statistical favorite hold up? In other words, does the expected percentage match the actual percentage? Im sure that the poker tracking software can tell you, but the number of hands required to get a reliable percentage, at least in my eyes, would be over a million, for a single player. Ive read several articles that claim that the stats hold up, but have never actually seen the hard data.

Ive seen some wierd things happen in online poker. But I dont think its because its rigged. I think its because its online poker and its to be expected. Factor in all the loose play, complete idiots, and radical number of hands played, and youll come to find that online poker is its own beast. Nothing controlled by computers will ever be as good as the real thing. Just ask all those porn addicts.

Right, that was I was saying earlier.
If you have AA and someone calls your raise and you get sucked out well to bad. Thats not rigged just a donky getting lucky.
Besides how much money does that get on the board anyway compared to boat vs flush vs quads.

Like I said earlier yes I play more hands online than live but it still doesnt add up.

I get quads almost everyday on Bodog, sometimes more than once. Only had quads twice in live play.

For those who play on Bodog how many times you see trips on the flop (example 888), multiple times a day right. How many times does that happen live? I seen it once.

Live play I have seen 2 monster get into it alot but not as much as online. Online 3 monsters at once is no surprise.
Live I have only seen it once. A high straight vs A high flush vs K high straight flush (I had the K hi str8 flush).

So no I dont think its rigged to screw your AA but I do have to wonder if its not tweaked for rake.

As for the Freerolls when its sponsored by the site I just dont get cards.
I think Ill trfack my cards the next free roll Im in and post them.
If Im in a Cards Chat or other sponsored freeroll I get a normal range of cards.

OMG! I just spent the better part of an hour respondind to the last 2 posts only to lose it all when I switched from quick reply to advanced. This always happens when I use the laptop. I don't have the energy to re-create it now, maybe later.

Table updated for ChuckTs last PokerStars contribution.
Site: Pokerstars
Number of ring hands: 7962
Dealt AA: 30
Expected: 36
I am dealt AA: 1/265.4 hands
Won: 93.33%

Looks like you got shafted Chuck. But you made up for it with a fantastic win rate.

Attached Images

Rigged AA.JPG (26.7 KB, 121 views)

win

Really wish I remembered stat, but wikipedia refreshed my memory. How about a test to see just how likely it is that poker is rigged.

Null hypothesis: AA(expected) = AA(actual)

Pokerstars:

AA(expected) = 1497.14
AA(actual) = 1491
Total hands = n = 330869

P = .0045063152
P(hat) = .0045248724

One proportion Z-test

z = (P(hat) - P)/sqrt((P(1-P))/n)
z = 0.159

P = .8736

In English, if we assume poker is not rigged, there is an 87.36% chance of seeing these results or stranger (further away from the mean). If this were to be under 5%, we would say it was sufficient evidence to show that online poker is indeed rigged. Since you cannot prove a negative, you can simply say that this data fails to reject the null hypothesis that poker is not rigged. There is no reason to believe that poker is rigged based on this data.

For full tilt, I'll spare all the calculations, but

P=.061, 6.1% chance of seeing data like this or stranger. This is awfully close to say we are bordering on being able to claim that full tilt is rigged by giving less action hands (or perhaps because the sample is skewed towards better players, they don't need to give action to get them to keep coming back... I'm kidding )

For ipoker:

P=0.69 = 69%.

Total:

P=.2816=28.16%

Note that for example 28.16% has nothing to do with the fact that there's a 28% sites are on the up-and-up. It's just that IF poker sites are not rigging, and we did this sample an infinite amount of times, we'd see this much variation from the mean 28% of the time, certainly not low enough to conclude the site is rigged. Full tilt's 6.1% is rather low though. As mentioned, usually 5% is the threshold of suspicion, and almost all confidence intervals (ever see the + or - after the Presidential polls? Most likely they use a 95% confidence interval, which would get a P of 5%). So it's low enough to raise suspicion.

Note that just because the number of hands at full tilt is lower does not make this number less reliable, because n is factored into the z formula. As n gets higher for the z to stay the same the mean has to get closer and closer to the expected mean.

There is also the assumption with this test of independence that p*n > 10 and (1-p)*n > 10.

The first is basically the expected is > 10 because p can be expressed as expected/n, and the multiplication cancels the n's out. All expected values are over 10, so the samples can be said to be independent. Or normal, or whatever that's supposed to conclude. Also I believe there's a restriction about the sample being a sufficiently small sample (at most 1/10 or something) of the entire population. I think we can safely conclude that CC members have not been involved in over 1/10 of all hands at pokerstars, full tilt, ipoker, or all combined.

So in conclusion there is no evidence whatsoever that pokerstars or ipoker are rigging their sites, and there's no evidence there's a mass collusion against the poker players by the sites. BUT, there is some evidence that full tilt gives less AA hands than it should. But also remember this is hardly even close to conclusive proof. If we took 17 samples of this size from full tilt, we'd expect to see this one even if they weren't rigged. More data would surely help determine if they're RNG is in fact rigged intentionally or not to give players less AA hands than it should.

Dont confuse what I posted in my first reply with a willingness to believe that online poker is rigged. Its not. I dont think that seeing monster hands go against each other more than youre used to seeing live has anything to do with the sites or the way they are run. This has to do with the high number of loose players seeing alot more flops. Seeing more "sets" and "quads" has nothing to do with the RNG or the site, its the number of players willing to stay in a hand with weak pairs that accounts for this.
A good example of this is the occurance, or lack thereof, of ace high winning a showdown. I cant remember the last time I saw ace high take down a pot in an online freeroll. It just doesnt happen very often because of the large amount of players who will see a flop with any two cards.
Not that I see it ALOT in my live play, but it does happen. When is the last time any of you played a live TOURNAMENT with more than a $100 buy in where at least 5 players saw almost every flop? It doesnt happen nearly as much as it does online. This is the reason you see all these freak showdowns, suckouts, and multiple monster hands. Its not the sites, its the players playing on them. This can account for almost every complaint or conspiracy theory Ive seen yet. Looking at the stats of hands dealt just helps to confirm this.

eh, there's not a kathy liebert's chance in hell that any site would rig their program to give out less AA hands. this doesn't take a statistics background, just any sense of practicality. from a statistical viewpoint, the fact that it can be that much less over a (not huge) sample just further reduces the probability that it's "juiced" in that particular way

additionally there's a possible response bias, because we're not collecting the data from every member of the site first hand. it's just whoever wants to do it. partaking in this sort of study and giving the results is just naturally going to be more appealing to someone who feels unlucky, because they're probably more curious about that kind of thing (and more prone to think online poker is rigged). so it wouldn't be a shock if the samples contributed are more likely to deviate to the low end of receiving AA over all their relatively short runs

Like I said, counting the number of times you get specific hands wont tell you anything. The only stat that will tell you if online poker truly is on the up and up is the expected probability of a hand winning versus the actuality of the hand winning. And even then, there are variables in the actions of the players that account for so much of this stat, that it would be impossible to prove one way or the other, without just dealing a million hands and everyone staying in to showdown. But thats not really poker is it?

The important thing to remember is this. If its online, its run by a computer. And if its run by a computer, it can be controlled by someone, somewhere. That doesnt mean that it is, just that its POSSIBLE. Even live poker can be manipulated. But its a much harder trick to pull off when youre seated in front of the 9 other people youre trying to cheat.

Ah, good point. I was thinking that since it was preflop and not dependent on going to showdown that it could be thought of as an SRS, because although I'd assume CC to have better quality players, I wouldn't expect them to get AA any more or any less. But the fact that a person's short-term luck may affect them participating in the study hadn't really crossed my mind.

Of course you know what that means. Since it must be skewed towards more, because the people who participate are the unlucky ones, pokerstars must actually give more AA than our average as well, just as in the full tilt. And you know what that means...

omg pokerstars is rigged to give action hands. Our total scientific analysis that used logic invalidating our scientific methods have proven it. Run run, get all your money off pokerstars.

lol and yes I'm joking, although this fact does show that perhaps this entire test is not free of bias and thus cannot be on its own trusted as proof or even evidence either way of poker being rigged or not.