Reminder: Absolute poker and Ultimate bet were already caught cheating.

How can someone be absolutely certain that online poker is not rigged?

Cheating has already happened. It's been done. What makes you think that they can't or wouldn't cheat in other ways?

I hope that most sites are fair most of the time. I know people make livings at online poker, but to say that a poker site cheating is out of the question, is in my opinion ridiculous.

Ah, ok. No, these are all the tournament hands in my data base in which I was dealt a pocket pair and subsequently saw a flop. I didn't filter for any pre-flop action, number of opponents, or anything else.

Here's what I did:

First, I filtered for the times I was dealt a pocket pair AND saw a flop. No other filters. That resulted in 1260 hands.

Then, I retained those filters and added four more. Under the Hand Value tab in filter editor, I went into Any Flop Hand Value and selected the following:

Three of a kind - Set - High set
Three of a kind - Set - Second set
Three of a kind - Set - Low set
Four of a kind - With Pocket Pair

That resulted in 105 hands.

Actually, the first time I did it, I forgot to include the Four of a kind filter, but I caught it before I posted.

Your suspicion is incorrect. I suspect a phrasing problem.

Edit: Just to be sure, though, the filters I set for the 1260 number were these (this is an HEM database):

I went into the Hole Cards (HE) section, deselected all hands then selected Pocket Pairs.
Then I went into the More Filters section and added Saw Flop=True.

I assumed that this would filter for those hands in which *I* saw the flop, but now that I think about it, I suppose it could filter for any hand in which there was a flop. I don't think it does, but I'd definitely like to know if I'm wrong.

FFS read at least some of the thread.

First of all, someone with backdoor software access and rigging an RNG aren't the same thing.

But much more importantly, I don't think anyone here is saying that online poker is 100% not rigged, or that sites can't or won't cheat, or that cheating is out of the question. I've never seen one single person say that.

However, we do have a lot of people who are 100% convinced that online poker is rigged, without any evidence. That's what you see what I would consider reasonable people (and they consider shills LOL) arguing against ITT.

Personally, I fully expect that other scandals will come to light from time to time. Be it big collusion or botting rings, an insider cheating customers, or even a poker site itself doing it. One must always be careful where one plays, and trust in the vigilance of fellow 2+2ers to find the scams. That's why I think it's important that people air their grievances. But I also think it's important we separate the wheat from the chaff, so when a real scandal comes to light, it isn't missed among all the ridiculous conspiracy theories. This thread is meant for the chaff, but hopefully the debate that goes on helps some people understand why their theory ends up in here, and also helps people understand what they should be looking for.

That is all certainly true. I certainly understand your point. You discount all unknown cards exactly the same way. And oppenent hole cards and undealt cards are both equally unknown until we have some information about those cards. The moment those players go to the flop we know something about those cards. They are no longer random. They are (presumably) above average holdings. Namely they have a high value bias.

Let's take the extreame opposites Heads-Up compared to full-ring:

Heads-up:
- never more than two unknown hole cards will go to the flop with you. Thus a miximum of two cards will be removed. As well as the fact that the threshold for seeing the flop is reduced thus reducing the preiumum value bias. Both of these reduce the card removal effect.

Full-ring:
- potential for 18 unknown hole cards will go to the flop with you. But if you like you could compute the average. It will surely be more than 2. And because of decreased VPIP stats in full ring compared to HU play the premium card bias for those unknown cards that do see the flop is increased. Both of these lead to increased card removal effect.

What I do not have the answer for you on, is to what MAGNITUDE this difference exists. And that is why I found myself unqualified to answer your question. I do have a tendancy to perhaps overanalyze people's questions. If I did so here I apologize.

Good points, for sure.

The people who helped to prove Absolute Poker and Ultimate Bet cheating are pretty much the same people who are arguing in this thread that the shuffling in online poker is probably legitimate.

The quoted poster being one of the key players.

Well, that does make some sense. But there is also the greater number of people at the table to consider. The more players there are, the more likely it is that somebody will get a wild hair up his butt and decide to play 83o. This would tend to at least partially negate the card removal effect, since the main assumption of the effect is that if a flop is seen, the players in the hand are likely holding premium cards.

Not at all. This topic might be useless in a practical poker sense, but the act of thinking about and discussing these effects can't be. In my opinion, at least. Thanks for your thoughts.

I agree and it would do the game so much better for all if those obsessed with superbots and switch button off and on unguarded RnGs would think about these genuine threats instead of whether cashing out may cause mystic curses.

No doubt when the next scandal takes place, whatever it is, riggedologists will cling to it like a newborn even when it will have nothing to do with the RnG aspect of the deal.

You are correct, the Saw Flop = TRUE filter shows all hands where a flop was dealt and you were still in the hand.

I am not sure what the best filter to use for just 'A flop was dealt' is. One that certainly works is to select 'Players on flop is bigger than zero' in the 'more filters' options. Another is 'flop pot size is > 0' in the same tab. There doesn't seem to be a way of doing it directly, though!

Not for the purposes of making an EV decision in a hand, because you are always calculating the probabilities explicitly there, not using any kind of approximation.

The only time the CLT comes into play is if you want to calculate what a 95% confidence interval is of your actual winrate, given an observed winrate in HEM; or how lucky/unlucky you have been, or something like that. And all of those things are solved by simply getting a suitably large sample size.

You could realistically run into problems if you were, say, trying to calculate your risk of ruin when you only play large MTTs with top-heavy prize structures because sums of these take an extremely long time to look normal because of the skew and kurtosis in the underlying distribution. A 'suitable' sample size for MTTs is much bigger than for other forms of the game, and what's more, using the normal approximation will always underestimate your risk of ruin.

I seem to remember that guy got some obviously ludicrous result like 17 standard deviations away, then he disappeared when his error was uncovered

didn't you miss all the possible full house combos?

No amount of working on your game is going to help if you are getting ripped off. Hand histories are not going to show jack squat unless the gouging is outside the range of probability. We saw this during the Superuser scandal. The moron who got his hands on this decided to get greedy and stupid which showed in the hand histories that what was going on was almost impossible. I read that he was making impossible calls and folds that were un characteristic of even the best poker player. If he would have been a little smarter, you guys would have never caught him. But greed and stupidity are easy to spot. Not everyone or every organization is nearly as stupid as these guys were.

The moral of the story is that you should frame your view and your decision on more than just hand histories. This may be a leap of faith in some situations because there is no way to know with absolute certainty what is really going on behind the scenes or in the programming.

What the heck do you mean by this? This is not a term I am familiar with.

And as I've pointed out many times before, and will continue to happily point out to all of the nut heads, if there's a real phenomenon at work that has an effect large enough that your brain was able to discern the pattern I guarantee it can be found with statistical analysis.

It appears he might not have counted the specific full houses that include a set. So count all full houses, and then exclude those that have triplets on the board. I don't know how that filter works in the software he is using.

Why is your standard for burden of proof set at "absolute certainty?" At that standard you would be taking literally everything "on faith."

Heck, you would be taking it on faith that you should use "absolute certainty" as a standard for burden of proof as you cannot know with absolute certainty that that is the proper standard to apply!

Donko, since you're the boy who cried wolf (you admitedly like to tweak the debate here and there) its hard to see if you are serious here. But you are coming dangerously close to full rigdology here. If there is something going on - it will show up in the hand histories. It has to. Spadebidder has gone to great pains to show this. Just saying: don't base it just on the handhistories is nonesense, at least when it comes to rigging. Or, if its not nonesence then you need to back up your claim. If you don't know the math (nor do I), then you have to look at the arguments people like spade, pyro, QPW and Josem (among others) throw out there and see if they seem to be making sense. But the debate has to begin and end with the handhistories (barring some admission from someone working for the site).

Again, there are plenty of people now with massive databases who look at them VERY closely. Just because you don't have a big enough database does not mean there are not many many serious poker players who do. Even without the study people like Spadebidder are doing. You are a scientist, you should know better!

No problem, I guess if the world is after specifically you in ways that are impossible to detect then it is not your fault that you are losing.

Quit poker.

Not sure what else you want to be told at this point.


No, the moral is either work on your game to fix the holes, or quit if you think mystical forces are at work against you.

Since you seem to think the latter is in effect, just quit.

Same thing I would say to a guy who thinks people at McDonalds plot about him when he is there. Just go to Burger King.

Not very complicated.

So how great would you say the uncertainty is in live poker, where you are playing with unknown opponents and an unknown dealer, and the cards are either shuffled manually or go down in a hole in a black box and come back shuffled? How about outside the few dozen regulated U.S. casinos (assuming you live there), what about all the other games people play in? What about casinos in all the soverign island resorts or tiny countries run by a monarch? Do you ever have absolute certainty of a fair game in any poker game? I'm not trying to disparage any of the places I mentioned, but I guarantee there is cheating going on in live poker rooms somewhere right this minute. I'd also bet that somewhere today an operator or manager of a live poker room got a cut of some money won from cheating. I'm not saying it's common, but it happens.

Ack! You're right. That brings the total up to 112. Not as bad, but still over 3 standard deviations below expectation. I think that should do it. No straights or flushes are possible on the flop with a pocket pair, so 112 should be the final answer. To summarize:

1260 flops seen with pocket pair
100 sets flopped
105 sets and quads flopped
112 sets, full houses, and quads flopped

Thanks for catching that.

That's interesting. I've been to a few sites that talk about and help you calculate your risk of ruin, and they're not all the same. This might help explain why.

You need to exclude the full houses that have triplets on the flop, as those are not included in the 11.76% expectation of hitting a matching card on the flop. Or if you count those, just add 0.23% to your expectation and use 12%.