Why Online Poker Is Not Rigged

In this post I will attempt to address every argument I’ve ever heard regarding online poker being rigged. Mind you, I do work for a company that develops online poker software but my motivation is not to convince people to trust *my* employer’s software, rather it is to have a central place where people can point misinformed message board and blog posters who think somebody is pulling something on them. Just point them here and end the thread. If you have heard an argument or have your own argument not covered in this post, please email me and I’ll add it to the list.

Though it’s more than obvious to many people that most of the major online card rooms are legit, the internet is littered with literally thousands of claims that they’re somehow rigged. Regardless of how many people point out obvious counter-arguments such as the fact that most online card rooms have their games audited by well respected firms to ensure that the games are fair, people continue to insist that online poker is rigged.

Almost without exception, claims of online poker being rigged fit pretty consistent patterns. For instance, nearly every accusation is presented void of any sort of empirical evidence. Perhaps someone will post a particular hand history of a bad beat but I have yet to see anyone post the results of analyzing 10,000 or 20,000 hand histories to show a true statistical anomaly worthy of further research. Readers are asked to take a leap of faith that this person’s bad beat or his gut feeling about how often certain things happen are indicative of long-term trends.

This lack of empirical data should be the first clue as to the strength of the argument. First because with a large enough sample size (20,000 hands or more) one not need make any argument as the numbers will speak for themselves. But more importantly, it’s the fact that those making accusations don’t even collect hand history data which exposes a fatal flaw in their argument because nearly every serious online poker player uses a software package like PokerTracker to analyze their play. Not using a tracking software package immediately casts serious doubt on claims made by most accusers that they are experienced, savvy, online poker players.

Another common characteristic of online-poker-is-rigged claims is a fundamental misunderstanding of probability. Just because something has a low probability of happening doesn’t mean that when it does, something is wrong. In fact, in many cases, quite the opposite is true. Low probability events are expected to happen given a large enough data set. For instance, if you flip a fair coin ten times, the chances of having a run of ten heads (or tails) is .09765625% (0.5^10). However, over a million coin flips, regardless of how seemingly low the probability of a run of ten heads in a row, you expect to get nearly 1000 runs of ten heads in a row. Unexpected things do happen. As long as the chances are above 0%, there’s a chance that it can happen.

Obviously, this doesn’t apply only to online poker. At the 2005 WSoP, during a televised event, Mark Seif, in two different hands, faced opponents with quads. One flopped quad fives and the other flopped trips and made quads on the turn. The odds against this happening in so few hands that are dealt at one final table are fairly high. Yet it happened. In the 2004 WSoP Gallagher flops quad fives, his opponent with pocket nines pushes all-in and Gallagher obviously calls. The turn and river come 99 giving Gallagher’s opponent quad nines and the winning hand. According to PokerStove, Gallagher was 99.899% to win that hand. Both examples demonstrate that the improbable is not the impossible.

Yet another common type of accusation is that from a player who has recently changed sites or is relatively new to online poker (though they claim decades of live poker experience). In other words, Party must be rigged because I was a winner on UB or I’ve been playing at Commerce for 25 years as a winning player but I can’t beat the $2/$4 at Paradise. Usually buried in this accusation is that the card room purposely rewards poor play. Since the accuser obviously plays excellent poker, the only explanation for losing at this particular site is that the game must be rigged. The glaring flaw in this argument is that there are many, many solid players playing on the same site and winning. If the deck was being stacked in favor of the fish, how do other players work their way up from .05/.10 micro-limits to playing the $30/$60 game? Based on the accuser’s argument we would have to believe that their consistent winning is due to how poorly they play. That’s an easy argument to refute if you’ve ever read their posts on forums like 2+2 and elsewhere. They are anything but fish.

But rather than continue with generalities of the accusations being made, let’s jump right into it.

Claim: More (Quads/Flushes/Straights/Full Houses) get dealt online than is normal.

Fact: Completely false. As this link points out, and as many hundreds of regular online poker players have confirmed for themselves, given a large enough sample size of data, the numbers are within expected statistical norms.

Claim: Online card rooms stack the deck in favor of poor players to keep them playing while allowing better players to be sucked out on.

Fact: Completely false. Again, go look at the actual numbers. Over tens of thousands of hands analyzed by hundreds of people, high card holds up as often as it should, top pair holds up as often as it should, two pair holds up as often as it should, etc. There’s a simple and easy way to test your accusation, use hand histories! The numbers will either prove or disprove your accusations very simply.

Claim: Online card rooms deal action flops to get more money in the pot so they can make extra rake.

Fact: Completely false. One more time, go look at the actual numbers. Use the stats from any casino. They’re all going to come out about the same and that same is right about what they should statistically.

Claim: If online poker rooms were legit they wouldn’t need to base their operations in offshore countries.

Fact: False. Due to the legal climate in the US regarding online gambling, no site can be hosted or have a base of operations here.

Claim: If Enron teaches us one thing it’s that corporations are just out to screw people and this makes them prone to rigging games.

Fact: Debatable. While I, nor anyone, can debate the morality and ethical standards of every casino, the incentive structure for the online casino is highly weighted towards offering a fair game. Fist off, the profits are huge. Go look at Party’s financial statements. Why would one risk the almost instant death of their business if they were found to be rigging games just so they could rake an extra .50 a hand? They would be far more incentivized to just raise the rake than they would be to rig the game. Additionally, there’s far, far more money in going public or being acquired by a larger casino than there is in nickel and diming games. Of course, the most compelling argument is that enough smart people (mathematicians, computer scientists, etc) play poker and religiously track their results that any poker room that was doing something out of the ordinary would be quickly and easily caught thus providing sufficient disincentive.

In summary, the vast majority of online-poker-is-rigged arguments can be easily resolved by allowing others to view your hand histories. If anything it seems painfully odd that in the thousands upon thousands of claims of online poker being rigged, not one single person has offered compelling data which could be scrutinized by others. Poker is a numbers game and the simplest way to test a hypothesis (say, that online poker is rigged) is to compare the numbers against the expected outcomes. And since many people have and continue to monitor such statistics without discovering evidence of any serious anomalies, the burden of proof rests with the person making the accusation. Simply inferring motives or dispensing anecdotal evidence is not likely to receive any sort of serious consideration when access to empirical data is readily available. The lack of empirical data in any claim that online poker is rigged reeks of suspicion.

Links to additional resources:

Perry Friedman and Howard Lederer on Full Tilt’s random shuffling

PokerStar’s description of their random shuffle

PokerStar’s independent auditor’s opinion on their shuffle

Paradise Poker’s independent auditor’s opinion on the fairness of their shuffle

PartyPoker’s independent auditor’s opinion on their shuffle

Kahnawake Interactive Gaming Regulations (Kahnawake regulates many of the most popular card rooms)

How We Learned to Cheat at Online Poker - Though this article is how a security team broke PlanetPoker’s random number generator (RNG), at a higher level it demonstrates that despite a poor RNG the researchers didn’t discover any intentional rigging of the game.

Fluctuation and the Rigging of Online Poker by Bob Dainauski - Although his intent was to prove or disprove theories regarding the cash-out curse, one can easily see why players who start playing on a new site, perhaps underfunded, experience similar results.

Shuffling and Randomization by Daniel Kimberg (CardPlayer Magazine)

Updates

Eddie points out that bot software is used on some sites. True. However, most bot programs are specifically banned by nearly every cardroom. PartyPoker and others have specific policies geared at stopping bots from playing on their sites. Most sites have adopted terms of service agreements that allow them to freeze the accounts of anyone suspected of using bot software on their sites. Obviously, for every move that the cardrooms make to prevent bots, bot writers will try to come up with a counter-measure.

There have been some rumors about smaller sites using bots to fill tables but anecdotal evidence suggests that the bots play softly and usually leave as soon as the tables fill. Not exactly above board but if you stick with the larger poker rooms you can avoid such practices.

10.12.05 I recently saw someone argue that one cannot prove online poker is not rigged. This person is correct. However, one can prove it to be rigged very easily. Since there lacks any crediable proof of it being rigged, despite the ease of proving it rigged via statistical analysis, one can correctly assume it is not rigged. If it were difficult to prove online poker being rigged then I could see the difficulty in the argument that it can neither be proved or disproved (like God) but this is not such a case. This is more like the theory of relativity in which scientists may not be able to prove it to be valid but they can devise plenty of experiments that would prove it to be false.

On a side note, despite readers of this site posting the link to this thread on message boards all over the place (usually in threads where some yahoo is claiming online poker is rigged) and plenty more coming to this site to debate me personally, NOT ONE has accepted the challenge of providing their hand histories as proof of what they claim.