Online Poker Fairness Evaluation Project


I have been playing online poker for many years and have been a winning player over this time. I have been on the worst streak of my life recently and consequently have questioned the fairness of these sites. Reasearch online has led me to many forums where people are experiencing the same things that I am. The general conspiracy is that these sites have unfair algorithms programmed into their software to keep certain players from winning and also to help new/bad players to win by manipulation of the cards. It is possible that I am just on a very unlucky streak but it has been too long and too often that I am questioning the integrity of these sites. I have contacted the website that I play on about this and they referenced two studies that were done. Neither of these studies look into the issues that many people are experiencing, they are based mainly of the actual/probability of certain cards and how well they hold up. Here they are:

What I would like to see is a comprehensive study on such things as "action" flops, "dominating" hands and how well they hold up, as well as other issues to determine if these programs are indeed fair or not. I might even be willing to compensate people for their time as I am no statistician and could not do this on my own.

If anyone is interested in this project and would like to help, it would be greatly appreciated.

Thank You
Last edited:
sounds kind of interesting if someone had the time. So you would want to see if there is any type of correlation between a player's skill level and the hands they are dealt?

If your theory holds true, you would expect that the lower the skill the better the dealt hands and on the contrary, the better the skill, the worse the hands.

Out of curiosity, how would you plan to collect data? I cannot imagine poker sites just giving this away for free.
Hi Lance,

Correct. But not limited to just skill level but other factors as well. In online poker "hand histories" can be saved and there are many programs available for the storage and collection of this data for analysis. As well some of these poker programs let you save "hand histories" of specific tournaments were you are able to see all the players "hole" cards.

I would also like to see if there is a tendency for more "action" cards then there should be, such as aces, kings, queens as well as cards that connect to make big "hands". I hypothesize that if there were algorithms manipulating the play, they would be complex as to not be noticable and to be blamed on variance of the game itself. I would think if this were the case they would be cards to entice you into play so that you will more easily put your "chips" into the "pot", as is the nature of the game.

Example from a "hand" earlier today:

I have AQo (off suit) vs opponent's 10 10.
The "Flop": 10 10 A
The "Turn": 7
The "River": Q

This "hand" gives me top two pair.
My opponent has quad 10's
Opponent wins.

The odds of getting four 10's on the "flop" is 0.25%

These types of "hands" seem to happen more often than they should with myself on the losing end.

This could be a great project for someone in school or a resume. Any help would be greatly appreciated!
Last edited:
Do you have any research experience in applied statistics or econometrics? I've been interested in doing an econometric study using a Hand history database but not on this topic, as the result will almost certainly be that there's no rigging (i.e. a boring result).

You really shouldn't have the expectation that you'll find anything. There is absolutely no evidence adduced so far (that I'm aware of) that any site is rigged, and the human brain is well known for seeing patters where there are no pattens (e.g. 2000+ gods invented that we've recorded, all with 0 evidence, etc). There are so many "rigtards" (not my word) who all believe there's different ways that the sites are rigged. All with no evidence. You should see the dedicated rigtard thread on twoplustwo forums and read of all their imagined patterns. Too much dopamine (in experiments, whether people will see patterns in completely random dots depends on the amount of dopamine in their brain). E.G. there's a guy there that thinks Q and 3 comes way too much, so he'll always open pre-flop with Q3. Lol.

A priori we should expect there to be no rigging, as there are many tens of thousands of people with hand histories in the hundreds of thousands/millions and it wouldn't take much for a statistician to spend 5 minutes with these hand histories and completely destroy the reputation of the site. Also see:

And it would really only take 5 minutes to do some basic analysis. Each HH is in a nice, small memory .txt and you can grep() the flop or the hole cards really easily.

I tend to notice these types of hands happen more often then I would expect and hence would like to examine the data that is available to find if there is any statistical evidence to suggest unfair play.


Having a pre-determined outcome of the study is bias. It is entirely feasable that these poker programs are as fair as possible and there is no intentional manipulation of cards. These programs would be wise to create as safe and fair of environment as possible such as the casinos in Las Vegas, to attract customers and in turn keep the endless revenue stream flowing as smoothly as possible. Yet, the opposite is also feasable. Through research, I've found that there are many cases of fraud, theft and other criminal activity in the online gambling industry coupled with the fact that these poker programs adhere to almost no forms of independant examination or regulation. I seek to examine the data that is available in a controlled, unbiased study to help conclude whether they are fair or not.

Humans have the ability to discover patterns and assume probable outcomes, this is one of the reasons why we are the most intelligent species on this planet. It is in my best interest to find that there is no foul play among these programs as I love the game of poker and would like to continue playing it online with confidence. But these topics are beyond the scope of statistics, which I intend to use to discover if there are discrepancies between what should happen and what does happen. I would like to keep this discussion strictly on statistical evidence.

It would not take much time to do basic analysis, but what I seek is a study examining the more complex intricacies of the game and whether they hold up to their corresponding probabilites. I will continue to research this topic to try and find any relevant studies that have been done. This is a study that I have been thinking about recently and would need help to create as unbiased of a test as possible. As I have said before I am no statistician and my education in this matter ranges from a few basic courses in college. If anyone would like to help me create this study it would be much appreciated.
Just trying to play devil's advocate :))), if weaker players tend to get better hands than stronger player (at some rate that is not simply due to chance) wouldn't that then suggest that the weaker player's skill level would increase while the stronger player's skill level would decrease, eventually meeting in the middle?

I guess maybe a better question is how would the computer know a player's skill level so that it could assign the better hand to the weaker player and assign a slightly worse hand to the higher skilled player?
Hi Lance,

I suggest there would be a middling effect creating a balance or equilibrium where there are more players playing but not every one is making or losing much money while possibly placing a handicap on certain players based not only on skill but profitability. Such as people who tend to deposit more money onto the poker program increasing it's economy. Thus increasing the the profitability of the poker company by increased use as every time a you play in a tournament or in a cash ("ring") game the poker company collects revenue.

In theory one could create algorthims in the program itself which target certain people based on how much they have won or lost, how often they play, skill level, how often they deposit money etc. and produce somewhat static results. With even more real-time manipulation tactics one could even determine complete outcomes such as placings and money earned.

Example (in tournament play):

Player X will finish 1256th for $0.00
Player Y will finish 124th for $37.50
Player Z will finish 1st for $7,541.68

I appreciate your input as every good test will stand up to reasonable criticism.
I suspect the best way to truly determine if some type of conspiracy is going on would be to get your hands on the actual computer code. Then you could say definitively whether or not there is some foul play.
Hey Derk,

So I guess the first things we need to get our hands on is some good poker anylisis programs and we can use my hand histories and I'm sure we could get people in forums to give us theirs as well.

Do you use any online chat programs?