pandafreeroll
Enthusiast
Silver Level
Hey there, everybody!
I have repeatedly come across people floating suspicions regarding the fairness of online poker site's card dealing systems.
Because I care about all of you and your concerns as well as the climate surrounding online poker, I have decided to definitively address the issue to the best of my ability, but since that will require quiiite some effort, I first want to check out whether anyone is genuinely interested. For that purpose, I will now only try to draw a preliminary sketch of what I will include in a follow-up post – please do let me know if you want me to go ahead or if you have any questions!
The plan would be to…
- Explain how our personal experiences are prone to psychological distortions by relating findings of empirical research in psychology to elements of Poker and why, as a result, subjective assessments are not a sufficient measure by which to judge the fairness of RNGs – regardless of how extensive the experience they draw upon.
- Point to a statistical method (specifically the chi-squared test for goodness of fit) that allows us to double-check whether the objective observations (hand history records) corresponding to our subjective experiences do in fact contain patterns indicative of unfair card dealing and sharing my understanding of how it works.
- Explore where my approach can be applied by describing basic statistical problems encountered in poker and further provide examples of contentious situations and how the fairness of their observed outcomes can be objectively measured. That is, to outline what results I can produce and how they can be interpreted - as well as their limitations, i.e. how they cannot be interpreted.
- Finally, I want to offer to anyone willing to share their hand history with me to perform the statistical analyses I can and share the results, either privately or publicly. I would be using Python to convert the hand history files into .csv and SPSS for the analyses. Unfortunately, SPSS is far from affordable and I can only keep my hands on a licence through university, so in case this thread/topic gains traction, I will have to try and brush up my python skills and/or scoop for a script on github to copy+paste. Then, I could share some everybody can use to replicate the method with their own hand history if they like, otherwise I will share the python string converter and SPSS syntax anyway, the 30-day free trial should do. All that will take some time tho, I am kinda busy anyways!
Thx for reading and thx for letting me know if you are interested!
I have repeatedly come across people floating suspicions regarding the fairness of online poker site's card dealing systems.
Because I care about all of you and your concerns as well as the climate surrounding online poker, I have decided to definitively address the issue to the best of my ability, but since that will require quiiite some effort, I first want to check out whether anyone is genuinely interested. For that purpose, I will now only try to draw a preliminary sketch of what I will include in a follow-up post – please do let me know if you want me to go ahead or if you have any questions!
The plan would be to…
- Explain how our personal experiences are prone to psychological distortions by relating findings of empirical research in psychology to elements of Poker and why, as a result, subjective assessments are not a sufficient measure by which to judge the fairness of RNGs – regardless of how extensive the experience they draw upon.
- Point to a statistical method (specifically the chi-squared test for goodness of fit) that allows us to double-check whether the objective observations (hand history records) corresponding to our subjective experiences do in fact contain patterns indicative of unfair card dealing and sharing my understanding of how it works.
- Explore where my approach can be applied by describing basic statistical problems encountered in poker and further provide examples of contentious situations and how the fairness of their observed outcomes can be objectively measured. That is, to outline what results I can produce and how they can be interpreted - as well as their limitations, i.e. how they cannot be interpreted.
- Finally, I want to offer to anyone willing to share their hand history with me to perform the statistical analyses I can and share the results, either privately or publicly. I would be using Python to convert the hand history files into .csv and SPSS for the analyses. Unfortunately, SPSS is far from affordable and I can only keep my hands on a licence through university, so in case this thread/topic gains traction, I will have to try and brush up my python skills and/or scoop for a script on github to copy+paste. Then, I could share some everybody can use to replicate the method with their own hand history if they like, otherwise I will share the python string converter and SPSS syntax anyway, the 30-day free trial should do. All that will take some time tho, I am kinda busy anyways!
Thx for reading and thx for letting me know if you are interested!