Organizing a project to determine which sites are legit or rigged

[ryanj247]

i understand all that. i think he'll offer more specific details if someone asks who isn't wasting his time. based on the information he's provided so far, even if it raises some questions, i just feel that it merits someone looking at it in more detail.

edit: didn't mean to suggest you were wasting his time with your questions in the other thread, just that i think he was providing general information in the forum, but would likely provide specifics to a qualified individual who contacted him privately.

[helter skelter]

[ QUOTE ]

well, here's the thing: that analysis HAS BEEN DONE. the result was conclusive.


[/ QUOTE ]

Sorry, I'm not sure what you are talking about here. Are you saying an analysis has been done to conclusively prove that an online site would profit by rigging their software to keep fish happy?

An analysis would have to show that the rake gained from returning fish would more than offset the rake lost from disgruntled pros. I don't know how an analyst would model that, but if that has been done, I suppose that would suggest motive.

I am new to this debate, so I have just read theories of other types of rigging, like having rigged house players sitting in games. That type of rigging would require the same type of software development and oversight, though. Again, there would have to be cost/benefit analysis.

Even if it is proved that there is a benefit to rigging, the cost of exposure is so great that the benefits would have to be pretty large.

The problem with the whole project of gathering and analyzing the data is ensuring the integrity of the process. How do you ensure that the HH's that are provided by volunteers haven't been altered by the providers. The only way would be to have the sites themselves independently provide those same HH's to the project manager so he could verify them. I just can't see that happening.

Unless the providers of the data are beyond reproach, the conclusions reached by the analysis couldn't be trusted.

[ryanj247]

no, the analysis i was referring to was the one done by a 2p2er concluding that favorite hands are winning less than their expectation in all-in confrontations.

if the results of his tests are accurate, then it seems unnecessary to even ask the question of whether it would be sensible or profitable for a site to purposely cause such a thing to happen.

[helter skelter]

From what I understood of that, there is some question about the process, since it showed the same for all sites.

In any event, it would be a worthy project. Like I said, the problem is with the process. To ensure the integrity of the process, it has to be audited and for that, you need the co-operation of the sites. If someone could get the co-operation of the sites, then I'm sure there would be people willing to volunteer their time for data collection and analyses. However, even if a site wanted to co-operate, there may be legal and/or ethical reasons why they couldn't, so we would all be reduced to trusting the data providers.

[nineinchal]

[ QUOTE ]
From what I understood of that, there is some question about the process, since it showed the same for all sites.

In any event, it would be a worthy project. Like I said, the problem is with the process. To ensure the integrity of the process, it has to be audited and for that, you need the co-operation of the sites. If someone could get the co-operation of the sites, then I'm sure there would be people willing to volunteer their time for data collection and analyses. However, even if a site wanted to co-operate, there may be legal and/or ethical reasons why they couldn't, so we would all be reduced to trusting the data providers.

[/ QUOTE ]

Agreed, I think it would be a worthwhile project to the 2+2 universe.

Now how the hell do we do something like this so that we know that we have a fair game when we play on line?

[ryanj247]

i don't know. from what i can gather, there's no reason why a proper analysis can't be done on a set of data from one or more players saved HH. i can't see why we would need cooperation from the sites in order to do a simple analysis on HH. if a statistically inclined individual verifies the testing procedure MT2R used, repeats the procedure on another set of data, and the results are the same, then that would be that. it may not be definitively conclusive. some may question the integrity of that process. but it's something that i would like to see done nonetheless, just as a way of resolving the matter of whether MT2R's results are accurate.

[Misja]

Some tests can be done very easily.

For instance: Count all cases where a player went all-in. Of course discriminate between different stakes, number of opponents, etc. Then count what percentage of those times the all-in player won.
Do the same procedure for another site. Compare the results.
If there is like 5% difference for over 1000 cases, then there is something to investigate.

You really don't need millions of hands for that, your own database is probably already enough.

[Henry17]

[ QUOTE ]
Some tests can be done very easily.

For instance: Count all cases where a player went all-in. Of course discriminate between different stakes, number of opponents, etc. Then count what percentage of those times the all-in player won.
Do the same procedure for another site. Compare the results.
If there is like 5% difference for over 1000 cases, then there is something to investigate.

You really don't need millions of hands for that, your own database is probably already enough.

[/ QUOTE ]

What? This test couldn't be any more flawed.

[StitchNV]

I am extremely interested in putting some form of dedicated organization that examines this issue in full depth.

So far the agenda would be a compellation of a lot of the questions brought up by this forum.

I have registered the following domain for the use of this project. www.BetSure.org (the name just seemed fitting)

If you would like information regarding the deployment of this project please just PM nineinchal, myself, or email BetSureOrg@gmail.com for more information.

So far there are a lot of great ideas that I think will really please a lot players in almost every aspect of online poker (from fish to pro rounder)

The project is going to be non-profit and beneficial to all members.

[Misja]

[ QUOTE ]
What? This test couldn't be any more flawed.

[/ QUOTE ]

Well, tell us what could be improved in it then! [img]/images/graemlins/wink.gif[/img] That is the whole purpose of this thread, to get some discussion going about what tests could be done.