Organizing a project to determine which sites are legit or rigged

[Sciolist]

[ QUOTE ]
The fact that there isn't even agreement on something as basic as how many hands should be required for such a project

[/ QUOTE ]
It depends how you measure the randomness. Once you have the criterea, you can calculate the sample size needed for various degrees of confidence. If someone wants to run a project like this, they would have to first come up with criterea.

[Sciolist]

[ QUOTE ]
I believe Absolute has been compromised in some way- whether from the inside or out.
What leads me to believe that is actual data and a strange convergence of happenings.

[/ QUOTE ]
Right, and this kind of thing would be impossible for us to catch using datamined hands except for people we have large sample sizes on. Presumably we'd be looking for small biases in the % of time a player wins at showdown and ptbb and so on. You need huge samples for both of these.

[Henry17]

[ QUOTE ]
[ QUOTE ]
I believe Absolute has been compromised in some way- whether from the inside or out.
What leads me to believe that is actual data and a strange convergence of happenings.

[/ QUOTE ]
Right, and this kind of thing would be impossible for us to catch using datamined hands except for people we have large sample sizes on. Presumably we'd be looking for small biases in the % of time a player wins at showdown and ptbb and so on. You need huge samples for both of these.

[/ QUOTE ]

With the Absolute situation since the assumption was that someone could see the other player's hole cards you'd be doing an analysis similar to what they do when they are trying to prove that a corporation back dated option grants. It is a little more labour intensive but not very hard to do.

[StitchNV]

[ QUOTE ]
If someone wants to run a project like this, they would have to first come up with criteria.

[/ QUOTE ]

Project and criteria are pretty much key in this situation. Honestly it sounds like something an origination should handle.

Standard guidelines for which an online poker site should follow in order to be approved?

It sounds like alot of work on both ends (the sites, and who ever keeps them in check)

I guess the alternative is just as acceptable… hand over your cash and HOPE they do the right thing.

[Sciolist]

[ QUOTE ]
With the Absolute situation since the assumption was that someone could see the other player's hole cards you'd be doing an analysis similar to what they do when they are trying to prove that a corporation back dated option grants. It is a little more labour intensive but not very hard to do.

[/ QUOTE ]
How do you do this? How would you apply it to online poker hand histories?

[helter skelter]

Maybe this has been discussed elsewhere in more depth, but

The first thing that comes to mind is to define "rigged".

What is "balancing", how would it be used and why would it be used.

Are we talking about a site identifying individual fish and then helping them in their future sessions by substituting the RNG with a program that analyzes their hand and then gives them a better outcome?

Or vice/verse. Identifying pros and doing the opposite.

I can see where a site might have motive, but it seems that the whole rigging process would be rather complex.

First, you'd have to have software to identify fish and/or pros.

Then, you'd have to have the "balancing" software.

So you have software developers.

Some group of people have to manage the process.

You'd have to have someone watching those people.

You'd have to have someone watching the watchers.

A site would have to weigh the consequences of the secret being exposed against some calculated benefit.

Which goes to motive. A site makes money by raking regulars. Unless a site is just assuming that keeping fish happier somehow profits the site, some individual or group of persons must have done a financial analysis to show that. That's even more people that would know the secret.

I guess I would like to see some proof in terms of numbers that a site would even have a motive for this type of rigging.

I worry much more about a site being apathetic to cheating and collusion than I worry about the site actively participating in rigging.

[Henry17]

[ QUOTE ]
[ QUOTE ]
With the Absolute situation since the assumption was that someone could see the other player's hole cards you'd be doing an analysis similar to what they do when they are trying to prove that a corporation back dated option grants. It is a little more labour intensive but not very hard to do.

[/ QUOTE ]
How do you do this? How would you apply it to online poker hand histories?

[/ QUOTE ]

With stock options you look at the stock price and the date. Basically looking for behaviour that is too unlikely to be coincidence. So if someone was issued options on the lowest stock price day repetitively then odds are they actually were issued those options much later and simply pick that date to put them underwater.

A player who can see other's hole cards plays differently. You would need to think about what those differences are. For example off the top of my head his ability to call down bluffs and fold strong but losing hands would increase.

So you take the hand histories and break them down into similar categories with the difference being the relative strength of the suspects hand. Then you compare them. If a player calls with A high when the opponent is bluffing but folds with A high when he is beat that is evidence. If he folds strong hands that in theory he should call when beat by a stronger hand but calls them in similar situations when he is winning then that is evidence. If you get enough of these situations it becomes conclusive. It is just very labour intensive since you need to make a number of categories and then parse all the hands into categories and compare them.

[ryanj247]

[ QUOTE ]
Maybe this has been discussed elsewhere in more depth, but

The first thing that comes to mind is to define "rigged".

What is "balancing", how would it be used and why would it be used.

Are we talking about a site identifying individual fish and then helping them in their future sessions by substituting the RNG with a program that analyzes their hand and then gives them a better outcome?

Or vice/verse. Identifying pros and doing the opposite.

I can see where a site might have motive, but it seems that the whole rigging process would be rather complex.

First, you'd have to have software to identify fish and/or pros.

Then, you'd have to have the "balancing" software.

So you have software developers.

Some group of people have to manage the process.

You'd have to have someone watching those people.

You'd have to have someone watching the watchers.

A site would have to weigh the consequences of the secret being exposed against some calculated benefit.

Which goes to motive. A site makes money by raking regulars. Unless a site is just assuming that keeping fish happier somehow profits the site, some individual or group of persons must have done a financial analysis to show that. That's even more people that would know the secret.

I guess I would like to see some proof in terms of numbers that a site would even have a motive for this type of rigging.

I worry much more about a site being apathetic to cheating and collusion than I worry about the site actively participating in rigging.

[/ QUOTE ]

all of that makes a hell of a lot of sense. but suppose a statistical analysis was done and the result clearly indicated that the deal was in fact unfair. somehow, you would have to reconcile that with the assumption that it wouldn't make sense for sites to particpate in anything unfair.

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

that being the case, i would like to see people put the debate about whether sites would have any incentive to do this on hold, and instead take a close look at the analysis that has already been done indicating that something unfair is happening.

[nineinchal]

[ QUOTE ]
Maybe this has been discussed elsewhere in more depth, but

The first thing that comes to mind is to define "rigged".

What is "balancing", how would it be used and why would it be used.

Are we talking about a site identifying individual fish and then helping them in their future sessions by substituting the RNG with a program that analyzes their hand and then gives them a better outcome?

Or vice/verse. Identifying pros and doing the opposite.

I can see where a site might have motive, but it seems that the whole rigging process would be rather complex.

First, you'd have to have software to identify fish and/or pros.

Then, you'd have to have the "balancing" software.

So you have software developers.

Some group of people have to manage the process.

You'd have to have someone watching those people.

You'd have to have someone watching the watchers.

A site would have to weigh the consequences of the secret being exposed against some calculated benefit.

Which goes to motive. A site makes money by raking regulars. Unless a site is just assuming that keeping fish happier somehow profits the site, some individual or group of persons must have done a financial analysis to show that. That's even more people that would know the secret.

I guess I would like to see some proof in terms of numbers that a site would even have a motive for this type of rigging.

I worry much more about a site being apathetic to cheating and collusion than I worry about the site actively participating in rigging.

[/ QUOTE ]

"Balancing" would come under my definition of rigged. My working definition of rigged is anything that would cause a poker site would alter to deviate from a live poker game as provided in a casino to create an advantage or disadvantage to one player over another.

Perhaps some of you would like to formulate the the working definition of rigged. Then we can vote on our definition of "rigged" by taking a pole...

[Henry17]

[ QUOTE ]

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


[/ QUOTE ]

No it hasn't. One person did some work on it. A few of us already suspect possible errors and even he admits to some issues with his own work.

He has also been vary vague about the details. When I asked him how many hands he said several hundred thousand which I found very odd. If I done something like this I'd have a specific number or at least be much more specific then limiting it to the range of 200k-999k. He also said it was way off from what would be expected but didn't specifically use any quantitative measures other then "a lot" and "so far off".

Third and probably most damning is that all the sites were bias by the same amount. That is very odd. Even if sites were rigged they would not be rigged exactly the same way. This strongly suggests the issues is with the analysis and not the sites.