Bad Beats Post

obv this is not a "bad beat" issue.  This is a discussion grounded in statistical analysis of a large sample.  Try to be open minded here and look at the hard numbers and facts, and lets have an open discussion of these issues.

 Not that it makes a huge difference but its 1 in 221 not 220.

 For tournaments I have 139,344 hands played and ironically the pair that I have had the least is AA.  Each pair should have shown up about 630.5 times.  I had AA 590 times.

I saw KK 618 times

AK suited and off a total of 1695 expected was 1681.

KK and AK are well within range but I was surprised how far off AA was.

obv really small sample size but still.

My cash hands I posted earlier were on ftp btw.

I've posted some things similar to this in the past and found it swiftly moved to the "bad beat" section as well. To me these are legitimate questions, correct or incorrect, and pocketfives is doing everyone a disservice by basically moving you to the "crackpot weirdo" section for suggesting something could be out of whack on certain sites. There would obviously be a huge upside to the poker sites for rigging hands, so why is it so preposterous to suggest this?

The problem is that every publication/site involved with poker is so determined for the game to "make it" that they turn a blind eye when it comes to whistle blowing. Card Player will report various scandals after the fact (Ultimate Bet, Absolute Poker, etc), but they're sure as hell not going to tell you ahead of time to steer clear of certain sites. But this attitude destroyed boxing, and nearly ruined baseball as well - everything can't be about escalating profits and ignoring any potential problems.

It's been over a year so I don't remember specifics but I was so incensed by one scenario at one point that I ran the numbers on over 400,000 hands on Bodog, looking for results in one specific test group: someone all in preflop against me with A-x, and I'm holding A-K. I was appalled to find out that against A-4, for example, I was running about 50/50. When I emailed them with my findings, their explanation was that because they also have an online casino, there's a large number of "gambling" types and thus more loose play. Which obviously doesn't answer my question at all! I sent back another response that I'm not talking about how often I get the call, or who moves in one who, or who reraises, because in this example it doesn't matter: this is a hot/cold proposition, you're just seeing how two hands fare in a showdown, both players all in before any other cards come out, and no one else in the hand. My exact quote was, "it doesn't matter if it's Phil Hellmuth, or some player from my home game, or some 'gambling' player who wandered over from your casino, A-K is not a 50/50 hand against A-4." But of course all they sent me back was the standard form letter.

Again, maybe you're wrong, and maybe my 400,000 hands aren't enough of a sample size, etc, but sweeping you into this "bad beat" section is pocket five's way of marginalizing you. They get kickbacks on clicks, and ads, etc, and they think they're hurting their bottom line if they air this stuff. But I don't know about you - Poker Starts is doing really well, but it seems like all the other sites I'm on have fewer people than they used to and shrinking guarantees.

To show you how far off you are (and statistics sometime seem counter-intuitive), I'll take the exact example you give here which you said would not be alarming: a 10% deviation over this sample size.  So here are the inputs:
1/82 expected rate.
236317 sample size
2881 (expected AKs) - 288 (10% of expected) = 2593 occurances, which is 10% off the mean.
Plug this into the equation (or calculator on the link) and you'll see that this case would not "easily" occur.  The probability is a miniscule 0.000003% that this would occur naturally... not exactly as common-place as you initially thought.  So mathematically speaking, the sample size is large enough to draw a strong conclusion from the deviation I am seeing.  And once again, note that it is compounded by seeing this for ALL THREE of the top three hands.

a2z1to3:

obv this is not a "bad beat" issue.  This is a discussion grounded in statistical analysis of a large sample.  Try to be open minded here and look at the hard numbers and facts, and lets have an open discussion of these issues.

 I dont understand why researching the hands dealt would indicate the site being rigged as that would seem to target particular players.  I would be way more interested in seeing how many times you were the shorter stack and lost with AA or KK or how many times you had AA and someone called an allin with A5 spades and hit the flush on the river.  Or how many times you lose on the river card.  Situations like that would seem to imply that a site could be rigged a little to speed up games. 

I agree.  As I look to the ad banners (for online poker rooms, including Absolute) at the top of the page and to the left, its no surprise that this type of discussion is marginalized.  There is a LOT of money flowing around in the industry.  All of the large forums also make most of their money by ads from poker sites.

We know for a fact that rigging occurred on at least 2 large sites (UB and AP).  For anyone to suggest it is not possible again is very naive.  We also know that it was concerned and intelligent PLAYERS (not the sites) who discovered it by anaylzing their hands mathematically and discussing it openly.   Statistically, they found for example that the results virtually HAD to have been due to an oppenent who could see hole cards.  And even in that case, had the cheater been mathematically smart enough or less greedy, he could have skimmed profits as a super-user for much longer without exposing himself.

There is most definately conflicts of interests and lack of transparency in online poker.  Like you say, it is ironic that this ultimately is a bad thing for the game.   Transparency and regulation should be welcomed, as should OPEN discussions like this one.  I am a profitable and skilled player.  The last thing I want to see rigging of any kind in online poker.  But I would much rather see A LOT of light shined their way than too little.  I agree that boxing and baseball have had similar issues.

I was not even suspicious of the amount of AA KK or AK i had received.  My initial concern was the exact same as yours.  It seemed to me like dominated hands won at least 50% of the time.  I understand that statistically, I can lose 20-25 SNGs in a row with hands that are fairly big favorites, and that only analysis of the larger sample means much.  So my intention was to look at the stats of similar scenarios that you described. AK vs Ax, etc.  But since that took more effort, I just on a whim decided to look at something very easy to see, right there in my HH database: the number of premo PPs vs expectation.  Again, since they occur so rarely it would be difficult to even notice without looking for it.  I just looked on a whim because it was an easy thing to do... and sure enough:  98% get more AA than I do... fine, I said.... probably just bad luck.  But then 98% get more KK than I do.  Ok... so I run bad... really bad... still possible.  Maybe I also get fewer 33?  Nope.  AA the least, KK second least.  Hmmmm.  Ok... so I'm sure it cant be the same for AK I said, but I checked.... 98% of players get more AK than I do.   Its hard playing with my hands tied behind my back when virtually every single oppenent gets dealt more of the big three hands than I have, assuming no others see this.  Definately not looking good.  I'm still looking into the other scenarios such as AK vs Ax, but need more time to dig into the results of such specific hands.

At any rate, it is sad that this gets put into the "bad beats" board, as it is obviously not even about beats.  I didnt mention a single thing about beats or RNGs.  But like I said, this can not be an unbias site by definition, since they get their funding from the very sites in question.   Thanks for the feedback and intelligent discussion.

toddm902:

 I dont understand why researching the hands dealt would indicate the site being rigged as that would seem to target particular players.  I would be way more interested in seeing how many times you were the shorter stack and lost with AA or KK or how many times you had AA and someone called an allin with A5 spades and hit the flush on the river.  Or how many times you lose on the river card.  Situations like that would seem to imply that a site could be rigged a little to speed up games. 

lol, that happens to me in 85% of my tourneys and I have a huge hand beat and it is always on the river.

PLEASE look at small stack vs big stack all-in pre situations on pokerstars, big stack wins 101% of the time...

 you do realize that the only way to get knocked out of a tourney is for a bigger stack to take you out?

So you probably are just remembering these instances.

 make sur your playing on Poker Stars.

Since reporting this to Stars, I have lost my last 36 of 37 (and 46 of 50) hyper-turbos, the majority of which were to bad beats such as my set running into runner-runner gutshot straights, dominated hands, etc.  While I know that losing 46 out of 50 hyper-turbos is possible and 50 in an of itself is not a worthy sample, this anicdote supports the larger picture I'm presenting.  Through the first half of this year I was on multiple shark scope leader boards for total profit at the SNGs.  So I'm not without some level of profitable skill here!  But there is a clear and dramatic point where my results have free-fallen over many thousands of SNGs, and 1-for-37 runs such as the current one, comprised of many improbable beats in a row, have been common-place for some time now.  

When I wrote stars, they did take the time to personally respond which was appreciated.  But, for some reason they chose to only look at hands from the past 45 days, a microscopic sample (and which still showed me underdealt top hands).