A disturbing trend in my hand history

Hi all,

I'm new here, so I'm not sure if I should be mentioning a site by name. So for now, I'll just leave which site I'm talking about out.

I just did some analysis on my play for the past couple of months.

last year, I kept a hand written log of every single pre flop all in I was involved in and after compiling 500+ hands by hand, there were several things that I found upsetting. This lead me to withdraw my money from that site and start playing on another site. I had started playing online poker on this site and I withdrew 3x what I deposited, so I can't complain, but I couldn't shake the feeling that I should have made more and I was robbed by losing more hands than I should.

Time passed and I lost that hand written log in a move. I have gotten much better at poker. I turn a nice profit at local casinos, both in SnGs and NL cash games. I also started turning a profit on another site playing only PLO8, which is my best game, but is not offered in my local casinos.

So my brother convinced me to deposit again so that we could satelite up to some higher stakes tourneys and go halves with eachother on the winnings....we do this in live SnGs and it really helps to mitigate suckouts because one of us places top three at the casino in 90% of the games we play.

So I've been playing there for several months in relatively low stakes tourneys....I never risk much in online poker.

This time, I logged every single pre flop all in I was involved in that was heads up or three way (most are heads up of course). And this time I did it in excell because my hand written log was difficult to work with.

Here is what I found:

First, I had almost 500 hands where one player was ahead, defined as 60/40 or better....I counted 45/55 type hands as a coin flips and tossed any coin flips asside in the sort.

Second, out of 500 hands, I was ahead about 400 times and behind about 100 times. I was pleased to see that in the history because it means that I have been getting my money in with the best of it...and these hands include the hands where you are forced in with a bad hand at the end of a tourney when you get blinded out. I thought 80% ahead was pretty good anyway.

Third, out of the 100 hands that I was behind, I lost 90% of them. That seemed high to me, since many of those hands are 60/40 type hands...I thought I should have won at least 20% of them, but 100 hands is a very small sample and things could even out with another few hundred hands.

Fourth, out of 400 hands that I was ahead, I won exactly 203 and lost 172...the rest were split.

That's right...I lost 90% of the time I was behind and only won 50% of the times I was ahead.


I realize this sample size is small, but I plan to continue the database until I have over 1,000 hands where there is some advantage. Then, I plan to write a letter to this site asking for an explaination of why the odds in my hand histories are so screwed up. Then I plan to publish my findings to the online poker community and on the forum of the site in question.


I've always thought online poker is less than fair, but this is just sick. And what gets me is that I lose when I'm behind, but it's a coin flip when I'm ahead and I still break even on the site. I have to wonder why it looks like this. If the odds were skewed accross the board, I would still win many of the hands I was behind in. But I lose the vast majority of these....I wonder if the fact that I have withdrawn a large sum of money from the site in the past, or that my current account is slightly higher than when I deposited a few months ago has anything to do with this.


I don't mind playing these unregulated unfair games for low stakes, but anyone who puts more than $50 on a game online is just giving away large sums of money.


By the way....you have to do this kind of analysis yourself. Pokertracker does not offer the ability to look at all of your pre flop all ins, sort them by odds in your favor and against you, and then look at how many were won, lost, or split. You have to look through your own history and create your own spreadsheet. If anyone knows of a program that will save me the time of doing all this myself, please let me know. For now, I take a few minutes after each tourney to enter the data into the spreadsheet.

Regards,
Julio

You'll have to re-do figures deleting the 3-way hands for a start.

I'm skeptical that you did actually log and input almost 500 hands, why do you say "almost 500" then "out of 500" and "about 400" times? Either you've got precise figures or you haven't.

Can you post the spreadsheet? Did you log the actual hands, and work out the actual pre-flop odds? Do you really know what hand confrontations are 60/40 and what ones are 55/45?

If you really have done this work then it is interesting, but you have to present it in a more accurate way to us, and certainly to the poker site as there are HUGE holes in the data at the moment.

Of course you are only the first in a long line of posters that will call into question the validity of the figures or my integrity in compiling them. I posted this for some thoughts and reactions, not for close scrutiny, else I would have written up an executive summary of the analysis and posted the full data. Still, I will respond to your post point by point.

First, there is no need to redo it without the three way hands because (1) there are only a handfull of them, less than 20, and removing them would not affect the overall trend, and (2) I only used three ways in which one player had a 60% or better chance of winning against the other two players, who shared the remaining chance of winning, which, in effect, makes it statistically analogous for the purpose of determining if the outcome of winning matches the probability of winning over time.

The reason I said almost and then 500 is that I did the analysis last night and I'm posting this from memory in some free time I have today. I remembered how many I won and lost because the numbers jumped out at me last night, but I can't recall if it was 493 or 498 hands I did it from, so I just said "almost 500." For the purpose of describing this here, there is little difference between 490-something and 500.

I will not post the spreadsheet until I have many more than 1000 hands where one player has a marked advantage. 384 is the magic number in stats for analyzing one variable...so 1,500 hands will be more than enough to look at individual odds (like all 70/30 situations) and the big picture, with no room for dispute of validity. Until then, the numbers are still questionable and can be dismissed as being part of a small sample size. Additionally, the numbers may still even out over the next few hundred hands. and I might actually end up winning 60%, which would still be a bit low considering the number of 70/30 hands in the sample, but would be arguably straight. When I have a larger sample, I will post it or email it to anyone who wants it.

I did the odds analysis with the cardplayer.com odds calculator. Of course for the obvious situations where I knew the odds by memory I just filled them in. So yes, I am sure that I am not including 55/45 hands in the "hands with an advantage" category.


As for the poker site, I doubt they will even respond to an email accusing them of cheating. It is their policy to ignore such communications. However, it is my hope that bringing something like this to their attention will force them to look at their RnG....there may even be some people in their organization doing things against company policy that lead to this type of error...who knows. I will say this. I am currently working with a PhD in mathmatics at a major university who is writting a paper on online poker and some of the things he is finding are completely whacked.

Like I said...I still play, but not for the kind of money I play in the casino.