Probable?

[Gubus]

This may belong in a bad beat forum, but I'm looking for a stat/game theory whiz (Andy Block type) to give me a little insight. Not about a hand, but rather a very long stretch of what appears to fall outside expected variance (at least to my somewhat limited knowledge of the subject).
I've been keeping track of my all-ins (NL Holdem) for a long time. I've got a lot of different catagories/situations, but will just post a common situation all can relate to. Pre-flop, all-in (me or the other player all-in): under pair vs 2 over cards. The classic ("thank you Mike Sexton") race situation. We all know the pair is usually favored by a few percentage points, but basically even money. As of this writing, when I'm holding the pair, my record is 61 wins, 252 losses or 19.5% (61 divided by 313 total situations). On the other hand, when I'm holding the over cards, I've won 28 out of 199 (14%). That's a combined 17.3% in a 50-50 spot. Maybe 512 times in that situation isn't a large enough sample, but it's going to take me a LONG time to get even.
I've got other much more improbable records, but believe me, you wouldn't believe me!
Gubus

[pzhon]

[ QUOTE ]
Maybe 512 times in that situation isn't a large enough sample,

[/ QUOTE ]
That sample is huge.

[ QUOTE ]

I've got other much more improbable records, but believe me, you wouldn't believe me!

[/ QUOTE ]
I still don't believe you.

The odds against that type of result happening with fair cards are astronomical. I'm not even going to bother calculating it. The odds against you having a recording bias which would lead to such a report are not astronomical. I have checked over the records of a few people who claimed unusual streaks like this, and every single time I have found a huge recording bias, though not usually one so severe.

I suspect that when you get knocked out of a tournament, you are much more likely to record the hand than when you survive (win or lose) and have to play the next hand. When you lose, you definitely see what beat you, but when you win, it may take extra work to see whether you were in the pair vs. overcards situation or not.

[qpw]

Recording bias is what gives rise (and continued credence) to many of the more improbable old wives tales.

Just thought I'd mention that - some examples of the phenomenon are hilarious.

[jogsxyz]

[ QUOTE ]
[ QUOTE ]
Maybe 512 times in that situation isn't a large enough sample,

[/ QUOTE ]
That sample is huge.


I suspect that when you get knocked out of a tournament, you are much more likely to record the hand than when you survive (win or lose) and have to play the next hand. When you lose, you definitely see what beat you, but when you win, it may take extra work to see whether you were in the pair vs. overcards situation or not.

[/ QUOTE ]

I agree with this poster.

If someone else recorded your play, the results may look different. Less biased. More like expected outcomes.

[Gubus]

Update-
OK, I took the comments to heart. Maybe I was hideously off with my recoed keeping. Started tracking anew, with a different approach. This time strictly as a statisitical endeavor, with the sole purpose of keeping the records... not playing & keeping track on the side. Played at UB and Absolute, 50 cent Sit'N'Gos. Guaranteed accurate.
All in pre-flop, under pair vs overcards - 23 wins, 66 loses= 26%. Over cards vs underpair - 10 wins, 44 loses= 19%. Not as whacked, but still weird. Side note...not sure if the following can be assigned a probability - flop a set, all in before the turn, 1 win 5 loses. Same setup, but flop a straight, 0 wins 4 losses.
There...don't I feel better now...

[Lego05]

Gubus....get Poker Tracker.


Then over a good enough sample size people will start to take it seriously......maybe evn if you keep pushing an it's not fair agenda (tough prolly not) But if you're looking for someone to tell you it's rigged and the results are not your fault it ain't gonna happen.

First off...were all those hands all-in pre-flop? Or not? Huge diff.


Either way if you ever plan on taking poker seriously get poker tracker...it can only help if you utilize it properly. It can't hurt, but it can be usesless if you don't use it correctly (actually I guess it could hurt if you use it really wrong.) Fortunately Pokey made a ridiculously good post on how to properly utilize poker tracker to analyze your own game. (Somebody else might point it out for you but w/e....it shouldn't be too hard to find if you wanna find it. I'm not quite sure where it is.)

[holdem2000]

The outcomes described in your original post (assuming each hand was actually exactly 50-50) most likely wouldn't happen once if you ran a sample of 512 coin flips billions of times each second for the entire age of the universe. Like everyone else said, recording bias.

In your new sample (again with the same assumption), the probability of losing at least that many flips out of 110 hands is approximately 1 in 10 billion. The chances are astronomically higher (but nowhere close as bad as the first set of data) that you either again are misrecording despite your efforts, or that you're misunderstanding something you're saying.

[ev_slave]

Just a final question: you said that you recorded YOUR all-ins, but what about when you have someone covered and call? Were these recorded?

While it would be strange that a subset of the whole (PP v Overs and vice versa ONLY while all-in) would have different numbers from the complement set (same while NOT all-in), it could be that overall you're actually even, but haven't recorded these situations.

Also, yes, do get PT.

[jay_shark]

Gubus , you are referring to a situation called conditional probability . Your stats are only recorded when your hands are all in but it fails to recognize the times when you've won a pot with your underpairs to overcards without going to showdown .

[mr_hanky]

These stats could be true.
Are ALL those all-ins heads up? Are you certain you are not including any hands where a super short stack was all-in with you making three players??