1. 2008-01-07 08:11:25

   Hand Analysis for 01/04 Tournament

   I'm writing this post mostly to reflect on the hand that crippled me in Friday's tournament at the group I frequently play with. I had managed to make it to the final 12, but I probably should have been home already. The blinds were getting really high and we were playing with antes after the first hour or so of play, which means each hand takes almost a minute longer - which means we see about 7 or 8 hands each blind round. Anyway, I went all in with Q8s and was called by QQ, but I made a straight. That same opponent had suffered some pretty bad beats this tournament. :/

   So, the blinds are 600/1200 and each of the 6 players has a 200 ante. That puts 3000 in the pot before the hand starts. The average chip stack is around 17k and I have about 25k, but I'm in the big blind. My opponent raises to 4.8k from middle position and everybody folds to me - I look down and see AQd. At the time, it seemed like an overbet, but looking at the numbers now, I don't really feel the same way. Anyway, I decided to put my opponent to the test, since he had about 24k before the start of the hand and I figured he wouldn't want to commit with a hand that at the time seemed to show some weakness. My opponent thought for a minute or so, and called with 99. I lost the race, and went out in my small blind the next hand.

   In retrospect, I obviously wish I would have played the hand differently. I think it was coincidence that I was mostly correct on my "read" of my opponent's hand, too, which is a little disheartening. Just for fun, though, I thought I'd do a little mathematical analysis on whether or not my play had a positive expectation, and/or how some details may change that expectation.

   First, the hand groups I thought my opponent might have and some percentages that he might have had them. These are just rough estimates that probably unsuccessfully try to reflect my opponents playing style, weighted a bit towards the probabilities to receiving those hands, and then subjectively weighted more yet towards the hands that make my decision look better ;)

   • AA: 5%
   • AK: 15%
   • KK, QQ: 20%
   • Dominated ace (AJ, AT): 10%
   • Complete bluff with suite undercards: 10%
   • Smaller pair (most likely JJ-99, but possibly others): 40%

   My approximate win percentages against those groups:

   Group	Win %
   AA	12
   AK	28
   KK, QQ	33
   AJ-AT	72
   Undercards	64
   Underpair	46

   Multiply the percentage chance that he holds each hand by the associated win percentage to get my aggregated expectation in this situation:

   5% * 12% + 15% * 28% + 20% * 33% + 10% * 72% + 10% * 64% + 40% * 46% = 43.4%

   Even if the expectation was positive, I'm not sure if I would make the same play again. Unfortunately, the fact that the average chip stack is a very small multiple of the blinds and antes makes the game seem a lot more like playing a slot machine. Had I hit that 50/50, I would have had enough chips to play (and it turns out my opponent won the tournament with those chips). I guess it's better than coming down to a coin flip an hour later and ending up on the bubble.

   In other news, I lost half a buy-in playing Full Tilt tonight :/

   Posted by Jason M at 2008-01-07 08:11:25

   View all posts

Comments on “Hand Analysis for 01/04 Tournament”

1. • Posted by Jason M g 2008-01-11 18:11:50
   • 

   • Thanks, dog. I keep meaning to check that out, but I should really just do it.

2. • Posted by uzjedi (unverified) g 2008-01-11 09:00:06
   • 

   • Did I tell you about "pokerstove"? It's a great poker calculator that lets you input hand ranges instead of just individual hands. You click card buttons (from 52 obv) for all the hands you think your opponent might hold, and it then tells you your hand's equity vs his range. Very useful. It would take me too long to do all that math on my own. ;)