Finding mistakes using math, by Steve Paul-Ambrose
The hand: Day two of the WSOP $1,500 NL and there are 270 left. We're in the money, average is around 45,000 and I have 11,100 after losing a flip on the bubble the night before.
Dustin Dirksen opens under the gun to 3,000 at 600-1,200 (100), I'm the cut-off with A-10 offsuit. I shove. Now against a tight under-the-gun raiser this is an easy fold even with nine big blinds. But I felt like Dustin would raise a pretty wide range here so I put it in.
Results: He calls with J-J and I bust. Now it's easy to say one of two things: 1) He had J-J, bad shove, 2) He's probably raising a lot of hands worse than A-T, good shove. But there's more to it than that.
The math: I'm putting in 11,100 to win 13,800 (11,100 from Dustin and 2,700 from blinds and antes) so I need to win almost 45% of the time ignoring the times someone behind me has a hand. I think we can safely assume Dustin is never folding pre-flop since he'll be getting 2:1 to call my shove.
Using Pokerstove (a really handy program which is free to download) I get that he has to be raising at least any pair, most suited aces, A-T and up, suited broadways and some smaller suited connectors just for my shove to break even. And while it's possible he's opening that wide (or a little wider), I haven't even dealt with the three players still to act.
Conclusion: Probably a small mistake to move in there, especially since I think I play a short stack better than most and there's some small value to surviving, despite my short stack.
Most importantly, though, doing math like this away from the table will give you a much better understanding of short-stacked play and make you a better overall tournament player.