1-22-06 [poker]Another thing I'm working on is stack size theory in multis. One thing I've come to grasp is that in crucial pots, you need to think about how it affects your stack category. There's no need to take a big risk in order to just slightly change your stack size. Also, say you have a medium stack, you don't really want to tangle with a shorter stack if losing would make you a short stack, but winning wouldn't push you up much, you'd rather double up.
For example, say you have 2000 chips, and are around M = 10. Average is 3000 chips, so you're a bit low. If you double up, you get to 4000 and can be in control. Should you tangle with a guy with 1000 chips? No. If you lose to him, you go down to 1000 and are now a desperate shorty in big trouble. If you win, you only go to 3000 and are just average. That's not a good gamble. Much better to take the risk when you have a chance to double up.
Similar ideas apply when you're a big stack. As a big stack you can be perfectly happy to take any gamble that doesn't risk putting you below average. So if average is 3000 and you have 6000, take any gamble of 2500 chips or less.
Once the pot gets big compared to my stack, I want it bad, even if the odds are not ideal. Say the pot is 1k, and I have 1k left, so if we get all-in and I win we'll got to 3k. If I fold I go to 1k. Presumably that's near an inflection point, in which case I definitely want to gamble for that pot. The only way I wouldn't would be if average was way over 3k and it was near the bubble, or if average was way below 1k, so I could fold and still have a big stack.
An example of this is when the board comes with a 4-flush or 4-straight. If the pot is around 1k and you have 1k in your stack, push all in. Push whether you have the hand or not.