[pzhon]

I really don't like the analysis in problem 12, with AJo after an UTG raise.

[img]/images/graemlins/spade.gif[/img] There is far too much apologizing about having a short stack. Getting crippled, or simply failing to accumulate chips as the blinds rise, is a normal part of tournament play. It's not a reasonable excuse to panic. Unfortunately, time and time again, when people ask a reasonable question about playing with a short stack in one of the tournament strategy forums, the glib response is that you should never have let yourself get short stacked. This is particularly wrong for a player with a skill advantage who can double up more than 50% of the time.

[img]/images/graemlins/spade.gif[/img] Reraising probably will clear out the players behind you. That doesn't make it right to assume that that is what will happen. The possibility that someone else will get involved is significant, and you need to consider whether this will help you or hurt you.

[img]/images/graemlins/spade.gif[/img] The conclusion was not justified mathematically. The excerpt seemed to suggest that if you are desperate, you should throw your chips in. While that might be the deciding factor here, a much better attempt should have been made to quantify it.

In this hand, you are in UTG+1, so folding means you get a free hand (or one where you pay only the ante of 25), and then the blinds probably hit you. Pushing as a 60-40 underdog means you give up that free hand 60% of the time, and the blinds hit you 40% of the time anyway.

How much does it cost to go through the blinds? Significantly less than the blinds themselves, since you are allowed to find a good hand while in the blinds, and after you make it through you are in late position where your stack is worth more than its face value.

The value of the free hand and the costs of going through the blinds may depend on the size of your stack, but let's assume that the dependence is negligible. I think a reasonable assessment of the net cost is about 100 chips; if anyone has some real data I'd love to see it. So, folding is worth about 1400-100=1300 chips, while pushing gives you 3300-100=3200 chips. That changes your threshold from 1400/3300 ~ 42.4% to 1300/3200 ~ 40.6%, which suggests that this is still a fold. This is a much smaller "desperation" effect than many players would assume, and it doesn't seem right to call this the dominant factor. For it to be right to push according to this calculation (which ignores the skill advantage which argues against pushing), your position must cost over 133 chips.


I'm afraid that this passage will encourage people to panic and play badly when short-stacked instead of making the most of what should be the most valuable chips.

Here is an example of a similar hand from a live tournament. I was short stacked in the small blind (due to playing well, btw), and considered pushing over a button steal raise with a weak hand. It seemed like a close decision numerically if I were sure that the big blind would fold, since the dead money was significant in comparison with my stack. However, even without skill advantage consideration, chips in a short stack on the button are worth a lot more than chips in a short stack in general, and there was no guarantee that the big blind would stay out. So, I folded. I was very happy after the BB reraised, and showed down AA. I bounced back to an above average stack.