ICM analysis of A,Q Facing (Relatively) Early Push

[Jbrochu]

PokerStars No-Limit Hold'em Tourney, Big Blind is t150 (5 handed) converter

saw flop|<font color="#C00000">saw showdown</font>

Hero (t3480)
BB (t1542)
UTG (t3140)
MP (t2780)
Button (t2558)

Preflop: Hero is SB with Q[img]/images/graemlins/diamond.gif[/img], A[img]/images/graemlins/club.gif[/img].
<font color="#666666">2 folds</font>, <font color="#CC3333">Button raises to t2558</font>, <font color="#666666">2 folds</font>.

Final Pot: t2783


I laid it down to the push, but felt like I played it weak so I decided to try an ICM analysis for the first time.



Situation

Blinds were 75/150 with about one minute to go until they went up to 100/200. The button still had a decent stack, as did everybody else, so it was nearing the time for the aggression level to ratchet up a bit. Up to this point, the button has played fairly solid ABC poker as far as I could tell.



ICM Analysis

The ICM calculator provided $EV based on the current chip stacks and results of the hand as follows:

Fold: 0.24
Call Win: 0.36
Call Lose: 0.09
Split: 0.25

I gave the button four possible ranges of cards and the likelihood he was playing that range, as follows:

1) AA through QQ,AK at a likelihood of 20%
2) JJ through 99,AQ,AJ,A10 at a likelihood of 40%
3) 88,77,A9 through A4,KQ,KJ at a likelihood of 30%
4) 66 through 22,K10,QJ,Q10,JT at a likelihood of 10%

Based on the probability of winning, losing, or splitting against each of those categories of hands, adjusted for the probability of the button pushing those hands, I found the following $EV for A,Qo in this situation:

Fold: 24.0
Call: 23.47

If I did a reasonable job estimating the likelihood of the button pushing the given ranges, then folding A,Q is the best choice.

Do you believe based on the information provided that I properly weighted the likelihood of the button pushing each range of hands, or would you adjust the likelihood differently?

If this is a reasonable estimation, then I think I need to tighten up my normal calling ranges a bit.

[Nottom]

I'm pretty sure if you always folded AQ to a 17BB all-in and weren't heads-up that wouldn't be far from ideal.

[microbet]

I think it is a pretty easy fold and I think you numbers make it look closer than it is.

I think you are much more likely to be facing 66 than A4.

[Jbrochu]

If I change category 3 hands to 88 through 55, A9, A8, and KQ and assume 0 percent chance of the button pushing a category 4 hand, I get the following results:

Calling is $EV 23.11
Folding is $EV 24


For those familiar with ICM methods, is this actually close? I have no idea if an $EV difference of .9 is significant.

[Nottom]

A difference of almost $1 in a $10 toureny is pretty significant. That's about a 9% difference in ROI.

[Bill Poker]

i would probably fold AK too.