ITM situation: ICM does not apply here

[Slim Pickens]

Party Poker No-Limit Hold'em Tourney, Big Blind is t500 (3 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

Button (t6640)
SB (t800)
Hero (t560)

Preflop: Hero is BB with 7/images/graemlins/club.gif, 3/images/graemlins/spade.gif.
<font color="#CC3333">Button raises to t1000</font>, SB calls all-in [t800], Hero ????

[Moonsugar]

You should call. Trying to luck into a shot at winning 30% of pool, not luck into an extra 10%.

[45suited]

I would actually fold here. Even if you get extremely lucky and win the hand, you're still far behind 1st anyway. On the 11s and 22s where I play, SB could easily be calling with a "monster" hand such as AQ and lose the pot a good 40% of the time. Fold and hope for 2nd here.

[Nottom]

I mean you can fold and give yourself maybe a 45% chance of taking 2nd and basically a 0% chance at first. In a $10 SNG the EV of folding is about $4.5.

The other option is to call. If you win, you have about 1700 chips and have a decent shot at 1st. I don't think either players range here should be particularly tight (Pairs/Aces/Broadways/+ a few more) so I'll guess you win somewhere around 20% of the time. When you win your stack is worth around $33 on average. So $13/5 = $2.6.

So asuming my math is even remotely close, this looks like an easy fold and I need to go back and change my vote. As for the title of your post, I do think ICM still applies here, but mostly becasue the math is so easy with the chipstacks in question (lots of 0 and near 0 stacks).

[curtains]

my instincts are to fold.

[Slim Pickens]

As for the title, I'm no ICM expert so ICM may in fact apply here. My understanding was that ICM was great only with one short stack, where survival for additional hands is only contigent upon whether or not you, as the shorty, bust. With two short stacks, it doesn't account for your additional share of the prize pool if the other short stack has to go all-in on the blinds soon, or is already all-in.

[Nottom]

You are misunderstanding what ICM actually is. ICM is simply a model that gives the expected $ value of a stack of tourney chips based on the size of all remaining stacks. Heres a link an ICM calculator (http://www.bol.ucla.edu/~sharnett/ICM/ICM.html) if you don't already have it.

A lot of people seem to confuse ICM with all the stuff Eastbay's SNG:PT program does just becasue it utilizes ICM calculations to compute its results.

When you use ICM calcs to figure out whether a call is +EV or not you are basically just averaging the resulting ICM values based on the chipstacks after you win or lose and then comparing that to what it would be if you folded. If you look back at my first post, I'm basically just doing some ICM-style calculations but they are really simple enough that you don't really need to calculate the exact ICM number to understand the result.

[microbet]

I'm on 4 tables now, so I'll probably say something stupid, but...

This is interesting. Big stack probably has a pretty wide range here. If you think SB only calls with a monster, then if you fold, he's probably like 70% to win, which makes you about 30% to come in 2nd. If you call, you are probably still like 20% to win the hand, so not that big a diff and probably worth the still small shot at first.

If SB doesn't need a monster, he is probably only like 60% to win, making you 40% for 2nd and if you call, you are still probably only about 20% to win.

So, my guess is, the tighter SB's range is, the more you should call.

[mmbt0ne]

Here are my thoughts, random and disjointed.

If you fold, you move up in money only when the hand rank goes:
You -&gt; Button -&gt; SB

Also, anytime button or SB gets first, you end up in 3rd place, which is where you would probably finish anyway if SB wins and you fold.

One more thing. Let's say it goes:
You -&gt; SB -&gt; Button.

Next hand is:
Button (480)
Hero (1680)
BB (5840)

I don't know if this is right, but I'm folding my SB here unless I have something like TT+, or AQ+. So, assuming Button didn't go all in here, the next hand is:

Hero(1430)
SB (6090)
BB (480)

So you're still in a position where the big stack is dominatng both of you, and if BB gets lucky, you're both going to be even stacked again duking it out.

Basically, I'm trying to say there's almost no way you get 1st here, and folding gives you a much better chance at finishing 2nd.

[adanthar]

turbo super ultra fold.

[eastbay]

[ QUOTE ]
You are misunderstanding what ICM actually is. ICM is simply a model that gives the expected $ value of a stack of tourney chips based on the size of all remaining stacks.

[/ QUOTE ]

I'm pretty sure Slim understands all that. The point here is that sometimes it's more than just the stack sizes that are important for determining equity. When in a 3-way game, 2 stacks who are so low that they are nearly or completely all-in when they take the blinds, the stacks aren't the whole story anymore for getting an accurate picture of equity. Who's going to take the blind next can be very important. ICM doesn't "know" about this important positional effect - it only "knows" about stack sizes.

eastbay

[lastchance]

Fold Equity = .3 (if SB loses) or .2045 (if SB winz)
Call Equity if won = if SB loses .339 (if SB loses) or .31 (if SB winz)
Call Equity if lost = .2

Meh, for SB winz, you need to call any 2...
For SB loses, you fold any 2...

Assume you win 25% of the time.
SB = chance of SB winning

.25*.31*SB + .75*.2 + .25*.339*(1-SB) = .3*(1-SB) + .2045*SB

Let's take a crack at this.

Call Equity = FOld Equity
.0775*SB + .15 + .08475*(1-SB) = .3*(1-SB) + .2045*SB
.0775*SB - .2045*SB + .15 = .3*(1-SB) - .08475*(1-SB)
SB * (.0775-.2045) + .15 = (1-SB)*.3-.08475
SB * -.127 + .15 = (1-SB)*.215

Replacing SB with x, SB is too confusing.
-.127x+.15 = .215-.215x
.088x=.0665
x=75% chance for SB to win so you can call = autofold.