ICM says this is a fold. I always thought this was a push...wtf

[Unoriginalname]

NL Texas Hold'em $20 Buy-in + $2 Entry Fee Trny:14334452 Level:5 Blinds(75/150) - Thursday, July 28, 15:55:54 EDT 2005
Table Table 13810 (Real Money)
Seat 4 is the button
Total number of players : 5
Seat 4: pkrplyerAA ( $1470 )
Seat 5: shannonsoss ( $1545 )
Seat 7: RAF77 ( $3290 )
Seat 8: Hero ( $880 )
Seat 10: lacvanle ( $815 )
Trny:14334452 Level:5
Blinds(75/150)
** Dealing down cards **
Dealt to Hero [ Ad 7c ]
>You have options at Table 13973 Table!.
Hero is all-In [880]

I don't get this at all. A7o is a pretty decent push hand. I don't think I'm going to get much better at this point. I'm short stacked and starting to get desperate and those 225 chips will increase my stack by about 25%. I post the BB next round, which will make me even shorter. Why is SNG power tools telling me that this is a bad push unless my opponents will only call with the top 8% of hands? That's pretty damn tight. Am I going crazy? Is this really an easy fold?

[1C5]

You have to push this.

[YourFoxyGrandma]

It's not your cards so much as your position that might make this a fold. Determining the correct play on this hand depends very much on having accurate calling ranges as you're pushing into alot of people. Unless your table was unusually tight, I'd definitely fold this.

[BadMongo]

I would insta-push this, and probably a few hands that are even worse. I'm not sure why ICM says this is a fold (and a pretty easy one at that). You are UTG, but if you don't pick up a hand in the blinds you will be left with virtually no folding equity.

This seems like a flaw in ICM analysis to me. Maybe someone with a better understanding of the model can explain this?

[KramerTM]

Wow. I'm confused myself. I've been pushing this (my min pushing hand here is like A5).

[Jbrochu]

How close are the ICM numbers...? A clear fold or borderline...? When you open up the calling ranges beyond the top 8% of hands, you bring in many of the Ax hands that have you dominated.

I would push this if I've been tight or not shown down junk to this point, and the blinds have been tight. If I've been pushing several recent hands, been forced to show down trash, or the blinds are loose, I think I fold this.

[spentrent]

[ QUOTE ]
It's not your cards so much as your position that might make this a fold. Determining the correct play on this hand depends very much on having accurate calling ranges as your pushing into alot of people. Unless your table was unusually tight, I'd definitely fold this.

[/ QUOTE ]

What is your calling range in the other seats?

[BadMongo]

I don't think MY calling range matters as much as what the typical player's calling range is. They are very different.

[BadMongo]

[ QUOTE ]
How close are the ICM numbers...? A clear fold or borderline...?

[/ QUOTE ]

I put the other players on what I thought were reasonable estimations of the other players calling ranges, given that this is a 20+2. Something along the lines of A7+, KJ+, 44+. I widened the range slightly for the big stacks.

Still a clear fold, not even borderline.

[maddog2030]

[ QUOTE ]
I don't get this at all. A7o is a pretty decent push hand. I don't think I'm going to get much better at this point. I'm short stacked and starting to get desperate and those 225 chips will increase my stack by about 25%. I post the BB next round, which will make me even shorter. Why is SNG power tools telling me that this is a bad push unless my opponents will only call with the top 8% of hands? That's pretty damn tight. Am I going crazy? Is this really an easy fold?

[/ QUOTE ]

This is because you're misunderstanding what ICM is and what it says. It's not a perfect model by any means. It's not taking into account you're going to be UTG with big blinds (and possibly even rising blinds). It's telling you its immediate approximation of your equity if you push or fold right now.

To the forum: it seems a lot of new, and some old, posters are finding ICM and just using it as a be all and end all crutch. If you don't understand how it works, why it approximates your equity in a certain way, and what it doesn't take into account in its calculations, you shouldn't be using it. Shadow's thread has plenty of links with discussions about ICM's workings and what its limitations are. If you haven't read them then don't bother using eastbay's tool.

[YourFoxyGrandma]

[ QUOTE ]
[ QUOTE ]
It's not your cards so much as your position that might make this a fold. Determining the correct play on this hand depends very much on having accurate calling ranges as your pushing into alot of people. Unless your table was unusually tight, I'd definitely fold this.

[/ QUOTE ]

What is your calling range in the other seats?

[/ QUOTE ]

If you assume a TT-AA AQs-AKs, AKo calling range for all seats this becomes a clear push.

If you assume a 66-AA, ATs-AKs, AJo-AKo calling range for all seats this becomes a very marginally +EV push

If you assume a 44-AA, A7s-AKs, A9o-AKo, KJs+ calling range this becomes a clear fold.

If you want to come up with a "standard play" for this hand, you can usually assume a mix of the last 2 calling ranges at the $22s, making this a very marginally +EV situation at best (and usually a clear fold).

[Pokerlogist]

IMHO, you and SNG analyer are both right! Using the top 8% criterion is probably correct here. The top three stacks are favorites for ITM and should want to play safe. Top 8% calling criterion isn't that tight anyway.
So push.

[BadMongo]

Uh, this is a 20+2. No way the other 4 players are only calling with the top 8% (66+, AT+).

[bigt439]

[ QUOTE ]
IMHO, you and SNG analyer are both right! Using the top 8% criterion is probably correct here. The top three stacks are favorites for ITM and should want to play safe. Top 8% calling criterion isn't that tight anyway.
So push.

[/ QUOTE ]

This is definitely incorrect at any level, but definitely at the 22's. I routinely fold this without a whole lot of thought. You have 4 hands to get through and your hand is not very good. Think about what kind of shape you're in when you get called, and with 4 people behind you, you'll get called more than you think. Also think about how much easier it will be to steal from the sb and button. I am very surprised at the responses thus far.

[West]

I don't know much about ICM yet, but I did some quick and dirty calculations that I think are helpful. You're not in good position in that you're fold equity is about to be significantly reduced, there are three stacks significantly higher than yours, and even the other guy has about the same as you.

The only hands you are really worried about calling you are AA-77 and AK-A8, because those are the only hands that have you dominated. By my calculations that range is a little more than 9% of the possible hands any one of your opponents could hold (AA&77 = 3 combos each (6), KK-88 = 6 combos each (36), AK-A8 = 12 combos each (72), total of 114 combos, 114 out of 1225 is about 9.3%).

1225 - 114 = 1111. I estimated the chances of none of the remaining 4 players calling you with one of these hands at (1111/1225)^4 ~ .6765

If we were to assume when called you would always be called by just one person, and if they only call with the hands you are worried about, then 67.65% of the time you would win the blinds (225), and 32.35% of the time you would get called. My quick estimation was that when called you would win 28% of the time. If on average you win 1050 chips when you get called but win, and of course lose 850 chips the 72% of the time you get called and lose, then your chip EV would be:

(.6765 x 225) + (.3235 x .72 x -880) + (.3235 x .28 x 1050) = 152 - 205 + 95 = + 42

I don't think being called by a few more hands that you are either beating or flipping against is going to change things much. Given that the blinds are about to take a good percentage of your stack on the next two hands I think it's a definite push.

*edit*...I think it's *probably* a push, though my instincts are this is right around a borderline decision. From personal experience, (not in the $20s), I know I wouldn't push this from UTG if there were 6 players instead of 5.

Also, things obviously improve for you if some of the hands that dominate you won't call you (obviously player dependent), which I would expect to be the case.

[Jbrochu]

[ QUOTE ]
To the forum: it seems a lot of new, and some old, posters are finding ICM and just using it as a be all and end all crutch. If you don't understand how it works, why it approximates your equity in a certain way, and what it doesn't take into account in its calculations, you shouldn't be using it.

[/ QUOTE ]

Excellent point, this made me take a second look at the problem. Realizing that hero takes the blinds next hand (my bad for not taking this into account), I would push this even if the criteria I listed in my earlier post (I've been all-in several times, or forced to show down any two, or the blinds are loose) were true.

[Moonsugar]

ICM has no way to value the big negative hit to equity you are about to take from the upcoming blinds. You should make -ICMEV plays when your stack is low relative to the blinds and you are about to be in blinds. How -ICMEV depends but in a situation like this I think the number is around -1% to -2%.

[jskinn04]

Maybe think about folding this hand if there are four people left, but with five people left you're probably going to have to get a little lucky to strike the money. Push.

[Matt R.]

2 quick questions: I haven't bought eastbay's program yet, and I'm wondering what the output was. How negative was the $EV? I'm planning on buying it soon, so this is just for my curiosity. My other question is, when you guys are saying "top 8%" of hands and such, is this the Karlson-Sklansky rankings? If not, could someone tell me what range we're referring to?

And I think this is a push despite eastbay's tool (wow, that sounds a little weird doesn't it? I mean eastbay's program, but we'll leave it for humor value). As others have said ICM doesn't take into account relative position of stacks and blinds (therefore, I don't think eastbay's tool does (I think?)). Since you're in the blinds next hand I think this is definitely a push.

[nWirb]

I havn't read all responses yet, but this is a push because even though it's -$EV, not pushing here is even more -$EV.
What I am saying is sometimes you have to take -$EV situations because the alternative you have is even more
-$EV.

[The Don]

It's defintely a push... I don't think ICM takes into account that you are being hit with the blinds next hand.

[Nick B.]

I wouldn't push A7o utg 5 handed, there is too great a chance that you will be called and only have 3 outs. I would push KQ though.

[eastbay]

There's an important thing to consider for SnGPT analysis UTG, and the current version of SnGPT does not do it automatically, although it's something that's going in very soon, to be more user-friendly for newer or less familiar users for hands exactly like this.

Here's the deal: Your "$EV fold" which provides the baseline for comparison against your push is computed from your stack after you fold this hand, but before you post any blinds in any future hands.

If you are going to be taking a big blind hit in the next hand, you really should be discounting your fold equity.

This is very easy to do. Just take the big blind out of your stack NOW and recalculate. You should consider your fold equity to be somewhere between the nominal value and the discounted value. In a very passive game, where you are likely to get a pass in your BB, you should consider it to be closer to the pre-blind value. In a more aggressive game, were you are not likely to get a pass, you should consider it to be much closer to the discounted value. Exactly where to choose is a matter of judgement based on the game conditions, but it is probably reasonable to just pick the discounted value as a conservative default.

Now compare THIS new "$EV fold" to your original "$EV push". I think this answer will be less surprising, although you are still subject to horrid position here, so the true range may be far tighter than you realized even with the discounting adjustment.

Look for an update soon that automatically takes care of this position based equity discounting.

eastbay

[derdo]

I wouldn't push this for 2 reasons.
1) big stack is the big blind. he'll call a good % of the time.
2) Big stack is to your right and You have two medium stack to your left. Even if you eat the blinds you have some folding equity against them. I think, it is safer to push with two worse cards when big stack folds.

Cem