What are ICM calculations?

[RedGladiator]

some one mentioned ICM calculations on another thread. please excuse my ignorance, i have no idea what this means.

[Slim Pickens]

http://sharnett.bol.ucla.edu/ICM/ICM.html

Start there. If you still have questions, just post them and I can probably answer them.

[toms2866]

ICM calcs compute equitable "deals" for tourney final tables.

[KPL]

[ QUOTE ]
ICM calcs compute equitable "deals" for tourney final tables.

[/ QUOTE ]
Yes, but it also helps to calculate the EV of your play in tournaments.

[RedGladiator]

how important is it to learn this??
is there any book that explains in depth?
i play mostly STT NLHE.

[Prodigy54321]

[ QUOTE ]
how important is it to learn this??
is there any book that explains in depth?
i play mostly STT NLHE.

[/ QUOTE ]

go to the STTF and search for post with ICM and you should start to get the picture..

It is VERY important to learn if you are playing STTs

in most STTs you will be faced with situations that demand know about ICM in order to make the right decision...and making the wrong decision in many of these situations is enough to decrease your $Equity by a huge margin.

there's a link in the STTF FAQ on the topic of ICM..also there is a link in the FAQ..

SitNGo Power Tools

a program written by Eastbay that uses ICM

here's quick explaination of ICM and how it's used

ICM(Independent Chip Model) is an equity model for tournaments..It gives you the value of your chips stack in a given situation (considering # of players left, payout structure, etc)..this model assumes equal skill and is by no means perfect..but it is considered to be the best we have and serves our purpose with great accuracy.

this model illustrates the point that..in tournnaments, chips do not equal money...not only do they not equal money..but certain chips are worth more than other chips...so winning 4000 chips is not necessarily worth losing 4000 chips..

it shows that cEV(chip Expected value) and $EV(Expected vlaue in dollars(which is what we care about..not chips) are not equal in tournaments

here's an example that I used to show a few friends who strated playing STTs...

it's the BUBBLE(4 handed..top 3 get payed $20/$30/$50) in a standard 10 person $11 STT

Stacks..
Hero - BB - t8000
Villain - SB - t8000
Button - t2000
CO - t2000

it is folded to the villain..then he pushes allin..what do we call with??

we must first estimate our $Equity here(we can use ICM).. ICM calculator

Hero = 34.44%
Villain = 34.44%
Button = 15.56%
CO = 15.56%

since these are the percentage of the total prize pool that our chips stacks are worth here, we can give the $value of our stacks as well

Hero = $34.44
Villain = $34.44
Button = $15.56
CO = $15.56

alright now..assume we call the villain's push here...

if we lose - our $Equity drops to $0 since we are out with no money
if we win - our chips stack will double...if chips = money, we would need to win this hand >50% of the times to win in the long run...but chips do NOT equal money..here's an explaination

if we double up here ..we will have t16000...so in order for the t8000 that we won to be worth the same as the t8000 we already had in our stack, our $Equity would have to double from winning this hand..but it doesn't...our chips stack would have to have a value of $68.88..but even if we WIN this tournament, we will only win $50..a $Equity of >$50 is impossible..heres our real $equity after winning this hand.

Hero = $45.78
Villain = $0
Button = $27.11
CO = $27.11

*NOTE* notice how the two short stacks $Equity nearly doubled by simply letting the others fight it out.

so lets go over the outcomes again..

Starting $Equity(if we fold) = $34.44

if we lose - our $Equity drops to $0 since we are out with no money
if we win - our $Equity rises to $45.78

notice that if we lose, we lose $34.44 equity, and if we win, we win $11.34 equity

so the only way to make up for losing more than we win with this confrontation is to win it more often..it is simply not enough to win this hand >50% of the time..we must win this hand MUCH more often to make it right for us to call here..

we would need to win this hand 74% of the time or greater to make it correct to call here (with 300/600 blinds)....and the ONLY hand that will win that often against his likely range is AA...that means this is a situation were we should fold AK even if we KNOW that he is pushing ANY 2 cards here..even 32...and we will also fold KK here even if we know that he is pushing as low as 22, A2, and KJ.

that is how powerful the difference between cEV and $EV are.

this was thrown together a while back so there might be a couple things off (I think that the "74%" figure is a close estimate)..I think assuming blinds of 300/600...but it's close..also there are a few factors that would shift this slightly but the major idea remains that same

there are no blinds in this example because I wanted to illustrate simply how this works...when folding, you will have less than 34.44% equity..so you lose a little more by folding than my numbers show..and win a little more by calling than my numbers show..however my 74% figure is right for 300/600 blinds..which is the typical level for the bubble in low level stts

..now the typical player in these STTs who have no idea what ICM is will likely call here with as low as maybe ATo or 88...and in doing so they lose over $6 in $equity in a $11 tournament even if they are pushing crappy hands like K2 and Q6!!! which is huge..but since WE know about ICM..we will not make that same huge mistake

this turned out longer than expected [img]/images/graemlins/tongue.gif[/img]

[Mustafa]

That is a great explanation. Let me ask this since I haven't touched game theory and duality for 15 years. Does that make villian's move of pushing all-in (in that same situation) correct over 99% of the time since hero is incorrect to call unless hero has aces?

[Prodigy54321]

[ QUOTE ]
That is a great explanation. Let me ask this since I haven't touched game theory and duality for 15 years. Does that make villian's move of pushing all-in (in that same situation) correct over 99% of the time since hero is incorrect to call unless hero has aces?

[/ QUOTE ]

I am assuming that the OP is playing low buy-in STTs ($5-$33)

If villain knew that we were only calling AA and maybe KK here, he would certainly push any 2 cards and be correct in doing so..

however If I were in the villains situation and deciding whether or not to push allin, I would not even consider doing it with anything less than 77+, A9+, KQs...

there is simply too great a possibility that the BB has no clue as the what ICM is and has no clue as to just how tight he should be here...I wouldn't be surprised to be called here by A9 or 66 which would be devastating if we were pushing even a fairly tight range

Now just because WE lose when he calls loosely like this, doesn't mean that HE wins..he loses as well..just as if we were to call with a mediocre hand in the BB....the ONLY people who really gain from the two bigs stacks going against eachother are the two small stacks who suddenly have their $Equity nearly double.

what you pointed out creates an interesting Nash Equilibrium game theory type situation where it is mutually beneficial for to the BB and SB for the BB to play tightly..this would mean that the SB has an edge here if he could ensure that the BB knows what he is doing, but 95%+ of the time in low buy-in STTs, the BB will take an action that is not only damaging to his own $Equity, but yours as well.

[Mustafa]

That makes plenty of sense. If you are villian and you've decided BB is an average player, how do you determine your bet (assuming the above scenario) wrt the strength of your hand?

[Prodigy54321]

[ QUOTE ]
That makes plenty of sense. If you are villian and you've decided BB is an average player, how do you determine your bet (assuming the above scenario) wrt the strength of your hand?

[/ QUOTE ]

I think the average player is at least decent enough to see that he's not too excited to play a hand with you so you should still be fairly aggressive IMO..I think it's between.

raising with the intention of folding to a push...(depending on what you think of the opponent, you may be able to do this with any 2 cards)

or simply pushing allin..even with premium hands like AA and KK, you may just push allin considering that the BB simply calling your raise with the intention of only plyaing if he flops 2 pair or better would be pretty bad for you..you don't want the chance that this hand goes to flop to be very high.

these are just my quick thoughts on it..it's certainly not an expert opinion..if you are still intersted in knowing more about what to do in this type of situation, you can post a question about it in STTF if you like..you'll get the opinions of some of th best STT players out there.