ICM- The basics

[Longy]

I'm probably going to do 3 posts, the next 2 will be covering doing ICM calcs by hand and using ICM based software.

What?

ICM = Indpendent Chip Model, a mathematical model which allows us to assess how much our chipstack is worth in relation to any given tournament payout structure.


Why?

In 99% of Poker Tournaments, you don't get all the prize pool for getting all of the chips. Therefore our chip stack is not linear with how much we have paid to enter the tournament. Therefore we need a way of assessing how much our chips are truly worth, unlike in cash games where each chip is worth that value.

A basic example being you get 30% of the prize pool in a standard sng for eventually losing all your chips.


Where can ICM be used?

In all poker tournaments it is applicable. Yet its main function is in 1 table situations (Sngs and MTT final tables) and when shortstacked (less than 10bbs). Our two options are to push or fold. Using ICM we can assess which of these two options is best by applying the mathematical model and virtually play perfectly because of it.

In the current online climate, being ICM aware in sngs is virtually essential to be a winning player at anything but the micro limits. As an awful lot of shorstacked situations occur where ICM is applicable. For MTT players it is far less essential but it can only help your game to know generic shortstack push/call ranges.


How?

ICM problems can be worked out with pen, paper and pokerstove (http://blondepoker.com/forum/index.php?topic=36870.msg807120#new) but it is multi stage mathematics and is quite time consuming when trying to work out whether pushing or folding is correct.

So ICM calculators like sngwiz (http://blondepoker.com/forum/index.php?topic=36984.0) take the maths out of it for us and by simply putting in relevant chip stacks, ranges and our cards we will get an answer back instantly, whether to push or fold.


So it is perfect right?

Err no. ICM doesn't take into account the effect of the blinds hitting us, for example it can be right to take a -ev shove utg. As our stack will take a big hit next hand. Also it doesn't figure in things like a short stack sitting out and when a player is extremely short the mathematics falls down due to not assessing how much stack is truly worth.

OK give us an Example?

A post from a while back, about folding qq on the bubble.

http://blondepoker.com/forum/index.php?topic=23792.msg484950#msg484950

This is a very counterintutive hand where icm is extremely helpful.


Err I don't know why but I'm not overly happy with this post, feel like i have missed something or not explained myself very well. All comments welcome.

[T_Mar]

Well as someone who knows very little about ICM, I'm looking forward to reading the next two posts... Thanks for taking time to write them!

As far as this post goes, I think you got lots of things you want to convey... and its prob not the easiest topic to explain!  Its a decent intro though and sets up the next posts well I think, n1

[Simon Galloway]

Would like to see thoughts on the accuracy of hand range assessment and its subsequent influence on push/fold decisions.  i.e. how wrong do you have to be, typically, on hand range before ICM returns dubious advice?
Also, on software tools, what freeware (or low cost software)  is best at setting up a 20 question ICM quiz?

[Longy]

Would like to see thoughts on the accuracy of hand range assessment and its subsequent influence on push/fold decisions.  i.e. how wrong do you have to be, typically, on hand range before ICM returns dubious advice?

Good question.

Position and stage of the tournament especially near the bubble can make a difference.

When looking to shove a hand utg 7-10 handed with 8bbs in mid tournament. Getting players calling ranges wrong is no biggie as one mistake tends to get evened out by another. Plus with the money so far way, each player chip stack is worth less of the prize pool relative to a bubble situation and losing/winning a hand means less.

Whereas calling a shove from the sb on the bubble from someone has you covered getting your ranges right makes a massive difference. Also different players shove such different ranges in these situations, so knowing what they are like is important.

Knowing inexploitable shoves is vital, meaning no matter what your opponents range is you can expect to make money in the long term.

Also, on software tools, what freeware (or low cost software)  is best at setting up a 20 question ICM quiz?

Sngwiz has a quiz function on it and has a 30 day free trial - www.sngwiz.com

There is also bubble trainer which is similar and allows you to do around 100 hands a month for free (I think) -http://pregopoker.com/

[Charlie44]

Thanks - very informative - look forward to your next post.

[bhoywonder]

good stuff longy

im making ICM my new hobby for the next wee while

not tonight though..

I have a date with Mr.CArling...

[Laxie]

I need to work on this too.  Thanks Longy.  Reading and doing my best to absorb.  Will ask questions as we go.

[Robert HM]

Looking forward to this thread's development. I need to some work on this area.

[tao82]

gd stuff. will be keepin an eye on this thread.

[vegaslover]

Nice work Longy, look forward to reading the finished product.
Whilst I have some understanding of,and use ICM, don't think I could clearly explain it to others.

[TheChipPrince]

Dont teach too many fish Longy, muchos respect though for taking your time and teaching it...