What is ICM? (Day 14 Course Discussion)

[Debi]

What is ICM is a question I wanted an answer to when I first started getting serious about playing tournaments. It stands for "Independent Chip Model" - but what is it?

If you have not yet read Day 14 and watched the video for Day 14 - take a few minutes now to do that and then come back here to discuss it:
https://www.cardschat.com/become-a-winning-poker-player/day-14/
What is ICM?

ICM converts chips to dollar equities so you can make money winning decisions. Collins goes thought this process in an easy to understand way - if there is anything you don't understand about ICM feel free to ask him or Katie in this thread.

 

[Polytarp]

ICM software was not allowed during game play when they first came out because the algorithms essentially made any dope a winning player. Back when Titan Poker had their jackpot SNG's, my way of playing was to wait for the another player to make a first mistake or somehow contrive a situation to make this happen and then patiently cut them to ribbons.
Understanding what to do and how to do it is contingent upon knowing exactly when to do it.
For example, understanding when a player is on tilt or when a player has been "scared straight"...the opposite of tilt...etc. The ICM model initializes with all players existing at the same skill level which is never the case. Only when the skill levels are weighted appropriately can certain plays be made...GIGO (garbage in garbage out is expensive). Have you used SNG Wizard and if so, any opinions?

 

[cheeeer]

I remember there was another way to convert chips into money yet. Is it not used anymore? And I'm not sure how hands influence on ICM. I thought ICM is only about a deal?

Anyway, I asked to deal 2 times and noone wanted....

 

[PsychoVas]

I guess ICM theory is what we instinctively do when ITM and about to reach a pay-jump. If other stacks are short and more likely to bust we avoid confrontation with anything but premiums. I'm still familiarising myself with ICM and I would like to ask you as a fellow mathematician if there are any free accessible theory analyses about that and "Nash Equilibrium".
Keep up the great work!

 

[Collin Moshman]

ICM software was not allowed during game play when they first came out because the algorithms essentially made any dope a winning player. Back when Titan Poker had their jackpot SNG's, my way of playing was to wait for the another player to make a first mistake or somehow contrive a situation to make this happen and then patiently cut them to ribbons.
Understanding what to do and how to do it is contingent upon knowing exactly when to do it.
For example, understanding when a player is on tilt or when a player has been "scared straight"...the opposite of tilt...etc. The ICM model initializes with all players existing at the same skill level which is never the case. Only when the skill levels are weighted appropriately can certain plays be made...GIGO (garbage in garbage out is expensive). Have you used SNG Wizard and if so, any opinions?

Good question: SNG Wizard is good software but not as useful as more modern alternatives like ICMizer.

 

[Collin Moshman]

I remember there was another way to convert chips into money yet. Is it not used anymore? And I'm not sure how hands influence on ICM. I thought ICM is only about a deal?

Anyway, I asked to deal 2 times and noone wanted....

I don't know of another way besides ICM. And no, that's great you asked because it's a common idea that ICM is just for deals but it's a lot more than that Any time we're trying to make an equity-maximizing decision, like whether calling an all-in at a final table is profitable in dollar terms, we need ICM for that.

 

[Polytarp]

...and "Nash Equilibrium".

https://library.princeton.edu/specia...Games_Nash.pdf

 

[Vallet]

ICMizer programs and calculator will greatly simplify your life. But without such programs, the head is completely filled with math during tournament hands. I don't use programs while playing games. But it was interesting to read nevertheless. Thanks!

 

[cheeeer]

I don't know of another way besides ICM.

Oh, chip chop, remembered. Though rarely fair.

 

[cferdi]

Very interesting, however, I don't quite understand how this helps. It's not really something one can use at the tables, surely?

In some ways, though, I think I (probably most half-decent tourney players) do some of this stuff kind of instinctively? For example, say you have ATs and would normally raise or even reraise depending on position, player types, etc. Now say two, three or more other players look like they want to battle it out with allins (or seem prepared to) preflop; in this situation, unless I have them well and truly covered, I will quite often fold and let them, knowing that one or more players will leaave the tournament and I risk nothing. Isn't that, kind of, like ICM?

I've never really understood it. I've tried, but don't see how it helps beyond working out prize money - maybe I'm just dumb?

 

[Luvart]

Finished Day #14.

Good reminder that ICM doesn't apply to winner-takes-all tournaments. Definitely not a simple tournament concept, and it needs a a decent amount of hours of study for someone to become familiar with this.

Answers to today's quiz:

i)1.13/100 x 250$ = 0.013 x 250$ = 2.825$
ii)They will increased.
iii)Failed to answer this question. :hmmmm2:

Finished Week #2 of the Training Course.

A break the weekend and back on Monday with the first day of Week #3.

:date:

 

[Collin Moshman]

Enjoy the weekend off Luvart!

Cferdi, ICM is useful for figuring out shove/fold/call ranges at all stages of the game where stacks are shallow enough. But I agree that there's an instinctual component. You can learn more with software, but there are definitely old-school players who have a good intuitive grasp of these concepts. So that's a good point too.

 

[freddydr87]

This was a hard leasson,in cash we play with money,there are no chips or pay jumps u get excactly wath you win, but in Sitn Gos and MTT is diferent because the pay jumps gets very big when we aproach the top 3 players

 

[redboy23]

This chapter reminded me of a bubble play where I was about the 6th largest stack in the tourney and decided to play my AK hand to a rainbow flop of 2 8 K and busted out in tenth when villain tabled 88 . A lesson hard learnt. The player with 500 chips was very grateful I imagine!
Response to video question:

The player with 100 in chips has a dollar value of $2.83 (1.13/100 x $250)

If the player with 4500 busts out then the equity is as follows:

100 chips - 20.4 %
4500 chips - 36.5 %
8900 chips - 43.1 %

Another good reminder of the importance of discipline from this chapter. It was fun using those online tools as well.

 

[Phoenix Wright]

Almost everything in this one was new for me. Prior to this course, I knew that "ICM" stood for "Independent Chip Model" and that it had something to do with tournaments, but that was it. Now I understand it conceptually a bit better. As someone who has only played small home games so far, this was all new content for me.

Thanks for sharing; glad to learn more

 

[karmakoumas]

i don't like this video.
it's speak about common stuff and not very useful information while playing poker...

 

[Debi]

Almost everything in this one was new for me. Prior to this course, I knew that "ICM" stood for "Independent Chip Model" and that it had something to do with tournaments, but that was it. Now I understand it conceptually a bit better. As someone who has only played small home games so far, this was all new content for me.

Thanks for sharing; glad to learn more

Nice to hear you learned something new!

i don't like this video.
it's speak about common stuff and not very useful information while playing poker...

Do you not play tournaments? Cause if you do it is very useful.

 

[karmakoumas]

Do you not play tournaments? Cause if you do it is very useful. [/QUOTE]



yes i play, but us i said this video speaks about common things perhaps usefull information but it's not usefull to improve our poker skills

 

[Collin Moshman]

Karma, for many players this info will be new, but if you already know the content in a given Day's section then of course feel free to skip it over

 

[Nafor]

What should be our main take away from this lesson?
Let's say that a ICM-software gives us the optimal shoving range. Does that make our opponent's range and what we think it might be less important?

One can often hear on Twitch a higher stakes player thinking out loud if this or that spot was a good one ICM-wise. Would it be useful to learn some spots by heart or is there just too much room for mistakes?

 

[Collin Moshman]

What should be our main take away from this lesson?
Let's say that a ICM-software gives us the optimal shoving range. Does that make our opponent's range and what we think it might be less important?

One can often hear on Twitch a higher stakes player thinking out loud if this or that spot was a good one ICM-wise. Would it be useful to learn some spots by heart or is there just too much room for mistakes?

We should take the software-generated range unless there's a good reason to do something else, such as us shoving wider when we know our opponent calls much too tight.

Learning some very common spots by heart is a great idea, like shoving first in with 10bb in the small blind and button in an MTT.

 

[ronn6583]

The topic is interesting and not so simple. You can make the right decisions out of general considerations: for example, you need to call a shove in pre-flop or it is more profitable to wait until several players leave the tournament.
Chips get the value in dollar terms only if you get into the prize area.

 

[Tux97]

Thanks for this informative chapter and video. I just want to add to people new to ICM software like me that you should only use it while studying poker not whilst playing a tournament. I just discovered that most sites can actually suspend your account if you use ICM software during a game. Of course, it is very useful and educational before and after tourneys and thanks again for this easy to understand introduction [emoji106]

 

[johnnylawford]

I have also heard of this in terms of 'risk premium' when it comes to calling/making all-ins. Specifically how much "money" do you risk giving up if you bust with your current stack. For instance, in the book's example of calling a 10bb small blind shove w/A2 if you're in 5/10 place with 11bb you're essentially risking a high percent chance of laddering, so you'd need even more equity to call. It's probably beyond the scope of this book, but it's an interesting rabbit hole.

 

[Collin Moshman]

I have also heard of this in terms of 'risk premium' when it comes to calling/making all-ins. Specifically how much "money" do you risk giving up if you bust with your current stack. For instance, in the book's example of calling a 10bb small blind shove w/A2 if you're in 5/10 place with 11bb you're essentially risking a high percent chance of laddering, so you'd need even more equity to call. It's probably beyond the scope of this book, but it's an interesting rabbit hole.

Great point, there are different ways of looking at this including risk premium and bubble factor. ICMizer actually has a cool feature where it will display your bubble factor when you're doing calculations.

This would be a nice topic for future content