When they aren't dominated? Can you be good in tournaments without studying Icm? Any good poker pro comes to mind? Does Harrington or Doyle know ICM?

if you want to play mtt learn icm.

I'm no expert, but it's basically about calculating your stacks equity in a tournamount. Applicable in bubble situations and making decisions based around pay jumps generally.

But isn't ICM more useful for sngs? Or is out for MTT?

Where can I learn ICM?

Surely it applies to any format with a bubble adn tiered payout structure.
SnG players are generally a lot stronger in it as the volume of bubble and ITM finishes is so much higher, therefore it's a much more important part of their game.

read the stickies here:
http://forumserver.twoplustwo.com/36/stt-strategy/

they will contain links to further reading

also there is this:
http://http://forumserver.twoplustwo...turbos-541962/

ICM is applicable to every tournament that isn't winner take all. It is a method of converting your chip stack into a $ value. From there you use that number to calculate how you can increase or decrease that $ value. It is necessary because risk/reward of winning chips is not tied directly to the chip value.

e.g On the first hand of a 9 man sit and go two people go all in and one gets knocked out. The person who wins the hand doubles his chip stack but not his prize pool equity. This is because everyone else at the table receives equity from a player being knocked out. Everyone else at the table whose chip stack remains exactly the same gain equity as they are one step closer to the money.

It applies in the first hand of an MTT also but the difference is too negligible to calculate.

And to answer the question in the title ICM applies to every chip you win or lose vs an opponent. You don't have to be all in. Your equity changes by winning or losing chips and the person you win them from or lose them to also affects your equity.

Ok... Why do we have to know that?

When you make decisions in mtts and sngs, you don't want to maximize the amount of chips you win through a play. You want to maximize the amount of money you win through a play. $EV vs cEV. Making correct ICM decisions will maximize our $EV. Playing typical pot odds and equity will only maximize our cEV. These are not the same.

Yeah it's beginning to make sense. I guess my interpretation is totally off, huh.
I'm wondering if you know any good sources to learn ICM from scratch? I tried to learn it multiple times but have to give up because i don't understand s word that article says. isn't there, like a simpler non scary way to learn ICM? I'm not really mathematically inclined.

From an intuitive standpoint, try and think about it like this:

Imagine you could sell your seat to your twin brother at any point in the tournament. At the start of the tournament you'd sell it to him at exactly the buy-in, whereas when you have 99.999999% of the chips you'd sell it to him at whatever first place money pays.

Now imagine you're one-off the money on a large tournament that pays the top 10% (lets say the WSOP ME), and your chip stack is average. Your buy in is now worth approximately 10x what you paid for it. So in this case $100k.

Your opponent shoves all in UTG and flashes AK, he has the exact same amount of chips as you, and you find yourself with QQ. ICM helps explain why you would fold.

Here's why:
If you fold, your total equity in the tournament = $100k.
If you call and lose (48%) your equity in the tournament is $0
If you call and win (52%) your equity in the tournament - well, you'll need a calculator but it's going to be in the $120k range.

It's not necessary for you to work all the calculations for much larger tournaments but to simply understand that chips are worth more "real value" the fewer you have of them.

I agree that this is a fold but I think the estimate for the $EV value when you win is too small. I would guesstimate that the hero will jump to a $ value of approx. $170k. It would depend on a few of factors like blind size and if in them.
If it is closer to my estimate you can see you gamble a value of £100k to earn an extra $70k with a 52 to 48 hand equity lead.
Many may feel this is a gamble worth taking for the future edge you might gain by being armed with a big stack - it is quite a judgement and icm is not a perfect model but great to know.

For the OP question, players tend to have a natural in built icm calculator but it is often far from accurate, you see this when players nit up toward bubbles. As mentioned in previous posts the icm is apparent even in non all-in situations, the value of your stack in the tournament is always changing as the game progresses, even when you are not involved in hands.
Icm is considered a reasonable model for this effect. It is fairly easy to calculate at final tables due to you having the size of all the current chipstacks but the effect happens througout - it is very small far from the money.
Another interesting thing about icm is that it always adds up to 100% of the prize money so if your stack value has gone down compared to chip ev somebody else's must have gone up, you can notice this effect if you are say 4 handed and in third place but 4th is still close. If the chip leader goes allin against the small stack and loses it really hurts you, you drop into 4th and feel the pain especially if very low.
In this case you lost quite a bit of equity without even being in the hand, the chip leader didn't lose as much as the chips would indicate. In a way it's a bit like the chip leader has in effect used some of your equity to go allin (the other non involved player would also have lost a little equity, the 4th place gained loads, more than chip ev indicates). If the leader had won it would have knocked out the 4th place and boosted your equity without being in the hand.
To get a good feel for icm shoves at final table time it is good to do some work with an icm tool of some sort. Also playing several smaller SnGs like 45s can help as you get see/judge the effect more often than final tabling in big MTTs. Don't forget that icm can still be quite noticable when still far from the final table.
Single table SnGs are also a good place to learn icm but remember that for MTTs you will have to develop a shove/calling strategy for non icm cases as well as icm as they are quite different.

None math thinking could go something along the lines of:

If you are the big stack - punish the other players (esp the mid stacks)
If you are the mid stack vs the big stack - be very careful
If you are the small stack - take a sensible stand against mid and big stacks

and so forth