PDA

View Full Version : ICM is often flawed


curtains
02-23-2005, 05:37 AM
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks. To use them to guide your play in these situations is often flat out wrong, yet a lot of people seem to be doing this. Any comments or opinions?

AtticusFinch
02-23-2005, 05:48 AM
[ QUOTE ]
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks. To use them to guide your play in these situations is often flat out wrong, yet a lot of people seem to be doing this. Any comments or opinions?

[/ QUOTE ]

Agreed 100%. I'm working on an adjustment, and will post my new formula once I come up with a decent rev.

ICM fails to take into account:

1) Blind sizes (this is a ridiculously huge oversight)
2) Your position (also huge)
3) Calling/Pushing standards for you and your opponents
4) Your relative skill vs. your opponents
5) Your folding equity before and after
6) Your opponents' folding equity, especially for short stacks.

My plan is to weight the stack sizes based on the above factors, then run the adjusted numbers through ICM. I believe that if you truly value the size of your current chip position correctly, then ICM will work as it should.

Opinions?

eastbay
02-23-2005, 05:48 AM
[ QUOTE ]
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks. To use them to guide your play in these situations is often flat out wrong, yet a lot of people seem to be doing this. Any comments or opinions?

[/ QUOTE ]

Give an example.

eastbay

curtains
02-23-2005, 05:49 AM
Yeah Im sure theres some way to improve it like this and make it a lot more applicable. It's just disconcerting to me to see so many people turn to it, in totally absurd situations where you will be given false numbers, and then making their plays accordingly.
I'm going to continue to sound like a broken record whenever anyone does so, but I think it's important, because this ICM is just wrong so often. People need to just learn how to play good poker.

eastbay
02-23-2005, 05:55 AM
[ QUOTE ]
[ QUOTE ]
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks. To use them to guide your play in these situations is often flat out wrong, yet a lot of people seem to be doing this. Any comments or opinions?

[/ QUOTE ]

Agreed 100%. I'm working on an adjustment, and will post my new formula once I come up with a decent rev.

ICM fails to take into account:


1) Blind sizes (this is a ridiculously huge oversight)
2) Your position (also huge)
3) Calling/Pushing standards for you and your opponents


[/ QUOTE ]

I think 1 & 2 are overstated and 3 just doesn't make any sense to me. What's 3 got to do with it?

Blind sizes are the same for everyone. Why is that a "ridiculous huge oversight"?

As for position, unless you're down to a couple of blinds, I'm not sure why this is a big deal, either. Your position changes with each hand and everyone gets their turn.

My guess is that any set of heuristics you try to apply to adjust for factors X or Y will end up only generating more confusion and questions about validity. I'm interested to see what you come up with, though.

In any case, I think it's maybe important to recognize that the $EV figures generated don't have to be accurate in an absolute sense. They only have to give an ordering which is reasonable. Big difference.

eastbay

eastbay
02-23-2005, 05:58 AM
[ QUOTE ]
Yeah Im sure theres some way to improve it like this and make it a lot more applicable. It's just disconcerting to me to see so many people turn to it, in totally absurd situations where you will be given false numbers, and then making their plays accordingly.


[/ QUOTE ]

Again, maybe you can point out a specific instance of this "absurd" valuation that leads to bad decisions that you are railing against.

eastbay

TheAmp
02-23-2005, 05:59 AM
[ QUOTE ]
[ QUOTE ]
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks. To use them to guide your play in these situations is often flat out wrong, yet a lot of people seem to be doing this. Any comments or opinions?

[/ QUOTE ]

Give an example.

eastbay

[/ QUOTE ]

http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Number=1786320&page=0&view=c ollapsed&sb=5&o=14

curtains
02-23-2005, 05:59 AM
Yes the problem is that a lot of the time the numbers are not reasonable. If the blinds are huge and you have 200 chips and your opponent has 10.
You dont have a 20x greater share of the prize pool than your opponent, however this is what the ICM tells you.

eastbay
02-23-2005, 06:01 AM
[ QUOTE ]

Yes the problem is that a lot of the time the numbers are not reasonable. If the blinds are huge and you have 200 chips and your opponent has 10.
You dont have a 20x greater share of the prize pool than your opponent, however this is what the ICM tells you.

[/ QUOTE ]

Well, all I can say to that is "duh." I think it's clear that when you're talking about less than one blind, other considerations are more important than stack ratios.

If that's the main point of your objection, I think it's at best a footnote on a valuable technique.

eastbay

curtains
02-23-2005, 06:03 AM
Thanks eastbay, that was the fuel that led to this post /images/graemlins/smile.gif People shouldn't use the ICM unless they can understand which situations it's applicable in and which it isn't.

curtains
02-23-2005, 06:05 AM
I never said it wasn't valuable, but I see too many posters rely on it in incorrect situations. If there are 3 threads a day where posters use ICM calculations that are just way off, then obviously it's being misused.
btw despite everything, Ive never had any use for the ICM in it's current form, but I'm sure it has real value in some situations.

eastbay
02-23-2005, 06:12 AM
[ QUOTE ]

btw despite everything, Ive never had any use for the ICM in it's current form, but I'm sure it has real value in some situations.

[/ QUOTE ]

Doyle never had any use for "the scientists'" ideas like EV, either. That doesn't mean it wasn't important, or wasn't there in his plays, it just meant that he learned how to make the +EV plays through experience and intuition, rather than analysis.

Same thing here. Just because you "have no use for it" doesn't mean it isn't valuable.

eastbay

curtains
02-23-2005, 06:16 AM
I didn't say it wasnt valuable. I just said I don't have any real use for it.

eastbay
02-23-2005, 06:18 AM
[ QUOTE ]
I didn't say it wasnt valuable. I just said I don't have any real use for it.

[/ QUOTE ]

I didn't say you did. But a casual observer might dismiss the entire idea without understanding why that might be a mistake.

eastbay

curtains
02-23-2005, 06:19 AM
Honestly I think that in order to use it properly, one has to already be a strong poker player with good understanding of poker math and odds.

Gotmilk
02-23-2005, 06:24 AM
[ QUOTE ]

Same thing here. Just because you "have no use for it" doesn't mean it isn't valuable.

[/ QUOTE ]

it is basically worthless in its current form because it can only be used in situations where position and blind size don't matter, which is basically never when you are 10-handed. If you ever use it for a sit and go in any situation it is going to be off, and off significantly (save the situations where stack sizes are equal--did you really need a calculator for that?). It will never tell you anything except the average EV of those stack sizes over a million trials with all the variables being randomized as well. Any poker situation is a specific situation, and since the calculator doesn't apply to specific situations it is of very little use.

[ QUOTE ]
That doesn't mean it wasn't important, or wasn't there in his plays, it just meant that he learned how to make the +EV plays through experience and intuition, rather than analysis.

[/ QUOTE ]

How, in its current form, can it teach you +EV plays through analysis?

eastbay
02-23-2005, 06:49 AM
[ QUOTE ]

it is basically worthless in its current form because it can only be used in situations where position and blind size don't matter,


[/ QUOTE ]

I don't think that's an accurate statement.

[ QUOTE ]

which is basically never when you are 10-handed.


[/ QUOTE ]

It is least useful 10-handed. It is most useful 5, 4, and 3 handed.

[ QUOTE ]

If you ever use it for a sit and go in any situation it is going to be off, and off significantly (save the situations where stack sizes are equal--did you really need a calculator for that?). It will never tell you anything except the average EV of those stack sizes over a million trials with all the variables being randomized as well. Any poker situation is a specific situation, and since the calculator doesn't apply to specific situations it is of very little use.



[/ QUOTE ]

I don't believe that's accurate either.

[ QUOTE ]
That doesn't mean it wasn't important, or wasn't there in his plays, it just meant that he learned how to make the +EV plays through experience and intuition, rather than analysis.

[/ QUOTE ]

How, in its current form, can it teach you +EV plays through analysis?

[/ QUOTE ]

Because it can account for the implications of prize structure whereas chipEV completely ignores this. This is the basic value of the idea.

Consider the case of calling a SB push from the BB in two cases, one on the bubble and one 3-handed. In a cash game, for given chip stacks and cards, your EV for a call is your EV and any other considerations are of secondary importance.

In a tournament, it may be a clear call 3-handed (ITM) but a clear fold 4-handed because of the prize structure. Equity models like ICM can be a very valuable guide for seeing the differing values of the call in the two situations.

To make it painfully clear, say you're in a tournament where top 3 win $1M and 4 wins nothing. 2 players and you have 10k chips, and 1 player has 500 chips. SB moves in over your BB of 100 for his 10k and shows you QJ. You have AK. Clearly prize structure has a big influence on what you should do. Equity models like ICM connect your decision to the prize structure to lead you towards better decisions than considering whether you're gaining or losing chips on the hand.

Obviously it would take a very stupid player to think taking 60/40 here was a good idea, but I've just drawn up an extreme example to show why equity considerations matter. ICM is only one way of getting at that, but for many situations it certainly is good enough to lead you in the right direction.

I know you understand all this - I'm really pointing it out for more inexperienced players who are trying to understand the concept of equity and why it matters, and why quantifying equity matters and is a worthy goal, even if the methods at our disposal aren't perfect. ICM is a very nice first cut way of making that quantification. Sure, there's situations where it doesn't apply at all, but there's also, IMO, far more situations where it does give useful information.

eastbay

Irieguy
02-23-2005, 06:57 AM
[ QUOTE ]
ICM numbers are wrong a lot of the time, especially late in an event with big blinds, and short stacks.

[/ QUOTE ]

As long as they are calculated correctly, the ICM numbers are never wrong.

The ICM does not make decisions. It just is. It can't be wrong any more than a tree can be wrong.

Now.. YOU can be wrong. I can be wrong. But the evidence can't be wrong... it's just evidence.

Don't hate the player, hate the game.

Irieguy

curtains
02-23-2005, 06:58 AM
What do you mean by calculated correctly? The ICM in it's current form will often not calculate the odds correctly.

ChrisV
02-23-2005, 10:04 AM
He means the ICM is the ICM and gives correct ICM results. Just like calculating chip EV gives correct chip EV results.

Determining the actual value of a play in an SNG is not possible. There are too many variables. It matters how you and the other players are going to play in the following hands, which is not the case in a cash game.

All we can do is refer to models. Chip EV is one such model, but it is pretty poor. The ICM is a great improvement. If you have reasons, in a specific hand, why the ICM results do not apply then all you have to do is state them. But I think it's a stretch to say the ICM "often" gives wrong results. Also I think it's a waste of time trying to develop a model that takes into account position, etc. Those variables are too complex to be modelled.

curtains
02-23-2005, 10:10 AM
The ICM has problems with short stacks, especially very short stacks. A lot of people who ask about ICM calculations, give situations where very short stacks are involved, and the ICM results in these cases are often a bit far from reality.

rachelwxm
02-23-2005, 10:45 AM
[ QUOTE ]

Yes the problem is that a lot of the time the numbers are not reasonable. If the blinds are huge and you have 200 chips and your opponent has 10.
You dont have a 20x greater share of the prize pool than your opponent, however this is what the ICM tells you.

[/ QUOTE ]

unfortunately in this case ICM is correct, sometimes intuition could be totally wrong.

Here is the math if you care (assuming hugh blinds):
1. If your chips ratio is 1:1 you have 50% prize pool, plain and simple.
2. If you have 3:1 chip lead, and your opponent take 50% chance to double up to situation 1 and 50% chance OTM. Therefore, he has 25% equity and you have 75%.
Etc...

Some people like play poker using intuitions, some people like to use math. Nothing wrong with either approach imo.

Any other examples? /images/graemlins/smile.gif

curtains
02-23-2005, 10:49 AM
Rachel the ICM isnt correct in this case. The situation Im referring to is when there are 4 players remaining and two players are so short they cannot afford the blinds.
Just because one is 20x shorter doesnt mean they should have 20x less equity. Sometimes the player with ten chips will actually be the favorite to finish in the money, depending on who plays the blind first. Either way, the 10 chipper is never a 20-1 underdog to make 3rd place, with another short stack around.

gumpzilla
02-23-2005, 10:49 AM
How could you have 75% equity in the prize pool? Assuming you're describing heads up on the end with a standard 5-3-2 payout structure, the winner only gets 5/8 = 62.5% of the prize pool. Winning every time can't do better than that.

rachelwxm
02-23-2005, 10:53 AM
[ QUOTE ]

Rachel the ICM isnt correct in this case. The situation Im referring to is when there are 4 players remaining and two players are so short they cannot afford the blinds.
Just because one is 20x shorter doesnt mean they should have 20x less equity. Sometimes the player with ten chips will actually be the favorite to finish in the money, depending on who plays the blind first. Either way, the 10 chipper is never a 20-1 underdog to make 3rd place, with another short stack around.

[/ QUOTE ]

If you mean bubble with stack 4000 4000 50 20 with blinds 200/400, I agree I would be cautious with ICM. But Heads Up ICM and Chip EV converge and lead to same conclusion.

rachelwxm
02-23-2005, 10:54 AM
[ QUOTE ]
How could you have 75% equity in the prize pool? Assuming you're describing heads up on the end with a standard 5-3-2 payout structure, the winner only gets 5/8 = 62.5% of the prize pool. Winning every time can't do better than that.

[/ QUOTE ]

For all simplification, I assume winner takes all. As it really does not matter since you can subtract second prize in Heads Up situation.

curtains
02-23-2005, 10:55 AM
Just take a look at the thread "Interesting bubble hand" that was posted earlier in this thread. That's a great example of a situation where the numbers are just bogus.

se2schul
02-23-2005, 11:02 AM
ICM will often be wrong just as the weather forecast is wrong. ICM is just a model which is a simplistic representation of a complex problem. If poker situations were simple, we wouldn't need to use a simplifying model. If you were to make ICM perfectly accurate, taking into account position, tendencies of opponents and such, you'd be on the verge of writing a bot.

ICM seems like a good tool for its purpose. In many cases, it will tell you what the correct move is. Somtimes it won't work, and that's just the nature of a model. It's important to recognize when a model is applicable to real life and where it deviates, and make the appropriate play.

I'd be interested in seeing examples of ICM not working. Do you have several specific examples that we could look at and maybe find patterns?

Steve

FishBurger
02-23-2005, 11:22 AM
It seems like the main problem is that ICM is a linear function based on the number of chips you and your opponents have while the blinds are not linear. The blinds are some type of step function that can take a big chunk out of your stack with each orbit.

ICM breaks down near the "edges." In my limited experience with it, it seems like it works best in the fat part of the bubble where everyone has greater than 4x BB. Once you ener a situation where two players have less than 2xBB, position and who is in the best shape to survive their blinds becomes more important.

P.S. I'm talking out of my ass.

eastbay
02-23-2005, 11:27 AM
[ QUOTE ]
He means the ICM is the ICM and gives correct ICM results. Just like calculating chip EV gives correct chip EV results.

Determining the actual value of a play in an SNG is not possible. There are too many variables. It matters how you and the other players are going to play in the following hands, which is not the case in a cash game.

All we can do is refer to models. Chip EV is one such model, but it is pretty poor. The ICM is a great improvement. If you have reasons, in a specific hand, why the ICM results do not apply then all you have to do is state them. But I think it's a stretch to say the ICM "often" gives wrong results. Also I think it's a waste of time trying to develop a model that takes into account position, etc. Those variables are too complex to be modelled.

[/ QUOTE ]

Finally some sanity in this thread.

eastbay

Zelcious
02-23-2005, 11:33 AM
ICM is not linear unless it's headsup, continous is maybe what you're looking for ?

eastbay
02-23-2005, 12:00 PM
[ QUOTE ]
[ QUOTE ]

Same thing here. Just because you "have no use for it" doesn't mean it isn't valuable.

[/ QUOTE ]

it is basically worthless in its current form because it can only be used in situations where position and blind size don't matter, which is basically never when you are 10-handed.

[/ QUOTE ]

The more I read this the more I think you are really misunderstanding what it is or what it is used for.

You give the impression that you went to dethgrind's web page, saw the inputs, didn't see any for blind size or position, and said "oh, this is worthless, it doesn't consider all the important stuff."

It is just an equity model. It is not a "play calculator."

eastbay

lorinda
02-23-2005, 12:09 PM
Fact 1: If any tool were so powerful that it could tell us how to play, the whole net would be full of super-winning bots in a month.

Fact 2: If these tools were worthless, nobody would be talking about them.

Fact 3: The people who can't use the ICM will not see why I posted Facts 1 and 2.

Lori

curtains
02-23-2005, 12:22 PM
Below are two links from the last 24 hours where posters asked for or somehow received ICM advice where ICM was not important or not applicable.

This all started when I posted a hand and started getting absurd responses about how my play was proven wrong according to the ICM. It was absurd to use the ICM in the aforementioned situation, but I can't find the link as the thread is rather old.


http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=1786320&page=0&view=colla psed&sb=5&o=14&fpart=1

http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=1783705&page=&view=&sb=5& o=&fpart=1&vc=1

lorinda
02-23-2005, 12:27 PM
I agree with you curtains, people rely too much on tools sometimes, and often forget entirely why they are using the tool in the first place.

Lori

curtains
02-23-2005, 12:31 PM
You are my favorite 2+2 poster even though you had 77 on the JT7 flop when I had JT. I need to join some KotZ's and get revenge.

eastbay
02-23-2005, 12:32 PM
[ QUOTE ]
Below are two links from the last 24 hours where posters asked for or somehow received ICM advice where ICM was not important or not applicable.

http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=1786320&page=0&view=colla psed&sb=5&o=14&fpart=1

http://forumserver.twoplustwo.com/showflat.php?Cat=&Number=1783705&page=&view=&sb=5& o=&fpart=1&vc=1

[/ QUOTE ]

I think you'd be hard pressed to explain why applying it in the 2nd case is wrong.

This is beating a dead horse now, IMO. We know that ICM has limitations for ultra-short stacks and "folding war" situations. Don't throw out the baby with the bathwater.

eastbay

curtains
02-23-2005, 12:33 PM
It's wrong because it's so absurd to consider folding AK there, that to resort to the ICM is insane.

lorinda
02-23-2005, 12:35 PM
It's wrong because it's so absurd to consider folding AK there, that to resort to the ICM is insane.

A tweak to your argument is required... the ICM should not be the target of your blame /images/graemlins/blush.gif

Lori

curtains
02-23-2005, 12:43 PM
I have no problem with the ICM. The title of this thread was misleading. I was just overwhelmed by ICM follies.
The 2nd link is just an example of someone using ICM as a crutch. This player shouldn't even know what ICM is until they learn the basics of poker, it will just distract them from what's really important.

eastbay
02-23-2005, 12:45 PM
[ QUOTE ]

It's wrong because it's so absurd to consider folding AK there, that to resort to the ICM is insane.

[/ QUOTE ]

Hmm. A bit of a logic error there. Of course I know what you mean. But the clueless will still get the right answer by applying it. They just shouldn't have to. But that's a different issue entirely.

eastbay

TheAmp
02-23-2005, 12:45 PM
Curtains,
I'm afraid you might be in some sort of "loop", and I want to try to help you out of it.

The ICM model does not intend to suggest how to play a hand. Any hand. It merely gives you data that can prove useful - if, and only if, you use good judgment.

You pointed out that in order to make a correct decision, it is essential to weigh other criteria as well, such as when blinds will change, position etc...I am sure everyone agrees with you. At least I do, for what it is worth.

The conclusion shouldn't be that "ICM is useless.." but that we should understand how ICM figures are calculated, and how to use them in our decision process.

Regards,
S.J.

curtains
02-23-2005, 12:47 PM
I never said that the ICM was useless!!!!!!! /images/graemlins/smile.gif I basically agree with you TheAmp.

curtains
02-23-2005, 12:49 PM
Yeah really the title to this thread is not the best choice. I was inspired because I really felt like it was flawed in the thread where the guy has 99.

TheAmp
02-23-2005, 12:53 PM
[ QUOTE ]

I was inspired because I really felt like it was flawed in the thread where the guy has 99.

[/ QUOTE ]

The guy = The Amp.
A 1967 model marshall tube amplifier...(guitar lovers will know what I mean).

curtains
02-23-2005, 12:57 PM
oh yeah that was you /images/graemlins/smile.gif

AtticusFinch
02-24-2005, 12:18 AM
[ QUOTE ]
[ QUOTE ]

1) Blind sizes (this is a ridiculously huge oversight)
2) Your position (also huge)
3) Calling/Pushing standards for you and your opponents


[/ QUOTE ]

I think 1 & 2 are overstated and 3 just doesn't make any sense to me. What's 3 got to do with it?


[/ QUOTE ]

My heuristics are really all about estimating folding equity. When deciding to fold a hand that may technically be +$EV according to ICM, many top players (e.g. Gigabet) speak of potential to make more with their own steals, which comes down to folding equity. Good players seem to understand this intuitively, but I'm trying to come up with a more concrete estimate of folding equity with my stats. Thus pushing/calling standards are important.

[ QUOTE ]

Blind sizes are the same for everyone. Why is that a "ridiculous huge oversight"?


[/ QUOTE ]

Because the higher the blind sizes are, the less accurate ICM estimations of your "share" of the pot tend to be. At party especially the blinds are often enormous in situations where ICM is used. Plus folding equity depends on them. (See above).

[ QUOTE ]

As for position, unless you're down to a couple of blinds, I'm not sure why this is a big deal, either. Your position changes with each hand and everyone gets their turn.


[/ QUOTE ]

This also affects folding equity. The more players in front of you, the less folding equity you have. However, on the button you'll have more opportunities to catch a hand that meets your pushing standards. This affects the "value" of your folding equity. Conversely, as you say, when UTG and down to only, say, 3 blinds, this is the last time you can expect to have any folding equity at all.

[ QUOTE ]

My guess is that any set of heuristics you try to apply to adjust for factors X or Y will end up only generating more confusion and questions about validity.


[/ QUOTE ]

I don't doubt this. I'm just going to give it a shot and see where it leads me. The worst case is I gain a deeper understanding of all the variables and how they interact.

[ QUOTE ]

I'm interested to see what you come up with, though.


[/ QUOTE ]

As I said, I'll post my results if and when I have something that appears useful, or at least interesting.

[ QUOTE ]

In any case, I think it's maybe important to recognize that the $EV figures generated don't have to be accurate in an absolute sense. They only have to give an ordering which is reasonable. Big difference.


[/ QUOTE ]

Agreed. Thanks for your comments.

-AF.

AtticusFinch
02-24-2005, 12:28 AM
[ QUOTE ]

Just take a look at the thread "Interesting bubble hand" that was posted earlier in this thread. That's a great example of a situation where the numbers are just bogus.

[/ QUOTE ]

That's because that was a satellite where all payouts ITM were equal. ICM doesn't even apply there.

curtains
02-24-2005, 12:30 AM
That wasnt a satellite, it was a normal sit and go.

eastbay
02-24-2005, 12:33 AM
I think your points are valid so far...

[ QUOTE ]

As for position, unless you're down to a couple of blinds, I'm not sure why this is a big deal, either. Your position changes with each hand and everyone gets their turn.


[/ QUOTE ]

This also affects folding equity. The more players in front of you, the less folding equity you have. However, on the button you'll have more opportunities to catch a hand that meets your pushing standards. This affects the "value" of your folding equity. Conversely, as you say, when UTG and down to only, say, 3 blinds, this is the last time you can expect to have any folding equity at all.


[/ QUOTE ]

I think this misses the mark badly as long as we're not talking about a blind or two's worth of chips. These are hugely important effects but they don't enter the equity model strongly at all, as long as everybody's got a few chips to play with.

This enters the analysis in the likelihood of you being called. This is nearly orthogonal to the issue of equity for given chip stacks.

To make it more clear, say you're ITM at a final table and making a deal. Deals are all about getting at least your fair share of the equity. Would you say "hey, that deal would be fair, except I'm on the button so I deserve more." Unless someone is about to bust from the blinds, that's just a silly thing to say, because it hardly matters at all for your equity.

eastbay

AtticusFinch
02-24-2005, 12:33 AM
[ QUOTE ]

That wasnt a satellite, it was a normal sit and go.

[/ QUOTE ]

My bad -- I confused it with this thread: http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Number=1775947&page=&view=&s b=5&o=

eastbay
02-24-2005, 12:38 AM
[ QUOTE ]
[ QUOTE ]

Just take a look at the thread "Interesting bubble hand" that was posted earlier in this thread. That's a great example of a situation where the numbers are just bogus.

[/ QUOTE ]

That's because that was a satellite where all payouts ITM were equal. ICM doesn't even apply there.

[/ QUOTE ]

Whoa. What are you saying?

A satellite where all players get the same payout is a perfect way to show off why equity is more important than chipEV.

In fact, I used exactly that example to demonstrate the value of the idea in a reply to GotMilk.

eastbay

AtticusFinch
02-24-2005, 12:40 AM
[ QUOTE ]

To make it more clear, say you're ITM at a final table and making a deal. Deals are all about getting at least your fair share of the equity. Would you say "hey, that deal would be fair, except I'm on the button so I deserve more." Unless someone is about to bust from the blinds, that's just a silly thing to say, because it hardly matters at all for your equity.


[/ QUOTE ]

Great analogy. I think the difference shows up mostly with short stacks again.

I think folding equity is a roughly logarithmic function. If you double a short stack, you make a huge gain in folding equity. If you double the biggest stack at the table, however, your folding equity is essentially unchanged.

While you may get laughed at if you try it, I honestly think a short stack is entitled to a slightly larger share if on the button than UTG. Think about the reverse argument: "Well you're going to be all-in on the BB anyway, so you deserve a smaller share." Not a ridiculous notion, is it?

More abstractly, which position would you rather be in? Clearly with a short stack the answer cannot be "I don't care." So there is some equity value there.

AtticusFinch
02-24-2005, 12:43 AM
[ QUOTE ]

I think folding equity is a roughly logarithmic function. If you double a short stack, you make a huge gain in folding equity. If you double the biggest stack at the table, however, your folding equity is essentially unchanged.

[/ QUOTE ]

By the way, I believe this is the primary reason why ICM is inaccurate when short stacks are involved, and this is the problem I am trying to rectify.

AtticusFinch
02-24-2005, 12:44 AM
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]

Just take a look at the thread "Interesting bubble hand" that was posted earlier in this thread. That's a great example of a situation where the numbers are just bogus.

[/ QUOTE ]

That's because that was a satellite where all payouts ITM were equal. ICM doesn't even apply there.

[/ QUOTE ]

Whoa. What are you saying?

A satellite where all players get the same payout is a perfect way to show off why equity is more important than chipEV.

In fact, I used exactly that example to demonstrate the value of the idea in a reply to GotMilk.

eastbay

[/ QUOTE ]

I should have been more clear. I was referring specifically to the online ICM calculator, which assumes a standard payout structure. Clearly the ICM works here, but you have to make the necessary adjustment for the changed structure.

ilya
02-24-2005, 01:19 AM
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
[ QUOTE ]

Just take a look at the thread "Interesting bubble hand" that was posted earlier in this thread. That's a great example of a situation where the numbers are just bogus.

[/ QUOTE ]

That's because that was a satellite where all payouts ITM were equal. ICM doesn't even apply there.

[/ QUOTE ]

Whoa. What are you saying?

A satellite where all players get the same payout is a perfect way to show off why equity is more important than chipEV.

In fact, I used exactly that example to demonstrate the value of the idea in a reply to GotMilk.

eastbay

[/ QUOTE ]

I should have been more clear. I was referring specifically to the online ICM calculator, which assumes a standard payout structure. Clearly the ICM works here, but you have to make the necessary adjustment for the changed structure.

[/ QUOTE ]

Dethgrind has been kind enough to change this. You can now specify the prize structure in the online ICM calculator.

AtticusFinch
02-24-2005, 01:21 AM
[ QUOTE ]

Dethgrind has been kind enough to change this. You can now specify the prize structure in the online ICM calculator.

[/ QUOTE ]

Cool! Can you give a link? The one at the address I have still assumes the standard payout.

ilya
02-24-2005, 01:23 AM
[ QUOTE ]
[ QUOTE ]

Dethgrind has been kind enough to change this. You can now specify the prize structure in the online ICM calculator.

[/ QUOTE ]

Cool! Can you give a link? The one at the address I have still assumes the standard payout.

[/ QUOTE ]

Spiffy new version of ICM calculator (http://www.bol.ucla.edu/~sharnett/ICM/ICM.html)