Image Image Image




Post new topic Reply to topic  [ 10 posts ] 
Author Message
 Post subject: Strange ICM decision
PostPosted: Mon May 25, 2009 9:57 am 
Offline
Senior member
User avatar

Posts: 232
Favourite Bot: Poki
Why does the ICM-calculator always push crap hands when 2 players are in the pot before you?

Here is an example hand:
- player0 sitting in seat 1 with $1089.00 [Dealer]
- player1 sitting in seat 2 with $1036.00
- hero sitting in seat 3 with $3390.00
- player2 sitting in seat 4 with $1985.00

player1 posted the small blind - $50.00
hero posted the big blind - $100.00
** Dealing card to hero: 10 of h, J of h
player2 called - $100.00
player0 went all-in - $1089.00
player1 went all-in - $986.00
hero went all-in - $3290.00


Top
 Profile E-mail  
 
 Post subject: Re: strange ICM decision
PostPosted: Mon May 25, 2009 10:03 am 
Offline
Newbie
User avatar

Posts: 373
Location: The Netherlands
Favourite Bot: The Crusher v2
Probably it's calculating the win % of TJs against 2-3 opponents and calculating as well how much there can be won in chips. This value's lie pretty close to each other.

ICM also takes the price structure into account, so with this hand it would have 90%+ certaintiy that it will become first so that makes it a supposed to be profitable play :S

I think this is a nice example why those kind of things are never optimal cause for anyone else it would be a clear fold. ICM doesn't take into account the possibilty to wait for a better spot.

Other explanation can be crappy setup, buggy bot, misscrape, etc.


Top
 Profile E-mail  
 
 Post subject: Re: strange ICM decision
PostPosted: Mon May 25, 2009 3:19 pm 
Offline
PokerAI fellow
User avatar

Posts: 1673
Favourite Bot: none
ICM Nash Calculator Results

This is the resulting range that Holdemresources.net gives you [3.9%, 99+ AKs AKo].
I suspect that you either miscraped the hand OR, what I most likely think is the reason, SNGEGT can't deal with limpers and you misinterpreted the ranges.

Hard to say without knowing how you set this all up.

_________________
Cheers.


Top
 Profile  
 
 Post subject: Re: strange ICM decision
PostPosted: Tue May 26, 2009 2:46 pm 
Offline
PokerAI fellow
User avatar

Posts: 1115
Favourite Bot: Johnny #5
Coffee4tw wrote:
I most likely think is the reason, SNGEGT can't deal with limpers and you misinterpreted the ranges.

Hard to say without knowing how you set this all up.


Ditto.


Top
 Profile  
 
 Post subject: Re: Strange ICM decision
PostPosted: Tue May 26, 2009 9:50 pm 
Offline
PokerAI fellow
User avatar

Posts: 1673
Favourite Bot: none
And btw whatever newbee said is total crap. I don't think he knows at all what he is talking about here.

_________________
Cheers.


Top
 Profile  
 
 Post subject: Re: Strange ICM decision
PostPosted: Tue May 26, 2009 10:09 pm 
Offline
PokerAI fellow
User avatar

Posts: 7731
Favourite Bot: V12
As usual :)

_________________
indiana


Top
 Profile E-mail  
 
 Post subject: Re: strange ICM decision
PostPosted: Wed May 27, 2009 6:20 pm 
Offline
Senior member
User avatar

Posts: 232
Favourite Bot: Poki
Coffee4tw wrote:
ICM Nash Calculator Results

This is the resulting range that Holdemresources.net gives you [3.9%, 99+ AKs AKo].
I suspect that you either miscraped the hand OR, what I most likely think is the reason, SNGEGT can't deal with limpers and you misinterpreted the ranges.

Hard to say without knowing how you set this all up.


I didn't misscraped the hand, I set up the same situation manually and it got same result

Do you have any tips for how to get around this? maybe I should treat limpers as folders or something?


Top
 Profile E-mail  
 
 Post subject: Re: Strange ICM decision
PostPosted: Thu May 28, 2009 7:39 am 
Offline
Newbie
User avatar

Posts: 373
Location: The Netherlands
Favourite Bot: The Crusher v2
Lol, they all know it so well!

Maybe I had to add 90% certainty to win the tournament when we win this hand, but some dumb figures misinterpreted this ofc. Funny to see that most people don't even understand the concept of ICM.


Top
 Profile E-mail  
 
 Post subject: Re: Strange ICM decision
PostPosted: Thu May 28, 2009 9:12 am 
Offline
PokerAI fellow
User avatar

Posts: 1673
Favourite Bot: none
Having limpers in a hand is a huge problem for these calculations.
What it basically does is try to find an approximate nash equilibrium for the given stacks sizes on the base of ICM as $EV calculation. This ANEQ is restricted to jam and fold only, therefore limping is not supported by the algorithm at all. It is possible that this result it gives you is a product of the increased pot size or the change in position of the big blind (one could argue that the limper is now the big blind).
Anyhow the algorithm has to be completely rewritten to allow limping in the set of strategic actions. I don't know how though to produce good results as this would require some sort of post-flop acting aswell (imagine the big blind checking after the limper).

Now would it be a good approach to consider the limper either as fold or as shove? I think fold is bad because he might call you a couple of times (slow played monsters) and then your whole NEQ calculation is off. You might be pushing to loose if you do this.
On the other hand considering him as a shove will probably lead to a better profit because we are rather pushing to tight than to loose when we have a limper, this will not be an euqilibrium strategy either though.

Another huge problem SNGEGT has is that it only considers 2-way pots if I recall correctly. You can only push or call a pusher, there is no overcalling after two people are allin, I think, which of course makes this less reliable aswell.

@newbee... You have no clue about what I know and what I don't know. But I know that you don't know what sngegt does at all ...

_________________
Cheers.


Top
 Profile  
 
 Post subject: Re: Strange ICM decision
PostPosted: Fri Jun 26, 2009 10:36 pm 
Offline
Senior member
User avatar

Posts: 138
Favourite Bot: Custom
I think it may tell you to call a little light here is because if we call we are almost guarenteed to get 3rd place and the differance bettween 3rd place and 2nd is a very small equity increase where if we get first it is a much bigger jump. I will do the calcs later and see what it comes up with.

I dont think the limper will have much effect in the calculation.

What ranges are you using for each player?


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 10 posts ] 


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: