Re: ICM
 

Send a private message to Eastbay.

Re: ICM
 

[ QUOTE ]
Send a private message to Eastbay.

[/ QUOTE ]

Yes, eastbay is the man WRT ICM calcs. The SNGPT (SNG Analyzer) is a tool that helps run through the ICM calcs for SNG pushbotting in an automated fashion.

(BTW Exit - the avatar is my dog from halloween ) [img]/images/graemlins/smile.gif[/img]

Re: How not to deal at the final table
 

I watched this last night... Actually played on a table for a while with Phat before going out 58th. ( I'm frodolives on stars ).

The thing I thought I noticed beyond what is mentioned in your post, is that in the conversation prior to what you have listed here, Gunny actually OFFERS a chip chop... to which Phat quickly says 50/50... Don't think he understood what a chip chop is...

Re: ICM
 

[ QUOTE ]
[ QUOTE ]
Independent Chip Model - Basically, a method for determining chip value during a tourney.

[/ QUOTE ]

So, being ICM-clueless myself, I went to that page and read about it. Then I downloaded the C source, compiled it (Cygwin) without difficulty and ran it. It looked to my cursory glance like it was s'posed to determine the EV's for stacks of 18, 24, and 30 in a tournament down to three players with payoffs of .2, .3, and .5

So Q1: Are my assumptions correct?


[/ QUOTE ]
Yes.

[ QUOTE ]

When I ran it, it printed out:

72.000 0.25 0.3333 0.4166667

which seems to be:

total_chips %stack1 %stack2 %stack3


[/ QUOTE ]
Which is actually: total_chips and then the probability of finishing first for each player,
which is simply the proportion of the chips in play that they currently hold.

[ QUOTE ]

Then it went into a long waiting period and then spit out one row of numbers:

1816214400 0.250000 0.303571 0.446429 0.000000 0.000000 0.000000 0.305357


[/ QUOTE ]
The first number seems to be the number of times the innermost loop ran.
The second number is player 1's chance of winning.
The third number is player 1's chance of finishing second.
The fourth number is player 1's chance of finishing third.
I supect the fifth, sixth, seventh numbers are 4th 5th + 6th position probability.
The last number is player 1's ICM value. - this is the useful number.

[ QUOTE ]

and a table. The table appears to be probabilities of people finishing in certain positions. But there's no guidance on how to interpret that.


[/ QUOTE ]

The table seems to be:
Postion, percentage of prizepool for that spot, chance player1 will finish in that spot, chance player1 will finish in a lower spot.

I'm not sure why this table is useful.

The last number is player 1's ICM value. (again).

[ QUOTE ]

In short, this program (presumably) does the heavy lifting nicely, but doesn't cross the t's and dot the i's, enabling a lay user to make much use of it.


[/ QUOTE ]
It only seems to give the numbers for player1.
It also seems to be much more complicated and take much longer than I expected it should.
Although it does seem to be able to cope with 16 players and paying out 10 spots.

My ICM code only calculates for paying out 3 spots but is effectively instant.

Insty.

Re: How not to deal at the final table
 

[ QUOTE ]
Gunny actually OFFERS a chip chop... to which Phat quickly says 50/50... Don't think he understood what a chip chop is...

[/ QUOTE ]

Tragic.