problem determining optimal push/call ranges

[campfirewest]

I've been trying to build a table of optimal push and call ranges for various permutations of #players, stack sizes, blind levels. I started with the simple case of:
-2 players
-blinds 400/800
-6750 chips each player

I used SNGEGT and started with a call range of 10%, then looked at the optimal push range, then found the 0-EV call range for that push range, and on and on like sheets says to do. The problem is I end up in a do loop and don't converge on an answer. Here's what I came up with:

-call range 10% = push range 100%
-push range 100% = call range 69%
-call range 69% = push range 47%
push range 47% = call range 31%
call range 31% = push range 100% (now I'm back to step 2 and the loop repeats)

Does anyone see what I'm doing wrong? Any help is greatly appreciated.

[jukofyork]

[ QUOTE ]
I've been trying to build a table of optimal push and call ranges for various permutations of #players, stack sizes, blind levels. I started with the simple case of:
-2 players
-blinds 400/800
-6750 chips each player

I used SNGEGT and started with a call range of 10%, then looked at the optimal push range, then found the 0-EV call range for that push range, and on and on like sheets says to do. The problem is I end up in a do loop and don't converge on an answer. Here's what I came up with:

-call range 10% = push range 100%
-push range 100% = call range 69%
-call range 69% = push range 47%
push range 47% = call range 31%
call range 31% = push range 100% (now I'm back to step 2 and the loop repeats)

Does anyone see what I'm doing wrong? Any help is greatly appreciated.

[/ QUOTE ]
If it doesn't converge then try to take the mean of the previous two results, eg:

-call range 10% -> push range 100%
-push range 100% -> call range 69%
-call range (69+10)/2 = 39.5% -> push range X%
-push range (X+100)/2 = Y% -> call range Z%
-call range (Z+39.5)/2 = ...

If that still doesn't converge then try to dampen more by weighting the previous result more (eg: 69*0.4 + 10*0.6, etc).

Juk [img]/images/graemlins/smile.gif[/img]

[campfirewest]

Thanks dude! I came up with ~58% push and ~38% call a shove using your method. Anyone agree or disagree with this result?

[BradleyT]

Hmm, disagree?

ICM Nash Calculator Results

[jukofyork]

[ QUOTE ]
Hmm, disagree?

ICM Nash Calculator Results

[/ QUOTE ]
I think it must be due to the OP just using the top % of hands rather than the range itself?

Juk [img]/images/graemlins/smile.gif[/img]