Open Side Menu Go to the Top
Register
ICM Problem ICM Problem

07-09-2024 , 05:19 PM
I was playing a home game the other night and came into this kind of unique situation with 88's. I tried to figure out the math myself but am having a bit of a hard time. If anyone could help me with this problem I would be really appreciative of it. A bit out of my realm here.

(Ranges were calculated using ProPokerTools and all percentages were rounded down to the nearest whole number to simplify)

SCENARIO:

900 000 Chips In Play

1st Place Pays $125.00
2nd Place Pays $75.00

Greg: 200 000
Graeme: 300 000 + SB (330 000)
Cameron: 310 000 + BB (370 000)

RANGES:

-Cameron Vs Graeme Vs Greg-

Cameron: 88 42%
Greg: AA-22,A,KQ-K4 27%
Graeme: AA-44,Ax 30%

-Cameron Vs Graeme-

Cameron: 58%
Graeme: 41%

POSSIBLE SOLUTIONS:

***OPTION 1 - CALL***

If Graeme wins (30%) I win $75.00 for 2nd with 40 000 left over. ICM $9.00 + Prize of $75.00 = $84.00

If Greg wins (27%) and I lose to Graeme (41%) I am left with 40 000. ICM = $9.00

If Greg wins (27%) but I beat Graeme (58%) I am left with 300 000. I get $75.00 + $16.00 of ICM. Total : $91.00

If I win I knock them both out (42%) and win the tournament pocketting $125.00.

***OPTION 2 - FOLD***

If Graeme wins (55%) I win $75 for 2nd + $17.00 of ICM (34% of the remaining $50 prize pool). Total = $92.00

If Greg wins (44%) I am left with 310 000. ICM = $68.00

QUESTION:

What is the best option?
ICM Problem Quote
07-09-2024 , 09:48 PM
Uh... what are the blinds and what is the action?
ICM Problem Quote
07-09-2024 , 10:02 PM
The action is Dealer (Greg) Jams all in, Small Blind Calls (Graeme), and the BB has the 8's (Cameron). The BB is 60,000.
ICM Problem Quote
07-09-2024 , 10:21 PM
Generally,

EV =[Sum over j] [ Pr(Event j Occurs) * ($ Event j)]

Here, the events are the various win/lose cases.

For your question, you calculate the EV's of Call and Fold and choose that action which has the higher EV


Assuming your win probabilities and ICM values are correct:

EVcall = 67.7

EVfold = 0.55* 92 + 0.44 * 68= 80.5

So, on this basis, Fold has higher EV than Call.
ICM Problem Quote
07-10-2024 , 06:12 AM
Third pays nothing?

This is pretty tough. Being the stone bubble should dictate a very tight range from you to get it in, since you lose so much more equity by losing the pot than you stand to gain from winning it. Either Greg busts or Graeme is down to two big blinds on the button, while you still have a little over 5BB behind.

On the other hand, you only have six big blinds, so there's not really any time to wait. Very importantly, you do cover everyone, so it's only really bad for you if you finish third in the hand. So I think if you call there's a strong chance you lock up at least second-- about 88% by your calculations, between the 73% chance Greg busts and the times Greg survives but you bust Graeme. That's probably worth it with a hand as strong as 88.

(Those ranges you provided seem pretty tight to me, but I take it you know these players better than I do. I can't see Greg folding any two Broadway cards here, though, and probably goes with a lot of suited connectors. I'd be surprised if Graeme folded something like KQs; KQo/KJs I think are calls here too-- better calls than 22, anyway-- and maybe even some lighter kings or QJs.)

If the big blind was 30k I'd probably fold, both for the tighter ranges that would suggest from each player (particularly SB) and to preserve a viable stack. As it stands, though, when you're this short, nobody really has the power to wait out anyone else.

I think I lean call because everyone is so short, but I'm not sure at all.
ICM Problem Quote
07-10-2024 , 11:30 AM
Quote:
Originally Posted by statmanhal
Generally,

EV =[Sum over j] [ Pr(Event j Occurs) * ($ Event j)]

Here, the events are the various win/lose cases.

For your question, you calculate the EV's of Call and Fold and choose that action which has the higher EV


Assuming your win probabilities and ICM values are correct:

EVcall = 67.7

EVfold = 0.55* 92 + 0.44 * 68= 80.5

So, on this basis, Fold has higher EV than Call.

I think you calculated EVcall wrong. I got $93.25 30%($84) + 27%*41%*($9) + 27%*58%*($91) + 42%*($125)

its a clear call when you cover everyone. Its gonna be really close if you dont have the sb covered. But you can do the same EV exercises based on that situation and solve that yourself. You actually did most of the legwork already, just needed to actually do the EV calc.


Edit: I was bored at work and interested so I did the math if the sb and bb stacks were swapped to see how not covering the sb affected it. Its still a call, although closer. I got $85EV for calling in that scenario.

Last edited by ledn; 07-10-2024 at 11:54 AM.
ICM Problem Quote
07-10-2024 , 05:51 PM
Quote:
Originally Posted by ledn
I think you calculated EVcall wrong. I got $93.25 30%($84) + 27%*41%*($9) + 27%*58%*($91) + 42%*($125)

.
Correct! I used the same equation but failed to add the first term .30*84 = 25.2, the difference in the results (except for rounding)

Okay on the math, lousy on the arithmetic!
ICM Problem Quote
07-11-2024 , 01:02 AM
After talking to a couple of people who say your range should be very tight for calling off here, I'm even less sure. You can't really go wrong by folding because of the bubble, but with how shallow the stacks are you also don't have much chance to wait anyone out.

Probably then the right play is to fold. Even if Greg wins, Graeme is so short he's going to have to survive multiple all-ins. I'm really not sure, though. Like I said, easy fold if effective stacks were twice as deep and BB had 13BB.
ICM Problem Quote
07-11-2024 , 02:08 PM
Thank you!! That formula is very helpful! I do not know what some of those functions or symbols mean but that gives me a great start to learn this in the next day or two when I have more time.

Great posts all around tbh.
ICM Problem Quote

      
m