I’m a little surprised by these results but ICMizer wants to call with 91% of hands. Literally only folding 2x and 3x offsuit up to 8. Nash shoving range for SB is 78% of hands. As SB shoves tighter we can start folding more hands but it changes pretty slowly.

We start calling less than 90% if SB shoves less than 75%, but we still want to call with 81% of hands when SB shoves only 54.5%. We only start calling with <50% of hands if villain shoves <19.5% of hands.

I think that as much as we want to preserve our stack, we can’t turn down such good pot odds considering that we won’t have many hands before the blinds cripple us.

Turning Future Game Simulation on for 3 hands lowers Nash shoving range for SB to 70% and our calling range to 87%. Our ideal calling frequency declines at roughly the same rate as SB tightens up but falls off a cliff after they start shoving <22%. The cliff is less steep without FGS and doesn’t really start until they are shoving <19% of hands. 

I would expect most unknowns to shove slightly tighter than they should, and I prefer working with FGS than without. I think I’d be aiming to call ~80% of hands here. Every offsuit combo with two cards 5 or higher, J2+, 64o, 54o, and every suited combo other than23s-27s.

edit: BTW, Nash range without ICM has SB shoving 90% and us calling any 2. We can start folding a couple of hands if SB has 4bb but we’re still calling >90%. I’m not sure where you found a chart that tells you to call tighter than that.

That makes a lot of sense, can’t afford to play overly tight when doing so makes it much easier for villains to call us in the future. Thanks for the insight, ICM can be a strange cat. My chimp brain often wants to think ‘play tighter’ is the core of ICM but that’s about as helpful as “be aggressive” is to general strategy. Nowhere near enough information, and misleading if applied too broadly.

One thing to note is that the stack sizes above are listed a bit awkwardly, we’re actually squarely in 3rd. SB started the hand with 900 chips and we started with 2300, I only figured that out when I realized the total chips in play made no sense otherwise, unless 9 players started with 1433.3 chips. The 4th place stack is also in the BB next hand, so this may be one of those rare scenarios where FGS1 will deviate (ie. tighten up) further from our baseline strategy than FGS3. Still, your point stands. Our stack is dangerously close to having no fold equity in future hands, and giving uncontested chips to shorter stacks is much worse than allowing big stacks to steal.