Poker-AI.org

Something interesting when crunching a strategy for a 2BB stack size game, it doesn't always push/fold.

For example, with: 74o, 72s, 63o, 54o, 52s it's calling some high portion of the time from the SB.

That depends on what abstraction you allow it to play right?
What abstraction are you using for your number crunching here?

_________________
Cheers.

Same as the 6bb strat.

Interesting. This kind of supports my theory that no action really ever dominates another action for all scenarios.
If you add calling or raising a smaller (maybe even larger?) amount, then that action will likely get some percentage of play somewhere.

Too many people think P/F is OBVIOUSLY better than calling with small stacks but it's really not game theoretically. It just makes life easier and obfuscates the strength of the hand.

_________________
Cheers.

I guess in a fold/call/push scenario it's +EV to call sometimes and hope your opponent has crap too.

For the 2bb example: are the listed hands the only calling hands? If so, BB should basically raise any hand vs limps and thus we cannot fold anymore given the pot odds. This means, we should shove the hand as there we might have a tiny bit of FE which should outweight the cases where BB has 72o and checks vs our limp.

I think so. Are you saying by this p2bb that it's a flawed strategy to call at all from the button?

Not necessarily. Given the 2bb stacks and the fact that we basically can't limp/fold, I'd find it weird that its valid to limp only the worst hands (unbalanced): villain can make easy value shoves with most hands and his worst hands can play perfectly against us as our range is small and defined.
Did you run the strategy xM times and track the EV of SB and then compare it with a PoF SB strategy?

I haven't played it against my P/F player. What's your EQ look like for 2BB?

Unfortunately I don't have time to develop on the strategies currently However, I'd guess that 2BB is basically a PoF game, so to compare you could just find the optimal solution in CREV. Even though calling might be viable and part of a GTO solution, the difference between that and a PoF should be close to 0.