I have a question regarding a very weird outcome of a simulation:
Both times 100bb stacks, both times 23.5bb 4bets.
(1) Villain opens, I 3bet AKo, villain 4bets with a range of TT+,AK, I jam, villain calls.
My jamEV (when facing the 4bet) is -9.215bb
(2) Villain opens, I 3b AKo, villain 4bets TT+,AK, I call preflop. I openjam any flop blind, and have villain call perfectly (the alt+d filter). This results in him calling ~75% of the time (folding TT-QQ (KK) on K (A) high flops, obviously).
The EV of calling the 4b preflop and jamming any flop blindly and having him call perfectly is ~ -1bb.
How can this be possible? In the first scenario I get the money in preflop and he can't fold TT on an A high board obviously, so I either win or lose a stack vs TT depending on the board.
In the second scenario I lose the same amount of money when I don't hit (my stack), but when I hit a good flop for me TT will fold (A73r), so clearly my EV on calling preflop and jamming any flop but letting him fold some flops should be worse than -9bb. Yet cardrunnersEV tells me my EV is better
What is going on? What (if I'm correct) is the bug and how can I circumvent it?
I can send you the cardrunnersEV files if you want, but it should be very easy to reproduce