Hey guys... was trying out the "River - AJ" combat trainer and spotted a flaw.
Being that the UTG opponent is programmed to auto-call our turn bet, he can never lead or c/r. I quickly noticed he was showing hands that should clearly be a lead or c/r for protection/value/semi-bluff on the turn. This makes the range assessment on the river unrealistic.
Not necessarily complaining, but figured it was worth pointing out.
It looks like you're new here. If you want to get involved, click one of these buttons!
Thanks for the comment. We try to design them to only be in realistic ranges, so this would be something that needs attention. Can you give some specific examples so I can look at them? I'm not as sure there are things that would clearly be played a certain way on the turn - I see a lot of variance in human play that isn't necessarily what you would call "standard" or "optimal". But we try to have the opponents play in a way that matches what the normal range of humans do. Turn play becomes widely variant and I do see a ton of check/call on the turn. What we need to look at is whether it follows the play style of the bot in question and whether it is realistic they would have gotten to that point in the hand where c/c would be something they might do.
I'm not doubting that there are issues - it's very hard to get things realistic on simulations like this, but I need some specific hands to look at to be able to adjust things if necessary. Feel free to post links to specific hands.