As of November 2021, the APT Forum is closed to new posts. Like with many online forums, usage has decreased in recent years. All previous posts are still available.

AI 4-bet on the draw

edited July 2017 in Specific Hand Questions

KQo -- one of the classic "Trouble Hands" (Doyle, also cited in Harrington on Cash Games). KQo is not great multi-way OOP, but I'm not going to consider folding when I can see a free flop. So I was really playing "check/fold" more than "considering raising."

But check out the bot's 4-betting on the draw (flop).

Watch replay

I considered folding (flop), and I feel like I would have definitely folded if the opponent weren't a bot with known aggressive tendencies.

How would you rate her 4-betting the flop on the draw as a strategy? Does it look more like she's repping a set? Should I have gotten out of the way (not outcome-dependent, but strategy-wise)? Was it a non-optimal, overaggressive move on her part? Or is this something we should look at doing in cash games?

I suppose she's figuring it's a 12-outer (ace + FD, assuming I didn't hit my set), so if she can see turn and river, the pot odds are fine. But why not flat at least one of those bets and just peel off another card? If she'd flatted and waited for the turn and seen that Q hit, I think she would have saved 3/4 of her stack. And on the turn, she was down to 9 outs needing to call 20k into 75k with one card to go and no further betting. Or is she figuring 12 outs is worth it (i.e.: is she still thinking aces would be good)?

I suppose one could make a case for her shoving the 4-bet (flop) -- although you need 14 outs with 2 cards to come for a 50/50 all-in, but there is $600 in dead money (not much) and some fold equity (actually significant). I think she's got to either do that or be prepared to give up on the turn (especially with the Q falling). At least giving up on the turn would have saved her half her stack.

Or did I misplay it?

I definitely had her as being on the flush draw or maybe having hit a lower set (8's or 3's). I thought AA/AQ/KK were unlikely both because there was no preflop raise and because I had one of each of Q and K (limped/slowplayed aces were possible, I guess).


  • monkeysystem

    I might've read her UTG open limp as a set mine, a balancing move, or slowplayed AA. Her raise on the flop looked like a semi-bluff with nine clean (nut flush) and three dirty (aces) outs. Your 3-bet was the standard defense against a semi-bluff. Her 4-bet seems really questionable. Maybe she thought you were bluffing as a blind defense? What was your table image up that time, particularly regarding blind defenses? She set herself up for big trouble and the Q on the turn delivered it. It actually cleaned up her ace outs and turned her 9 into dirty outs. Plus it gave her correct calling odds in a bloated pot on a hand she shouldn't have been in in the first place.

  • monkeysystem

    @monkeysystem said:
    It actually cleaned up her ace outs and turned her 9 into dirty outs.

    Did I analyze this correctly?

  • think

    She had A9s (spades), so the only outs were really the FD or some kind of runner-runner draw.

    I hear you on reading her range from the open limp. I don't know, I guess I thought the FD was most likely by far.

    That bot is the most aggressive of the aggressive bots, in my experience. I don't even think my table image mattered, but in that session, I think I was trying to reel myself in to solid TAG poker with some position plays. Sometimes when I confront the bots, I'll be testing out some LAG tactics, so it might be a little (or a lot) crazier. But I don't think that was one of those times.

    But using the "Peek at End" feature, I would see who generally raises the stakes with weak holdings, and she's at the top of the list.

Sign In to comment.