[NL2-NL10] Unpaired AK vs. drawing hand at NL $10 (FR), how much to raise?

    • howard182
      howard182
      Bronze
      Joined: 30.10.2006 Posts: 416
      Party Poker No-Limit Hold'em, $0.10 BB (9 handed) Hand History Converter Tool from FlopTurnRiver.com (Format: FlopTurnRiver)

      CO ($10)
      Hero ($2)
      SB ($1.80)
      BB ($13.62)
      UTG ($3.30)
      UTG+1 ($18.27)
      MP1 ($9.40)
      MP2 ($17.01)
      MP3 ($1.75)

      Preflop: Hero is Button with K:heart: , A:club: . CO posts a blind of $0.10.
      5 folds, CO (poster) checks, Hero raises to $0.5, 2 folds, CO calls $0.40.

      Flop: ($1.15) 9:heart: , 3:club: , Q:heart: (2 players)
      CO bets $0.35

      At this point I am practically certain that my opponent has a drawing hand and no pair (I can't specifically pin down the reasons beyond this bet, but I was sure of it), which actually put me slightly to significantly ahead according to my estimate at the time. Raising seemed the only sensible option therefore, but how much?

      A raise to $1 gives him attractive enough odds that he might call or re-raise, but this is enough of my stack that we would both be committed as I see it. The short stack strategy articles suggest going all-in in this case, but I think he would certainly have folded if I did that; a single pair would be a great threat to him. Of course this isn't a bad outcome, but I thought I could do better.

      So it's a fairly certain +$1.40 / rather unlikely -$1.5 versus a likely +$2.65 / less likely -$1.5. Looking at those figures now, all-in seems more profitable but the $1 still has appeal. The rest of the hand is below (in white after my next action) including results. What would have been the correct action here and why?

      Hero raises to $1, CO raises to $1.30, Hero calls $0.15 (All-in).

      Turn: ($4.10) Qd (2 players)

      River: ($4.10) Qs (2 players)

      Final Pot: $4.10

      CO has Jd Ts (three of a kind, queens). (So my read was right, more or less; I thought he was drawing for the flush.)
      Hero has Kh Ac (three of a kind, queens).
      Outcome: Hero wins $3.95.
  • 3 replies
    • Korn
      Korn
      Bronze
      Joined: 14.01.2005 Posts: 12,512
      Let's assume he has a draw 100% of the time


      Condition 1:
      The amount you raise should be at least enough that if he chooses not to fold, he will have a negative expected value. Or quickly said: don't give him the odds such that he can call or reraise profitably.


      Condition 2:

      You want to raise more than mentioned in condition 1. You would like to maximise the expected mistake he is going to make. Say - for the sake of the argument - if you raise $1 his EV when calling will be exactly 0, i.e. the break even point for the call. What you want to do now is to maximise the following number:

      (Probabilitythat he calls your raise) * (Amount of money above the break even point)

      In that way, you will maximise your own expectation.

      Common sense conclusions: the more fishy or loose your opponent is, the more you should raise as he is likely going to make some horrible calls with nice looking draws.

      Now my suggestion:

      His bet is 0.35, the pot is 1.15. If you just call, the pot would be 1.85.

      If you raise the pot on top of that (i.e. a total raise of 2.2), you'd give him 1 to 2 Odds or 33%. That is still too good here, as a flush draw or straight draw gets there in about 33-35% of the time and also - on top of that - we can win by just pairing his cards. He might have many more outs than that, in fact, he might have up to 15 outs against you.

      So if you raise here, you need to raise big. I'd raise at least $3 (or in fact if your stack is $6 or less, I'd go all-in (if your stack is any bigger, I would really worry that my read that he has a draw is not correct and still fold)

      In your case, going all-in is a no brainer.

      However, having said that: We assumed that your read was 100% correct, which is never the case. Still, with just $2 in your stack, it's rather low risk, so if you trust your read, going all-in is the best way to play it.
    • howard182
      howard182
      Bronze
      Joined: 30.10.2006 Posts: 416
      Ah, I see why my reasoning is flawed now; even going all in I can't afford the level of protection I need in this case, so I should get the best I can or none at all, right?

      (And I was absolutely - well, 95% or so - certain of my read in this case. I'm certainly not arrogant enough to think that I know what my opponent(s) is/are holding most or even much of the time, but very occasionally I am sure.)
    • Korn
      Korn
      Bronze
      Joined: 14.01.2005 Posts: 12,512
      Originally posted by howard182
      Ah, I see why my reasoning is flawed now; even going all in I can't afford the level of protection I need in this case, so I should get the best I can or none at all, right?
      Yes, that's it :) But none at all would be by far your worst option here, given your read.