Donkey Test

[KipBond]

Here is the best case scenario I think we are looking at:

Board: 4d 5h 7c
equity win tie pots won pots tied
Hand 0: 46.874% 45.93% 00.94% 26828 551.00 { 4h4s }
Hand 1: 53.126% 52.18% 00.94% 30480 551.00 { 88-55, 86s, 75s, 63s, 86o, 75o, 63o }

However, if our opponent is a good tight player (with non-"tricky aggression" on flops), I think this is more likely:

Board: 4d 5h 7c
equity win tie pots won pots tied
Hand 0: 30.566% 29.67% 00.89% 11163 336.00 { 4h4s }
Hand 1: 69.434% 68.54% 00.89% 25785 336.00 { 77, 55, 86s, 63s, 86o, 63o }

If we think he might be 3-betting loosely (contrary to his image), then this is a marginal push. We have $110 of our $1000 stack, and it will cost us the rest of it to find out if we are right. The more I think about it, the more a fold seems right here. A very tight, non-"tricky-aggressive" player, in the BB, bets out on a coordinated flop, meets a raise, then pushes in 40% of his stack. I'd guess he has the low straight 35% of the time, the high straight 35% of the time, a higher set 25% of the time, and a bluf/semi-bluff 5% of the time. Just a guess, I didn't crunch any #s to come up with those. Let me know if you think they are way off. [img]/images/graemlins/laugh.gif[/img]

[AKQJ10]

Well, even if we can categorically restrict his range to sets and straights, I think your estimates are substantially off. There are 16 card combinations for each straight, three for each set, and the starting hand is random. So if I know nothing else it's better than 5:1 he has the made straight instead of an overset. (Note how this is VERY different than the same calculation for a limping hand, where his starting hand range would likely not include 63.)

Granted, the set-or-straight assumption is probably close to right for this player. But 30% is also close to 33% or whatever odds you're getting on your push. I suppose there's even a chance that a timid player would be less likely to 3-bet a set because you might have 86, although I admit that's a stretch.

So in general I agree that it's profoundly unlikely that this player has anything but a straight or overset, but you still almost have the odds you need to draw out on that hand range, because there are so many more made straights than sets.

If you add in dubious combo draws like A6 then I think you're getting just enough to put the money in. The real question is whether to call or push, and I don't know the answer.

[KipBond]

[ QUOTE ]
Well, even if we can categorically restrict his range to sets and straights, I think your estimates are substantially off. There are 16 card combinations for each straight, three for each set, and the starting hand is random. So if I know nothing else it's better than 5:1 he has the made straight instead of an overset.

[/ QUOTE ]

Thanks for doing the math. [img]/images/graemlins/grin.gif[/img] Here's what I had said:

[ QUOTE ]
I'd guess he has the low straight 35% of the time, the high straight 35% of the time, a higher set 25% of the time, and a bluf/semi-bluff 5% of the time.

[/ QUOTE ]

Now revised:

I'd guess he has the low straight 40% of the time, the high straight 40% of the time, a higher set 15% of the time, and a bluf/semi-bluff 5% of the time.

[ QUOTE ]
Granted, the set-or-straight assumption is probably close to right for this player. But 30% is also close to 33% or whatever odds you're getting on your push.

[/ QUOTE ]

It's going to cost you $890 to win $1125. You aren't getting the right odds.

[ QUOTE ]
If you add in dubious combo draws like A6 then I think you're getting just enough to put the money in.

[/ QUOTE ]

Sure, if that is "reasonable aggression" for a tight player in the BB. I don't think it is.

[ QUOTE ]
The real question is whether to call or push, and I don't know the answer.

[/ QUOTE ]

It probably doesn't matter -- all the money will almost always be going in here if you call.

If you call, do ever plan on folding? Do you think if you call that your opponent might later fold a better hand? If both of those answers are "no", then a push is at least as good as calling, and may be better if there is any fold equity or chance your opponent has a worse hand than you right now.

[AKQJ10]

Sorry, you're right. I messed up the math by not considering being pot-committed. Since you're only getting about 11:9 then I agree you can safely fold here.

[Davdob]

I think isolating his range to just the oversets and the straights is a bit too tight even given the parameters of the problem. But if that is his range, I agree this is a fold.

But, I think we need to change the pokerstove analysis a bit to get a fairer picture.

First, I think the pre-flop betting substantially reduces the chance you are facing 77 or 55. I have argued those are raising hands against two limper, and that didn’t happen. To reflect this, I would only include 77 in the pokerstove analysis.

Second, I think KK+ is a plausible, but less likely than just statistically possible explanation for the post flop bet. In a NL ring game, its certainly plausible he would just check in order to get more value out of these monsters pre-flop. Post flop, he made a small raise into a weakish looking flop, we raised, but not imposingly so. He is holding AA, maybe we are fooling around with something like A7 or a lower pair, and he is telling us we better get lost. Its not 100% our raise shows a straight or a hit set. Although, I concede, that he is paying a lot to know for sure, even tight players have trouble believing their AA is cracked. To reduce the possibility, and probably to more accurately reflect what he might have, I would still include AA in the mix.

We have to include all the straight possibilities for sure – 86 and 63.

For reasons I stated before I would also include 75.

With those assumptions, pokerstove comes up with a ~48% chance to win.

Still a -40ev play, but better than the -140ev from folding. (Or perhaps more accurately, you would be paying 860 for a shot to win 960, which is a plus EV move). Also, note if you include 55 as a possibility here the EV drops to 46.5, but push is still +ev.

The main criteria for deciding this question is whether or not you believe he could ever be playing high pairs or AA (and assume he never ever bluffs). If you believe he could possibly make the three bet with those holdings you push. If not, fold.

[TasteThePainbow]

I think you've read reasonable in his agression on the flop to mean: he is only agressive on the flop when he is confident he has the best hand. It's reasonable to raise with a fairly wide range here.

Before we analyze his range, lets consider from his perspective the range that we are raising with here.

I think its reasonable for us to semibluff any paired 6 and we semibluff a gutshot and a pair fairly often say 33 25% of the time and 88 75%, so 88, 67, 66, 65, 64, 33 are definitely in our range (although most of the time 80% we fold these preflop, so I'll discount these), we raise a set, two pair and even TPTK (44, 55, 77, A7, K7 (rarely), 54s, 57s, 74s (rarely)) to protect the hand, we raise an overpair as a semi-bluff and a raise for information (we can counterfeit his two pair, get him to fold some better pocket pairs and find out where we stand (but we will often have raised these before the flop so we should discount accordingly), we might raise bottom straight but we're really unlikely to hold 36 in the first place. With top straight we call sometimes and then make a valuebet with position makes sense to me. All in all I think we are raising with a straight or set not all that often here (20 of the 80 hands I have in our range)

Here's the range for our raise I came up with -- I've used suited cards as a placeholder for the idea that we're folding these hands before the flop at least 75% of the time.

99-44,3c3d,3c3h,A7s,9c6c,8c7c,8d7d,8c6c,8d6d,8h6h,75s +,7c4c,6c5c,64s,6c3c,6d3d,5c4c,Ad7c,Ah7c,As7c,Kd7c ,Kh7c

Basically, we will rarely have a straight and we will sometimes have a set and we will raise with quite a few hands that are scared by a raise (by my count we have 80 hands in our range 20 are a set or a straight, 30 set straight or two pair) I think we fold anything else pretty easily to a big raise). So he can expect to win $105 by raising (5/8)*165 and imagine he loses 2/3 of the hands when we call [either by folding to a later bet or by checking down to the river) -(3/8)*(-370)*(2/3) = -95.

The point is -- he can make this move and he only needs to have a better set or a straight 1/3 of the time in order to justify it.

I don't think this is very close.

A more interesting question is whether to raise or shove... I wish someone would write about that.

[Davdob]

Well, I dont think his range is as broad as that, but it doesnt have to be for push to be the right play. You basically just have to assume he would 3 bet AA and you are most of the way there.

As for raise or all in. What raise could be credible? If you go all in he still has 2:1 pot odds. He is going to call you with anything almost at that point. If he has an overset or the straight, he almost for sure is just going to instantly re-raise you all in, and you are going to call for the same reason he would call your all in bet. If he is working on a draw, you are just giving him another card at a reduced price.

I dont see any advantage to any raise other than all in.

[KipBond]

The previous analysis from you two would work if a) you were playing live $1/$2 NL (maybe $2/$5), or b) your opponent should be playing $1/$2 NL. Not raising with AA/KK in the BB? And now with a flop like that he wants to pot commit himself OOP? Horrible.

[Fiepoto]

So, I think this one is a clear fold.

Key information here is that the villain is tight/passive

I don't see the tight/passive player raising on the turn, without at least the K or pocket 22. Because the player is tight, the only thing we beat is KJ or maybe K 10 (but then, that's not what I would call tight). I really think we are up against KQ or pocket 22 here. He is obviously not pushing all in on the river without at least 3 kings, so at best we are hoping for a split AK.

But he's got a boat. Fold.

[Fiepoto]

10/20 NL. Approx $2,200 stacks. UTG limps. You are in middle/late position with:
A[img]/images/graemlins/diamond.gif[/img]Q[img]/images/graemlins/spade.gif[/img]

You raise to $80. The button calls and the BB calls. UTG calls. Flop is:
A[img]/images/graemlins/heart.gif[/img]10[img]/images/graemlins/heart.gif[/img]Q[img]/images/graemlins/diamond.gif[/img]

UTG leads out for $200 into the $330 pot. You raise to $600. The others fold and the UTG player moves all-in for $1,900 total.

The all-in player limps too much pre-flop but is not out of line with raising pre-flop, is generally passive post flop, and seems to be an inexperienced player.