Tuesday, January 17, 2006

You Make the Call

Scenario: You're in the BB in a shorthanded limit game. The button is a complete and utter maniac who will raise every time when it's folded to him, regardless of the cards he holds. The SB is very timid and folds nearly everything when facing a raise from the button. Post-flop, the button plays very poorly and largely relies on naked aggression. That said, he'll fold his cards like a hot potato when played back at and he's bluffing with nothing.

Theory: When everyone folds to the button (who naturally raises) and the SB folds, you should never, ever three-bet. Ever. Not even with AA. Simply call with any hand you intend to play, regardless of its relative strength.

Thoughts?

(Edited to add: I'm just posing the above question. I neither advocate nor deny the theory in question.)

15 comments:

d said...

I agree with the theory in the short run, with the caveat that you must pay attention to see if maniac adjusts to your strategy.

However, I think you are missing out on defining a very important qualifier. What percentage of hands will you be playing? Maniac is playing 100%.

Sean said...

I disagree.

The situation is essentially the same as heads-up play against an opponent who always raises the button (albeit an extra small blind in the pot). I'm sure there is a lot out there on HU strategies (although I don't have any links). Negreanu faced this type of player (never three-betting pre-flop) when he played DreamClown and I believe he mentions a bit of strategy in his blog.

I'm no expert, but I have played some heads-up against these types of players before.

I agree that certain hands (like big pairs) should not be three-bet. You'll be seeing the showdown with those hands unless the board is very scary anyway, so you might as well deny him the read and use his aggression against him. To compensate, other middling hands that are playable should be called to.

Certain hands (big unpaired cards, middle to low pocket pairs, high suited connectors) should be raised for value. The third bet is both a value play (you most likely have the best hand) and a way to take the lead since you'll be playing out of position. Furthermore, it helps define his hand to determine if he outflopped you. Raising big unpaired cards will allow you to take the pot when when you both miss but he has very live draws (which would be difficult to read when they hit).

The key is that he is capable of folding -- the three-bet will get more money in the pot when you are a favorite and set up bluffing situations later in the hand.

ScurvyDog said...

Dave,

Is your answer different if you're playing 30% of total hands as opposed to 60% of total hands?

Absinthe said...

What's the maniac's preflop tendency when he actually does have a decent hand (say top 10-15%)? Will he ever fold to a threebet (unlikely) or cap with a premium hand preflop? And is his postflop line predictable (i.e. autobets when checked to, folds when reraised except when holding a piece of the flop, jams with top pair or better or decent draw)?

I think so far my response to your theory is a solid "it depends". If the maniac is crafty enough to notice that you are only threebetting with a decent hand preflop, then you are giving up a lot of information when you threebet. If you threebet more often (say any hand Q7o or better) you're probably pushing a very thin theoretical equity edge when you're at a positional disadvantage, but you're giving up less information against an opponent whose postflop skills you clearly have no respect for.

That said, the turn and the river are where the big bets are and if his line is predictable all the way to the river you might be justified in giving up some premium-hand edge since you know you'll be getting the best of it on value bets.

Alan said...

It depends.

I don't know how to do the math to explain the decision (and thus I don't know what the correct decision is).

Basically, you are going to end up with the best hand 50% of the time, but you can't be sure when those times will be so you end up folding more than 50% of the time.

I think the correct answer is to raise at a percentage so that you win back the blinds he steals from you. If he is raising 4x, then reraise him with top 80% of cards.

When you hit a monster where you are a huge favorite to win, let him bet as much as he wants into you.

But them I'm not really sure about any of this.

ScurvyDog said...

Sean,

Doesn't calling with big pairs and raising with big unpaired cards, middle to low pocket pairs, high suited connectors give your opponent an awful lot of information?

If you feel you can outplay him and that he's too poor of a player to exploit the above betting patterns, shouldn't you simply three-bet with any hand that you decide to play, thus revealing nothing about your holding and still getting in the maximum amount of money?

ScurvyDog said...

Absinthe and Alan,

You both chose the cheater answer of "It depends". Robot monkeys have been dispersed to track you down and flay the flesh from your respective hides.

(Yes, of course, that's the right answer, given the complete lack of further details.)

WillWonka said...

Of course, I agree that it depends; but I'm not in the camp of getting too tricky. I would always 3 bet with a premium hand... as well as any real good hand. If I play it the same everytime, then how does he know what I have?

I've always scoffed at the idea of the change things around premise. Sooner or later, your changing is around isn't random and someone could potentially figure something out.

What's the worse that could happen? He folds? Better that than him getting a relatively cheap flop.

Get your money in preflop and play accordingly postflop. That is where the money is made.

ScurvyDog said...

Will,

Since he's playing any two cards, you have no idea what his range is. Pre-flop, you've got a lot of incomplete information.

You're also much better than your opponent. This advantage is magnified post-flop, when you have more complete information and can charge more on later expensive streets, especially against an opponent that will auto-bet into you when first to act or checked to.

Given that, can't an argument be made that you'd maximize your overall EV in this situation by getting less than maximum money in the pot pre-flop (i.e. just calling, never three-betting), when you have less complete information?

Egarim said...

I'd test the maniac out by 3betting preflop with a big pair and seeing how he plays after that. If he plays scared and gives you a lot of credit then start throwing in a larger range of hands. Maybe if he's giving you too much credit 3bet with a rag hand and show it to him if you pick up the pot. Since, he's a maniac that'll definitely get him steaming and give less credit to your 3bet.

The other alternative is to 3bet about 1/3-1/2 of the time with your big pairs and about 1/3 of the time with your high cards and mid pp's. Also, 3bet occasionally with suited connectors. If you take the lead you'll win more out of position. Of course, this depends on his reaction to your 3bets. Will he fold on the flop to your automatic lead out when he misses the flop?

SeanSkill said...

You are going to be playing this guy heads up in a lot of pots. You have the choice whether to play him in small pots or large ones. If you are a better player post flop, I would choose to play him in big pots. I would 3bet with any two cards that I would normally open with. consider it a 3 small bet ante.

ScurvyDog said...

SeanSkill,

Do you more often create big pots by three-betting pre-flop or by three-betting on the turn/river?

Could it be possible that not three-betting preflop from the BB might actually increase the overall average pot size that you win?

Or that if you're a superior player that you'd rather wait for the bulk of the money to go in until after you see the flop, at which point you have much more complete information and can extract full value from your relative advantage?

Alan said...

I really like the idea of making all of the pots big and out-playing him after the flop.

But then that's assuming I could out play him after the flop, which is doubtful at this point.

Sean said...

Scurvydog said:

"Doesn't calling with big pairs and raising with big unpaired cards, middle to low pocket pairs, high suited connectors give your opponent an awful lot of information?"

It does seem to give away a lot of information (i.e. if I raise I don't have a big pair). Obviously, in real life, you mix it up when you feel them adjusting.

But, it is still relatively balanced. For instance, if I call pre-flop I either have a relatively weak hand or a really strong hand. If I raise, I have better than an average hand with probably larger cards than usual, but not always. Ideally this strategy would be limp 50% and raise 50% or thereabouts.

I'm no expert, but against a maniac, I like to play relatively tight post-flop. The three-bet with the harder to play hands allows me to take the lead when I'm out of position, giving me more information from his raises and bets (since I've already shown strength).

SeanSkill said...

I equate this situation to playing someone heads up who always raises their button, you are giving up to much equity preflop by not reraising your top 20% or so starting hands, its slowplaying which really isnt neccesary against this type of opponent.

If I were the button facing a very laggy BB who was three betting a lot I would prefer to call and get some extra bets out of him on the big streets.