Bluffing Rob, a reconsideration

I've been thinking quite a bit more about that hand I played with Rob at Planet Hollywood, written up here. He describes it from his perspective here. Specifically, I've been going over what I would have done if I were in Rob's seat, playing his cards, against an opponent like me. My thought process got churning because of a series of comments posted to Rob's blog.

First, a reader calling himself "ohcowboy12go" said:

Rob, I think your better play would have been to take advantage of your position and use some pot control by checking the turn. This way you can turn your hand into a bluff catcher on the river or bet for value if they check again.

Rob's response was basically the same as mine would have been:

Well, ohcowboy, I never really considered that. With so many draws out there, giving a free card seemed way too dangerous. Of course, if he already had made his draw, that wouldn't matter, but, I thought protecting my hand was the way to go.

But then along came Vookenmeister with a more complete argument for the same point:

it's actually a valid suggestion by ohcowboy. Here's why!  

If you're planning on folding to a check raise knowing that Grump is highly capable of making this play with a wide range of hands that include a lot of semi-bluffs then basically you are making your own hand meaningless. 

Thus your bet now becomes a bluff if you are willing to fold in a spot where grump is often going to check/raise his draws. 

Against this particular crafty opponent it might be better for you to outcraft him. Go for pot control on the turn here (which makes your hand look like AK/AQ) and be prepared to call almost any river bet. 

Sure you might give him some free river cards to his outs but if he is so likely to bluff then it's worth it. You can also bet the river for value if he checks. 

Against clever opponents you need to mix up your plays and go another level. don't let grump put you on a polarized range of hands. 

Don't be predictable against him. Of course make sure you keep these plays to those spots where you are mostly heads up against him. Other players will not comply. 

In other words, go a level up. Think about what he thinks you might have based on your play and take a step higher. If you do this every once in a while then Grump will not be able to read you as easily and make plays like this. 

cat and mouse. 

PS. I look forward to a blog post where you try this and it fails and you curse my name. There are no rebates on my silly advice. 

I have read Vook's thoughts on his own blogs (here and here), and when commenting on other people's blogs. I have also met and played in a cash game with him. Through all these means, I have come to respect him as an analyst of tough situations.

And you know what? I think he's right. On a board where there can already be completed straights and flushes, unimproved pocket kings really is not much more than a bluff-catcher, so you might as well treat it that way.

Suppose Rob had checked back on the turn. I would still be unsure whether he had an overpair that feared the straight and flush, or an A-K type of hand that had missed completely. Assuming that the river card is not a fourth diamond and not an ace or king, I would indeed be very likely to bet. It would be some odd mixture of a bluff (hoping he would fold an overpair if that's what he had) and a value bet (hoping he would hero-call with ace-high, which my pair of 6s would beat).

How much would I bet? Probably right around $40, the same amount that Rob bet on the turn. So for the same $40 he bet and lost on the turn, he would instead get a showdown in a situation where he well might win, if my river bet is a bluff or a value bet with a pair lower than his kings.

Suppose a diamond came on the river. What then? Would I bet? I think I probably would. A $40 bet to win what's in the pot is well worth a shot, because (A) there's only about a 50/50 chance that Rob has a diamond in his hand, and without one he will never call, and (B) if he has any diamond other than the ace, he still might fold, fearing that I have the ace.

If an ace or king had come on the river, I think I would be much less prone to bet, for two reasons. First, any component of value bet is now gone, and what would have been a semi-bluff on the turn is now a pure bluff, which is harder to make profitable. Second, I would have to worry that Rob had just made a set. He would then be much less likely to fold, because it would remove all the two-pair and sets from my range of hands that he would have to worry about. If he only has to worry about being beat by a straight or flush, he'll be far more likely to call than if he has to worry about everything I might have that can beat one pair. Therefore, a bluff is much less likely to succeed.

The obvious objection to this line is the one Rob expressed--checking behind on the turn may give a free card that completes a draw to beat him, when a bet on the turn might win the pot. That's true, and David Sklansky says that in mathematical terms one of the worst mistakes you can make in poker is failing to bet when doing so would win the pot.

But weighted against that consideration is this one: You want to play big pots with big hands, and an overpair simply is not a big hand when the board is as scary as the one we're talking about. It took me a while to come around to Vook's point of view, but I have. I think the balance comes down on the side of playing conservatively and defensively rather than aggressively in this specific situation. A $40 investment is more likely to win the pot as a bluff-catching call on the river than as a bet on the turn, given the specifics of this board and this opponent (i.e., me), and it doesn't risk getting you pushed off of the winning hand.

Very early in learning about poker, I heard that when you bet or raise, you should always know in advance what you will do if an opponent raises back. You don't necessarily have to have a plan for what you'll do in case he calls, because you'll get another card and can re-evaluate then. And obviously you don't have to plan what you'll do if he folds, because the hand ends. But you should know what you'll do if he raises--whether it will be to fold, call, or reraise.

I doubt that any of us manages to do this every time, but I think it's a worthwhile aspiration. From Rob's reaction to my check-raise, it was clear that he had not thought through what he would do if I raised. (I don't say that as a criticism; I'm as guilty of stumbling into such "now what do I do?" moments as he is.) If he had, he might have realized that it was a situation in which a check-raise from me would put him to a very difficult decision, since my range would be an indecipherable mix of made hands (straights, flushes, and sets), semi-bluffs, and pure bluffs. Given that, he might have then realized that $40 or so would be better invested on an easy, automatic river call--no matter what came--than on a turn bet that could lead to a costly mistake.

For the last 24 hours or so, when my mind has drifted back to the question, I've asked myself whether I would have made what I now think is the right play (i.e., checking) had I been in Rob's shoes. And I honestly don't know. I think maybe I would have on my best, most analytical, A-game days. But I spend a lot of time playing on what amounts to auto-pilot, and the auto-pilot mode says to keep betting a hand like pocket kings--especially in position--until something grabs me by the throat and makes me stop. So I might well have done exactly what Rob did, and ended up in the agonizing no-man's-land that he did, unsure of where I was or what I should do.

Anyway, the more I've thought about it, the more I think it's an interesting hand, full of potential for learning and debate.