On your 99 example, I think it's a bluff if your intention is to fold out some winning hands, like maybe JT. I think (?) it falls into the blocking bet category, even if every time he calls you lose, if the intention is to prevent him from betting bigger.
But maybe it will help to look more closely at the blocking bets. I feel like I'm completely clear as to percentages of success relative to bet and pot-size on VBs and bluffs, but totally in the dark on blockers.
So, first let's just see when the blocker is even +EV (which still doesn't mean it's max EV relative to a check, as excession points out with the induced bluff). The purpose seems clear: To keep the pot small with a hand that has some showdown value.
I'll take a stab at it this way: Your hand wins some percentage of the time p. Pot is 100, and villain will often bet 80, which is more than you like. So, you bet 50. We know for sure that if villain raises, we're beat. So, the probability that villain raises is <1-p, since he may sometimes call when we're beat.
Hmmmm... This would make it look like a subcategory of value bet, but I don't think it is. What happens on a check is this: He bets 80, and now you need to call 80 to win 180, so you're making a bad laydown if you can win at showdown a bit less than 1/3 of the time.
So, let's say you win at showdown about 1/3 of the time. By making the blocking bet, you lose only 2/3 of 50 rather than 2/3 of 80. Hence, it's a good bet.
I think that gives us the criteria for a good blocking bet:
1) Your hand wins at showdown somewhere in the 1/3-1/2 range (assuming villain isn't going to overbet).
2) To a check, villain will on average bet larger than your blocking bet. And, if there's any variation in bet size, you don't have a sufficient betting tell to say whether a big river bet is one of his winners or losers.
3) To a check, villain will bet very few if any of the losing hands with which he would have folded to the blocking bet.
More than anything, I think 3) is what makes or breaks the blocking bet. If he's going to bet just about any of the losers he would have folded, then you're better of check-calling, despite the larger bet size.
I do think criterion 1 is often given when you have been betting TP solid kicker and a draw hits on the river. You don't know whether you've been called down with the draw or with some kind of made hand, usually weaker. The drawiness of the flop is certainly a big factor.
Also worth noting is that, given your fairly low probability of winning, the blocking bet looks -EV relative to a check on first glance. But here, the check means that you're going to have to call an even bigger bet. So, EV becomes better than a check if all the criteria are met. That also means that a fairly detailed read of villain is required for a blocking bet to be a good blocking bet--specifically, villain likes to bet big on the river, but these bets are almost never air.
I guess one final thing: I picked 50 in the 100 pot rather than 30 or even 25 because if you bet 30, he can still make it 80. So, in theory you haven't really blocked his bet. That's not necessarily the case from a psychological standpoint. If you show some strength, he may be less inclined to raise you on his weaker holdings.Statistics: Posted by Aisthesis — Sun Feb 24, 2008 4:24 pm
]]>