Ingredient 1 – Improvable EquityImprovable equity means having a non-made hand that has equity due to what it might become on future streets. If we are considering bluffing the turn, and we estimate an unknown or average amount of likely fold equity, then the non-made hands which can improve to something nutted a relevant portion of the time will make much higher EV bluffs than those which are drawing close to dead when called.
For example, let's say that we have c-bet the flop on a board of T835 with no flush draw present and villain has called our flop bet. On the turn, we are better off firing a second barrel with J9 than with A6. The former has the potential to beat top pair at showdown on eleven different river cards. The latter has only three outs for this purpose and can never cooler a hand better than one pair. Whatever our fold equity here might be, there will be occasions when we get called, and so our improvability is directly linked to our EV. Winning a large pot when we improve to a straight more than compensates for a slight deficit in immediate fold equity. A two-thirds pot sized turn bluff, for example, needs to work 40% of the time to break even if we had no equity whatsoever. The A6 gets a slight discount on this due to its three outs and can perhaps survive with 38% fold equity. The J9, however, gets an enormous discount and needs almost no fold equity to be a +EV bluff. The upshot is that, even against seemingly call-happy opponents, J9 is still likely to be a profitable turn semi-bluff.
The only time that improvability has no effect on a bluff is on the river when there is no longer any possible improvement to be made.
Ingredient 2 – Lack of Showdown ValueBluffing with too much showdown value is a mistake I witness very often in students at the beginning of their poker journey. The reason that this type of error is so serious is that it completely defeats the point of bluffing. In general, we do not gain a lot of EV by making Villain fold worse hands than ours. Sometimes, on the flop or turn, we gain some EV when he folds two overcards to our pair, surrendering his chance to hit one of six outs, but this is a relatively small benefit and betting for this purpose is a protection bet, not a bluff.
Villain checks a flop of AQ4r in the BB after calling Hero's BU open. Hero holds JJ and checks behind his middle pair as there is no real attraction to betting. On the turn, a harmless looking offsuit 7 rolls off and Villain checks again. Now the student will often mistakenly reason:
"I decided to take a stab because he showed weakness twice."
I will then ask what exactly the purpose of this so called 'stab' is supposed to be.
"To get him off the hand."
The student might reply, to which I will ask what exactly we are making our opponent fold.
"Weak hands, underpairs, air, king-high"
I will be correctly told, as there is certainly very little prospect of Villain folding either a Queen or an Ace. Meanwhile, if he happens to hold a one of these hands, we have just lost a bet we did not need to make. Herein lies the problem; we fold out only hands that are drawing to two or even less outs. Only Kx or a gutshot has as many as four outs against us and this is barely worth protecting against. "Stabbing because he showed weakness" is a terrible insufficient reason to bet, which can soon be reduced to absurdity once we have a clear definition of what a good bluff is:
A good bluff is one that causes the opponent to fold enough better hands for the amount risked.
Ingredient 3 – Positive RemovalSometimes we just don't know how likely Villain is to fold because we're playing against some unknown human halfway across the world. In this case, we might rely on how much equity we have to improve our non-made hand when selecting which hands to use as bluffs. The problem is that sometimes, none of the unmade hands in our range have any real chance of improving. Take a dry board such as K83r SB vs. BB. As the pre-flop SB raiser we c-bet the flop and Villain calls. The turn brings an offsuit 3 and we ponder our options. Which parts of our range make the best bluffs now that we cannot possibly hold a draw? Should we just give up all of our air and bet only with value hands?
Certainly not: to become fit or fold simply because you cannot hold a draw to a flush or a straight is ludicrous. Villain could simply fold all of his bluff catchers and exploit us by never paying us off. In order to create some balance and make our ranges tougher to play against, we must find some bluffs and with little else to separate trash from trash within our range, this is where positive removal comes in.
Firstly, 'removal' means holding cards that stop Villain from holding certain relevant hands. When we hold cards that Villain could use to form some strong hands we say that we 'block' some of his good holdings. This is known as 'positive removal' as it benefits us. In the case of bluffing, it helps us have higher average fold equity. That's right, we can increase our own fold equity via bluff selection, even when we know nothing about our opponent's tendencies.
'Negative removal' would occur when we remove something from the deck that could give our opponent some bad hands, for example, having the ace of a busted flush draw on the river.
Okay, now back to the K833 board. Having c-bet the flop and been called, we should fire the turn sometimes, but mainly with positive removal. We want to eliminate some of the combinations of our opponent's potential good top pairs, which are almost never going to fold. Therefore, it is a better idea to bluff in this spot with QJ, which eliminates KQ and KJ (positive removal) than it is to bluff with 76, which removes weak underpairs that might well fold (negative removal). Having no equity and no read does not force us to play in a face-up manner, contrary to popular novice opinion.