In this article, we're going to clear up a few misconceptions and tie up a few loose ends. Below is a further discussion of advanced principles of bluffing, especially with respect to optimal bluffing.
If you have read Bluffing 101, the following question should now be answerable.
Q: Assume that you're playing limit holdem and that you're on the river. You've been checking and calling. Assume that when you draw to a flush, you tend to check-and-call and bet out when you hit the flush. A flush card arrives, but you don't have that, you have some other hand that is normally not worth betting (perhaps you have a busted straight draw or were calling down with a weak kicker). If you want to camoflauge your made flushes, how often should you bluff in this other situation?
A: The bluff-to-non-bluff ratio here 6-to-1 because you're offering 6 to 1 odds to your opponent on a call when you make the bet. This means that for every six times you make your flush and bet out as a non-bluff, you should bet out with a bluff once in the same situation.
This particular question and answer should help clarify things. Many of your betting patterns are NOT bluffs. However, all situations like this one may be associated with a betting pattern that usually indicates you are not bluffing, but where you should have a counterpart bluff in your arsenal.
Now, is it true that you should really just religiously stick to bluffing at the ratio game theory dictates? The answer is no. There are some conceivable situations in which your opponents are set to call and it doesn't matter that you've NEVER shown a bluff, they are still going to call. In that case, your counterstrategy is to deviate from the equilibrium point and set your bluffing level to zero.
This is why players are advised to play "straightforwardly" (i.e. honestly without bluffing) at lower levels. It's because you can count on the calls and you'll earn more by taking advantage of them than by trying to bluff.
Another situation that comes up is when you're playing online at a place like Party Poker and you're almost never at the same table with people who know you or people who've even had enough time to log your tendencies with Poker Tracker. Any time that you're up against a field of unknowns, it seems that you can count on calls.
At least for a while.
My advice in the past (as given to a friend) has been "Don't bluff during the first two orbits of a session." Of course, I wouldn't really follow this advice to the letter, because sometimes I like to do the opposite.
The real time to turn on the bluffing is against players who seem to think through their decisions and seem at least capable of folding. The equilibrium point is great against players who will sometimes fold and sometimes call rather than against players who tend heavily toward one or the other.
It's also good to use bluffing of this sort when playing against the same old players day in and day out. If you don't do it, they'll know your every move. If you do do it, they can't ever be completely sure what you have. Especially if you eliminate your tells by sticking to making the same actions in the same way in all situations.
A point that comes up now is this: What's a good way to bluff at the proper ratio, but also keep the event of a bluff random? (You could just count off six non-bluffs and throw in a bluff every seventh time, but that would be deterministic.)
The answer is that you should pay attention to those times when you are NOT bluffing. Look at the pot size and determine the bluff ratio. In this case it was 1 to 6. Select a random event that is known to occur 1 out of 7 times and IF it occurs during the non-bluff, you can post a mental sticky note that you should bluff the next time that same situation with those same odds comes up.
One of the easiest things to do is just use a clock. For this example, if the second hand is in the first one seventh of the clock face during a non-bluff, turn the bluff ON for the next time the same situation arises but you don't have a hand. I don't usually use this particular method of randomizing, but it is highly effective.
My particular experience with the high-limit players at Foxwoods and in AC is that a lot of this is beyond most of them, although perhaps not all of them. It's certainly not beyond David Sklansky, but no one believes that even he religiously applies it.
Trust me. There's plenty of room to be the most knowledgeable player at the table.
The real trick at that level is learning how to throw chips around that are worth $100 or even $1000 each and still view them as just chips.
