When you get caught bluffing, you of course lose. However, you may not mind being caught and losing early in a session because you are considering your image for future hands. You may even make an ill advised bluff early so that you will get a lot more calls on your legitimate hands the rest of the night. (Similarly, an early ill-advised call against tough players may keep them from bluffing against you the rest of the night because they fear you're likely to call their bluffs.)

Creating an image that you almost never bluff can also be advantageous. I am generally considered a tight player, and I sometimes pass up an early, marginally profitable bluffing situation to enhance this image. What that does is allow me to steal some pots in the future with complete impunity. No one imagines I am daring to bluff.

When you are up against even average players, they are constantly studying the way you play. So considering the effect of any play on future hands should be an important part of your game, especially in no-limit and pot-limit poker and especially when you are playing against the same people all night or from one night to the next or one week to the next.

Some players go so far as to argue that bluffs should show a loss because those losses will be repaid with interest when they get a lot of action on their legitimate hands. Game theory, suggests that when you employ optimal bluffing strategy, you should break even on your bluffs. However, there is no reason not to develop a sense of your opponents and of betting situations so your bluffs show a profit. A successful bluff wins the whole pot, and it takes a lot of extra calls of your legitimate hands to make up for one pot. Therefore, against all but very tight players, you should bluff slightly less than optimally so your bluffs show a profit. The greater your reputation as a tight player, the more you will be able to get away with bluffs. At the same time, you will still get caught often enough to get paid off when you do have a good hand.