Bluffing and Betting
The number of online poker games hands anyone can have is comparatively limited, but in addition to the hands themselves there are so many other variables that rarely if ever is a particular play always right or always wrong. Your play is affected by the size of the pot, your position, the opponent or opponents you are facing, the way they have been playing, the amount of money they have and you have, the flow of the games, and other, more subtle factors. This point is particularly applicable to questions of bluffing and betting fair hands for value on the end. Here are some general principles that usually apply.

When you bluff, you are rooting for your opponent to fold because that is the only way you can win the pot. When you bet for value, you are rooting for your opponent to call because you want your legitimate hand to win one more bet from him. It is important to realize that it may be right to bet a fair hand for value, and it may also be right to bluff, but it is almost never right to do neither. If you decide you can't get away with a bluff on the end when you miss your hand, then you should bet for value when you do make your hand. (The only exception to this principle would occur in games like hold 'em and five-card stud, where your opponent can see your last card and might often have a good sense of whether it made your hand. In those cases, if you bet a hand for value, you are likely to get called - or raised - only by a hand that has you beat.)

Similarly, when you don't think a value bet is justified with a fair hand, since your opponent will only call if he has you beat, then if you miss your hand, you should usually bluff. For when you bluff, it is possible your opponent will throw away his fair hands.

 [ 1 ][ 2 ][ 3 ]
A 2 Percent Approximation

University of Texas math professor (and poker player) Charles Friedman suggests a simple method to approximate the chances of making a draw. Multiply the number of outs you have times the number of remaining cards to come, then multiply that product by 2 percent. The result is a slight overestimate of the probability of making the draw.
It's a good approximation if the product of the number of outs and the number of remaining cards is less than about 20 cards.