[AKQJ10]

The important point isn't the precise calculations, because it's always going to be a judgment call whether opponent is 10% or 15% or 20% to hold a set when she raises the turn, etc. While it's good practice to analyze hands based on such estimates as a post mortem, you can't expect to do that math at the table even if you could estimate the probability of each opponent's hand with pinpoint accuracy. IMO the point of the hidden outs chapter is to awaken the reader to the fact that these outs do exist.

Real world example of the top of my head: you pick up AA, and opponent has been playing passively as you bet or raised the flop. But suddenly, on the turn, opponent raises. (I believe this is similar to Ed's first example, don't have it handy though.)

Depending on the board you may be more or less scared that opponent has made two pair or a set. But whatever your course of action, you need to keep in mind that you have an additional 6 outs to beat a lesser two pair (in addition to the aces). Before I read Ed's book, I didn't always take these hidden outs into consideration. They're not full outs because you may be drawing thin to a set or better, but they're important to estimating whether to fold, call, or raise.

If you figure you have only two outs to improve AA in the situation described, you're underestimating your hand and will fold too often.