I think the term payoff is a bit clumsy. Perhaps it would be more exact to state that we need to compare our risk/reward ratio with our win/loss ratio.

The coverage of implied odds is the situation where the reward is not fully known - the opposite to that is when the risk is not fully known. Reverse implied odds. Let's say you have JsJc on a board of Td9d5c and someone bets into us. If we call any A, K and Q may give our opponent a better pair. Any K, Q, J, 8, 7, or 6 may give someone a straight and any diamond at all may give someone a flush. Even T, 9 and 5 may give someone trips.

If we're going to fold to all of these scary cards on the turn we would be wrong to call on the flop. That means we're either raising the flop or, in the context of this example, calling the flop bet and be willing to call a turn bet at a time when we may well be behind. This increases the risk component of our risk/reward ratio in a similar but opposite way to implied odds - called reverse implied odds.

Incidentally Spoon - I saw your post on the FR forum on bluffing (or ABCD ranges as I read it) and I think reverse implied odds can be a criterium in hand selection for your bluffing range. I think Fnord has an example along those lines. The conventional wisdom is to bluff hands with no showdown value and semi-bluff hands with implied odds going for them - and just check or call hands with mediocre showdown value. I think the next level is probably to bluff hands with mediocre showdown value because they are vulnerable to reverse implied odds and when the drawing hands do not improve they still work as bluff catchers.