Probability matching
Probability matching is a suboptimal decision strategy in which predictions of class membership are proportional to the class base rates. Thus, if in the training set positive examples are observed 60% of the time, and negative examples are observed 40% of the time, the observer using a probability-matching strategy will predict (for unlabeled examples) a class label of "positive" on 60% of instances, and predict a class label of "negative" on 40% of instances.
The optimal Bayesian decision strategy (maximizing percent correct prediction, see Duda, Hart & Stork (2001) ) in such a case is to always predict "positive" (i.e., the majority category). The probability-matching strategy is of psychological interest because it is a strategy found to be frequently employed by human subjects in decision and classification studies.