We study search games between a mobile Searcher and an immobile Hider in which the Searcher aims to minimize some payoff, which is either the time to find the Hider (the search time), or a normalized search time. We consider a new setting in which the Searcher has some potentially erroneous information, or prediction on the Hider's position. Specifically, we study tradeoffs between the consistency of a search strategy (i.e., its worst case expected payoff assuming the prediction is correct) and the robustness (i.e., the worst case expected payoff assuming that the prediction is adversarially generated). We show how to apply this framework in search games over both discrete and continuous, as well as bounded and unbounded spaces. Specifically, we prove optimal consistency/robustness tradeoffs for three fundamental search games, namely searching in a number of discrete locations, expanding search in a tree network, and searching in the infinite line. Our study is the first to address the full power of mixed (randomized) strategies; previous work focused only on deterministic strategies, or relied on stochastic assumptions that do not guarantee worst-case robustness in adversarial situations.