When are mixed equilibria relevant?

Mixed Nash equilibria are a cornerstone of game theory, but their empirical relevance has always been controversial. We study in the laboratory two games whose unique NE is in completely mixed strategies; other treatments include the matching protocol (pairwise random vs population mean-matching), whether time is discrete or continuous, and whether players can specify mixtures or only pure strategies. Comparing point predictions, NE always does better than maximin and often does no worse than Logit QRE. NE predicts better than Center (50-50 mixes) under mean-matching, but otherwise not as well. By contrast, in a dominance solvable game, NE predicts better than alternatives in all treatments. Qualitative and quantitative dynamic models capture regularities across all treatments.