This article reports two studies on the accuracy of flirting detection. In Study 1, 52 pairs (n = 104) of opposite-sex heterosexual strangers interacted for 10 to 12 minutes, then self-reported flirting and perceived partner flirting. The results indicated that interactions where flirting did not occur were more accurately perceived than interactions where flirting occurred. In Study 2, twenty-six 1-minute video clips drawn from Study 1 were randomly assigned to one of eight experimental conditions that varied flirting base rate and the traditional sexual script. Participant observers (n = 261) attempted to determine if flirting occurred. The results indicated that base rate affected accuracy; flirting was more accurately detected in clips where flirting did not occur than in clips where flirting occurred. Study 2 also indicated that female targets’ flirting was more accurately judged than male targets’ flirting. Findings are discussed in relation to theory and courtship context.