Fitness level contributes to a law enforcement recruit’s ability to graduate from a training academy. However, limited research has used receiver operating characteristic (ROC) curves to ascertain whether fitness tests can accurately predict academy graduation. The investigators analysed archival data from 311 recruits (260 males and 51 females) across 4 academy classes from 1 law enforcement agency. A robust, broad battery of fitness test data included grip strength for both hands, vertical jump, 75-yard pursuit run (75PR), 2-kgmedicine ball throw, push-ups, sit-ups, arm ergometer revolutions in 60 seconds, and 20-m multistage fitness test (20MSFT)shuttles. Recruits were split into those who graduated (n 5 269) or those who separated (did not complete academy; n 5 42). A univariate ANOVA, with sex as a control variable, determined between-group fitness differences. Receiver operating characteristic curves were plotted for each fitness test, and the area under the curve (AUC) determined accuracy. Except for grip strength, graduated recruits were superior in all fitness tests (p , 0.001). The 75PR (AUC 5 0.708) and 20MSFT (AUC 5 0.727) had fair accuracy in predicting academy graduation. The data suggested that, ideally, fitness tests should not be used for diagnostic purposes in relation to academy graduation potential. Fitness test data can be used to provide specific feedback to at-risk recruits. These results suggest that the 75PR (change-of-direction speed) and 20MSFT (aerobic fitness) could contribute to successful graduation in law enforcement recruits. Law enforcement training staff could use this information to provide targeted training for recruits lacking in these qualities.