The authors investigate thecorrelation between National Football League (NFL) combinetest results and NFL success for players drafted at three differentoffensive positions (quarterback, running back, and widereceiver) during a recent 6-year period, 1999–2004. Thecombine consists of series of drills, exercises, interviews,aptitude tests, and physical exams designed to assess the skillsof promising college football players and to predict theirperformance in the NFL. Combine measures examined in thisstudy include 10-, 20-, and 40-yard dashes, bench press, verticaljump, broad jump, 20- and 60-yard shuttles, three-cone drill, andthe Wonderlic Personnel Test. Performance criteria include 10variables: draft order; 3 years each of salary received and gamesplayed; and position-specific data. Using correlation analysis, wefind no consistent statistical relationship between combine testsand professional football performance, with the notableexception of sprint tests for running backs. We put forthpossible explanations for the general lack of statistical relationsdetected, and, consequently, we question the overall usefulnessof the combine. We also offer suggestions for improving theprediction of success in the NFL, primarily the use of morerigorous psychological tests and the examination of collegiateperformance as a job sample test. Finally, from a practicalstandpoint, the results of the study should encourage NFL teampersonnel to reevaluate the usefulness of the combine’s physicaltests and exercises as predictors of player performance. Thisstudy should encourage team personnel to consider theweighting and importance of various combine measures andthe potential benefits of overhauling the combine process, withthe goal of creating a more valid system for predicting playersuccess.