Context Medical school admissions committees attempt to select the most qualified applicants. In addition to traditional performance measures, committees often look favourably upon applicants who report previous clinical experience. Objectives This study aimed to determine if self-reported clinical experience is a valid indicator of future performance in medical school and internship. Methods We collected data for seven year groups (1993-1999; n=1112) and operationalised trainee performance in terms of five outcomes: cumulative medical school grade point average (GPA); US Medical Licensing Examination (USMLE) Step 1 and 2 scores, and scores on a validated programme director's evaluation measuring intern expertise and professionalism. We then conducted a series of analyses of covariance to compare outcomes in applicants who self-reported previous clinical experience with outcomes in those who did not. In these analyses, the independent variable was self-reported clinical experience (yes/no), the covariate was undergraduate GPA, and the dependent variables were the five performance outcomes. Results In four of five analyses, we found no differences in the performance of the two groups (clinical experience versus no clinical experience). However, on the cumulative medical school GPA outcome, applicants who reported previous clinical experience had statistically significantly lower cumulative GPAs upon graduation than those who did not report such experience (F 1,940=9.35, p=0.002, partial η 2=0.01 [small effect size]). Conclusions Our results suggest that applicants who self-report previous clinical experience may not be better candidates. In fact, on some measures of performance, these applicants may actually perform worse than those who report no clinical experience.