Objective: The purpose of this study was to determine the intrarater and interrater reliability of the Balance Error Scoring System (BESS). Design: A prospective observational study. Setting: Academic sports medicine center. Participants: Three scorers participated in this study. Methods: Three scorers experienced in using the BESS viewed a videotape depicting 30 consecutive individuals performing the BESS stance positions. The 3 scorers independently scored each of the 30 videotaped individuals using the BESS scoring criteria. A week later, the same 3 subjects viewed and scored the videotape again. Main Outcome Measurements: The interrater and intrarater reliability of the BESS was determined using intraclass correlation coefficients (ICC), reported with 95% confidence intervals. The minimum detectible change was also determined. Results: The interrater and intrarater reliability ICCs for the total BESS scores were 0.57 and 0.74, respectively. The interrater reliability ICCs for the 6 different stance positions were between 0.44 and 0.83, while the intrarater reliability ICCs were between 0.50 and 0.88. The interrater and intrarater minimum detectible change for the total BESS score were 9.4 and 7.3 points, respectively. Conclusion: This study suggests that certain subcategories of the BESS have sufficient reliability to be used in the evaluation of postural stability but that the total BESS score is not reliable. In addition, a change in score of greater than 9.4 (interrater) or 7.3 (intrarater) points is required before the change in postural stability can be attributed to the balancer rather than to the scorer.