Résumé : Objectives: The primary purpose was to examine the reliability of a new shoulder physical performance test -the Shoulder Endurance Test (SET)- in young healthy overhead athletes and sedentary adults and to provide preliminary reference values. The secondary objective was to determine whether there are differences on SET scores based on groups, sides and days. The third objective was to evaluate the relationship between the SET and shoulder rotational isometric strength in both groups. Design: Reliability and validity study. Setting: Laboratory setting. Participants: A total sample of 92 participants volunteered to participate in this study (30 healthy overhead athletes - 62 sedentary adults). Main outcome measures: We used a two-session measurement design separated by seven days to evaluate the reliability. We calculated intraclass correlation coefficients to determine relative reliability and used standard error of measurement and minimal detectable change to quantify absolute reliability. Systematic differences in SET scores between groups, days and sides were analysed with a two-way analysis of variance (ANOVA) for repeated measures. To check for systematic differences within groups between day 1 and day 2, a Wilcoxon Signed Rank Test was performed. Relationship between shoulder rotational isometric strength and the SET was determined using the Spearman Rank test (rs). Results: Relative reliability was high to very high in both groups (intraclass correlation coefficient [2,1] range = 0.78–0.93) and absolute reliability was clinically acceptable. The standard error of measurement varied from 10.7 s to 16.45 s. The minimal detectable change ranged from 29.6 s to 45.6 s. Weak correlations were found between the SET and isometric shoulder rotational strength (rs range = 0.309–0.431). Results: of the ANOVA for repeated measures showed a significant two-way interaction effect for day x groups (p = 0.020) and a significant main effect for side (p= < 0.001). Results of the Wilcoxon Signed Rank Test showed no systematic differences in group 1 between day 1 and day 2 for both sides (p = 0.79 dominant side; p = 0.66 non-dominant side). Conclusions: The SET is a reliable clinically applicable shoulder physical performance test in young adult overhead athletes and sedentary adult.