Relative reliability, construct validity, and concurrent validity of the Jiu-Jitsu Anaerobic Performance Test (JJAPT)

This study aimed to investigate the test-retest reliability of the Jiu-Jitsu Anaerobic Performance Test (JJAPT), as well as its construct validity according to experience level (intermediate and advanced) and belt rank (blue, purple, brown, and black), in addition to its concurrent validity with physical performance measures. Twenty-three amateur Brazilian Jiu-Jitsu (BJJ) athletes participated in two assessment sessions. JJAPT repetitions (mean, peak, and total) were analyzed under 4 and 5-set protocols, along with physical tests including vertical jump, Wingate test, and specific strength-endurance tests. Reliability was assessed using the intraclass correlation coefficient (ICC), construct validity through t-tests and ANOVA, and concurrent validity through Pearsons correlation. Results indicated that the JJAPT showed moderate to good reliability (ICC = 0.74 to 0.82). No significant differences were observed between experience levels or belt ranks for any variables (p > 0.05), with trivial effect sizes, indicating a lack of discriminant validity. On the other hand, significant correlations were found between JJAPT performance and isometric and dynamic strength-endurance tests (r = 0.39 to 0.62), with moderate to large magnitudes. It is concluded that the JJAPT shows measurement stability; however, it demonstrates limited ability to discriminate performance between athletes of different experience levels and belt ranks, being more strongly associated with strength-endurance capacities. Keywords: Brazilian Jiu-Jitsu; anaerobic performance; training monitoring