Assessing Isometric Hip Adductor and Abductor Strength in Elite Youth Soccer Players: Does Hip-Flexion Angle Matter?

Taylor, JM, Pallotta, H, Hamilton, A, Tears, C, and Wright, MD. Assessing isometric hip adductor and abductor strength in elite youth soccer players: Does hip-flexion angle matter? J Strength Cond Res 40(9): e983-e990, 2026-Isometric hip strength is a risk factor for hip and groin injuries in soccer and is assessed at various short lever positions (i.e., a hip-flexion angle of 45-90°). In this study, the effects of hip-flexion angle on peak adduction and abduction force, adduction-to-abduction ratio (ADD:ABD), and relationship between test positions were assessed. Twenty-four male soccer players (age 17.4 ± 0.8 years) completed 3 testing sessions across 2 weeks, during which peak adduction and abduction force and ADD:ABD were assessed at 45-, 60-, and 90° of hip flexion on a fixed frame dynamometer. Players completed three 5-second maximal voluntary contractions in each position, and the tests were completed in counterbalanced order. Robust repeated measures ANOVAs were used to compare peak force and ADD:ABD between positions, with significance set at p < 0.05. Pearson's correlations were used to quantify associations between peak force on each test. Significant differences were observed in adduction between 45 and 60° (right -25.9 ± 23.1 N [±95% confidence limit]; left -33.4 ± 19.1 N); 45 and 90° (right 46.8 ± 23.2 N; left 32.6 ± 19.9 N) and 60 and 90° (right 72.7 ± 22.9 N; left 66.1 ± 19.9 N). Significant differences were also observed in abduction between 45 and 90° (right 32.2 ± 20.8 N; left 25.7 ± 21.1 N) and 60 and 90° (right 42.4 ± 20.9 N; left 41.6 ± 21.3 N). Very large correlations between tests were observed in adduction and abduction, respectively ( r = 0.77-0.87). However, prediction errors were also large (∼40 N). Adduction-to-abduction ratio was highest at 60° (1.10). In conclusion, peak adduction and abduction force vary between short lever testing positions. Although strong associations between testing positions exist, large prediction intervals are evident, and these tests should not be used interchangeably.