Answers are grounded in this paper and cite their sources.
Pollitt, L, Pieske, O, Behm, DG, and Granacher, U. Six weeks of drop jump training on stable vs. unstable landing surfaces result in similar improvements in drop jumping performance in young healthy men. J Strength Cond Res XX(X): 000–000, 2026—To investigate how progressive drop jump (DJ) training performed on unstable landing surfaces affects jumping performance. Forty-two subjects were randomly allocated to no training (CONTROL, n = 8), stable surface (STABLE, n = 9), or 1 of 3 unstable landing conditions: mini-trampoline ( n = 9), both-sides-used balance trainer (BOSU, n = 8), or an equated volume of BOSU and STABLE (MIX, n = 8). Subjects in each experimental condition completed 1,080 jumps across 9 drop heights (0.2 m progressing to 1 m) during 3 sessions per week over 6 weeks. Pre- and post-testing assessed maximal DJ at low (0.2 m), moderate (0.6 m), and high (1 m) drop heights for reactive strength index (RSI), jump height (JH), and contact time (CT). Using baseline as a covariate, analysis of covariance revealed significant, large main effects of condition for DJ outcome measures across all drop heights ( 0.19 < η p 2 < 0.48 ). Bonferroni-adjusted post hoc comparisons identified significant postintervention differences between certain experimental conditions and CONTROL, depending on the outcome measure. No significant differences were observed between experimental conditions. At 0.2 m, differences in CT ( p = 0.016 ) and RSI ( p = 0.003 ) were observed for BOSU. At 0.6 m, CT for BOSU ( p = 0.033 ), JH for STABLE ( p = 0.034 ), and RSI for STABLE ( p = 0.012 ) and BOSU ( p = 0.013 ). At 1 m, JH for MIX ( p = 0.012 ), and RSI ( p ≤ 0.003 ) for all experimental conditions. Progressive DJ training on unstable training devices resulted in changes in performance comparable to a stable landing surface. When conceptualizing training programs, including plyometrics performed on unstable landing surfaces may provide an alternative neuromuscular stimulus without compromising performance.
No takes yet. Share an insight, caveat, or question.
Pollitt et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: