The load that maximises power is exercise-specific
≤30% 1RM for jump squats, 30–70% 1RM for the squat, ≥70% 1RM for the power clean and hang power clean
Participants
n=468 subjects in 27 studies (5766 effect sizes; databases searched to 30 June 2014); random-effects Hedges' g.
Loads grouped into Zone 1 (0-30% 1RM), Zone 2 (30-70% 1RM) and Zone 3 (≥70% 1RM).
Key Findings
Optimal load for peak power differed by exercise:
Squat: moderate loads (>30 to <70% 1RM).
Jump squat: lighter loads (≤30% 1RM).
Power clean and hang power clean: heavier loads (≥70% 1RM).
"Optimal load for power" cannot be generalised across lifts.
Practical Application
Pooling 27 studies (468 subjects), peak power came at ≤30% 1RM in the jump squat, 30-70% 1RM in the squat, and ≥70% 1RM in the power clean and hang power clean — match the training load to each exercise's power peak rather than using one "power %" for every lift.
Limitations
Coarse three-zone load bins; optimal-load estimates vary by method and population.
Peak power in a single session is not the same as long-term performance transfer.
How this entry was checked and changed
Content correction 2026-09-27
"~40-70% 1RM for the power clean" -> the abstract gives ≥70% 1RM for the power clean and hang power clean; 30-70% applies to the squat. Jump squat "near 0% (body mass)" -> ≤30% 1RM.
PMID 25998853 (resolves to an unrelated burn-healing paper) -> 26063470, the PubMed record returned by searching this DOI.
Added 27 studies / 468 subjects and the load zones.