The paper
Retrospective normative database study, N=101,898 drug-tested classic competitive powerlifters aged 21-49.
Found
Relative strength (bodyweight multiplier) declined non-linearly with body mass at every percentile studied (p<0.001); the smoothed male 99th-percentile deadlift multiplier fell from 4.37x bodyweight in the 59kg class to 2.65x in the 120kg+ class, with decline steeper at the 99th percentile than the 50th, and women showing an even steeper mass-related fall in bench and squat than men.
Appraisal
The statistical significance here is close to meaningless given the sample size; what carries weight is the magnitude, and a 99th-percentile deadlift multiplier drop from 4.37x to 2.65x across the weight spectrum is a genuinely large, practically relevant gradient that undercuts flat bodyweight-multiplier benchmarks. It is purely descriptive of what competitive powerlifters achieve, not a causal or interventional claim, so it should be read as a reference tool rather than evidence that changes treatment or training decisions on its own.
The gap
It's not clear how much the 'Best-Ever' peak-selection filter inflates the upper percentiles versus a single competition-day performance, whether norms from athletes who chose to compete (and stayed in the sport long enough to log a best) generalise to non-competitive strength athletes or clinical return-to-sport testing, and the sample is capped at 21-49 years old, so it says nothing about youth or masters lifters.
Landmark
This extends the same rationale behind long-standing formula-based normalisation systems such as the Wilks coefficient and IPF GL points, which were built precisely because relative strength does not scale linearly with body mass; this paper re-quantifies that relationship empirically with quantile regression on a much larger, modern dataset.
On Monday
S&C coaches benchmarking competitive or serious raw powerlifters can replace flat bodyweight-multiplier targets with this paper's absolute-load, percentile-and-weight-class-specific tables, particularly for heavier lifters generally and for women's bench press and squat specifically, where flat multiplier thresholds are least realistic. Because the norms are built from each athlete's lifetime-best single performance, treat them as a benchmarking and goal-setting reference, not as a single-session pass/fail cutoff for return-to-sport strength testing, and don't apply them outside competitive raw powerlifting.
