REF-764 · Seitz et al., 2014 · Sports Medicine, Vol. · Paper · PubMed

Increases in Lower-Body Strength Transfer Positively to Sprint Performance: A Systematic Review with Meta-Analysis

A-fairly strong evidenceMeasured on a mixed groupWritten before October 2026, being re-checked

Meta-analysis of 15 studies (510 subjects, 85 effect sizes, 26 experimental and 11 control groups) asking whether gains in back-squat strength transfer to sprint performance. They do: squat-strength effect sizes and sprint effect sizes were very strongly correlated (r = -0.77), and resistance training improved sprint performance by 3.11% on average (sprint ES = -0.87).

Level of practice, body mass, sessions per week and rest between sets moderated the sprint gain; age, height, method, %1RM, programme duration, exercises, sets and reps did not.

Participants

n=510 subjects in 15 studies, nested in 26 experimental and 11 control groups, contributing 85 effect sizes. Age range, sex and sport are not given in the abstract.

Protocol

Search: ADONIS, ERIC, SPORTDiscus, EBSCOhost, Google Scholar, MEDLINE and PubMed, plus reference lists.
Exposure: resistance training with lower-body strength measured by the free-weight back squat.
Analysis: correlation between squat-strength ES and sprint ES; moderators tested (subject characteristics and resistance-training variables).

Key Results

Transfer: squat strength ES vs sprint ES, r = -0.77 (p = 0.0001) — described as a very large significant correlation. Overall sprint ES = -0.87, mean sprint improvement = 3.11 %.

Moderator of the sprint improvementStatisticSignificant?
Level of practicep = 0.03yes
Body massr = 0.35; p = 0.011yes
Resistance-training sessions per weekr = 0.50; p = 0.001yes
Rest interval between setsr = -0.47; p ≤ 0.001yes
Agep = 0.86no
Heightp = 0.08no
Resistance-training methodp = 0.06no
Average load intensity (% of 1 repetition maximum)p = 0.34no
Programme durationp = 0.16no
Exercises per sessionp = 0.16no
Sets per exercisep = 0.06no
Repetitions per setp = 0.48no
The abstract does not state the direction of the moderator correlations in words; because a faster sprint gives a negative sprint ES, the sign of r alone does not show which way each variable pushes the gain.
The authors caution that the number of ESs available differs widely across moderator categories.

Practical Application

"Will getting stronger in the squat make me faster?" — yes: in a meta-analysis of 15 studies (510 subjects), larger gains in back-squat strength went with larger sprint improvements (r = -0.77, p = 0.0001), and resistance training improved sprint performance by 3.11% on average (ES = -0.87).
The variables that moderated the gain were sessions per week, rest between sets, body mass and level of practice; %1RM, programme duration, exercises per session, sets and reps did not reach significance. Inference, not tested here: the practical lever is raising the squat through a programme the athlete can sustain, not a particular load scheme.
Not significant is not the same as no effect: method and sets per exercise (both p = 0.06) and height (p = 0.08) were close to the threshold.
Not from this paper; the base's own practice: for endurance runners and cyclists, the strength dose and its place in the week are in SYNTH-44 and SYNTH-43; longer rests between heavy sets built more strength in REF-679 (a strength outcome, not sprint).

Limitations

Meta-regression across studies: associations between groups, not proof that a given moderator causes a larger gain.
ESs are unevenly distributed across moderator categories (the authors' own caution).
Populations, sex split and sprint distances are not given in the abstract; strength was measured only by the free-weight back squat.
Written from the abstract and record metadata only (full text not read); heterogeneity statistics are not in the abstract.
How this entry was checked and changed
Why this level

meta-analysis of mostly small trials (n=510 across 15 studies)