A regular stretching or mobility habit reliably increases range of motion, but daily frequency isn't clearly necessary and it doesn't reduce injuries or add meaningful strength
Interpretation
ROM gains are well supported: a 77-study meta-analysis (3,870 people, mean age 27) found a moderate effect, and a 13-study detraining analysis (556 people) found gains largely persist 2–6 weeks after stopping. Dose evidence disagrees: Konrad 2024 found frequency and total duration non-significant, while Thomas 2018 (23 studies) found weekly frequency and at least 5 minutes a week mattered. Strength and power gains are trivial-to-small (41 studies), and a 25-RCT meta-analysis (26,610 people) found stretching didn't reduce sports injuries (RR 0.963) while strength training did (RR 0.315). Strong evidence for range, weak or negative for the wider claims; no triathlete trials.
Practical implication
Use a short routine to build or keep specific range a triathlete needs, three to seven days a week, without expecting injury prevention or speed from it; keep strength training as the priority if time is short. Based on young, mostly non-endurance populations.
Sources
7 sources, independently verified.
Chronic stretch training produced a moderate ROM increase vs controls (ES −1.002, 95% CI −1.165 to −0.840; I² 75%). Static and PNF stretching beat ballistic/dynamic (ES ≈ 1.01 and 1.28 vs 0.55). Females gained more than males (ES ≈ 1.56 vs 0.88). Age (meta-regression p = 0.40), training status (p = 0.74), weekly frequency (p = 0.42) and total stretch duration (p = 0.73) were not significant moderators. By muscle: hamstrings ES ≈ 1.16, hip flexors ≈ 1.13, triceps surae ≈ 0.70. Substantial heterogeneity and signs of reporting bias; participants mostly young.
All stretching types improved ROM long term, static more than ballistic or PNF. Total weekly stretching time appeared key once it reached at least 5 minutes a week; time within a single session didn't matter; weekly frequency was positively associated with gains. Suggested static stretching at least 5 days a week, at least 5 minutes a week.
A single stretching bout produced a small acute ROM increase vs non-active control (ES −0.555, 95% CI −0.677 to −0.434; I² 33%; GRADE moderate). Gains for sit-and-reach, hamstrings and triceps surae, not hip adductors. Intensity, training status, technique and sex were not significant moderators; no relationship with age or stretch duration.
Stretch training gave a large ROM gain vs baseline (ES 0.93). After 2–6 weeks off, ROM stayed moderately above baseline (ES 0.55) but fell slightly from post-training (ES −0.41, p = 0.016). Weekly stretching frequency moderated the loss; only active participants maintained ROM. GRADE: high confidence for training gains, low for maintenance.
Chronic static stretching gave trivial-to-small gains in strength (SMD 0.21) and power (SMD 0.19) and a moderate-to-large flexibility gain (SMD 0.96). Sedentary participants gained more strength (SMD 0.58) than recreationally active (0.16); older mean age and more female participants predicted larger strength gains. Passive stretching beat active for flexibility.
No difference between strength training and stretching for ROM gains (ES −0.22, 95% CI −0.55 to 0.12, p = 0.206); subgroups by risk of bias, active vs passive ROM and joint showed no differences. Studies highly heterogeneous.
Stretching did not reduce sports injuries (RR 0.963, 95% CI 0.846–1.095; 3 trials); strength training did (RR 0.315, 0.207–0.480; 4 trials), as did proprioception training (RR 0.550) and multiple exposures (RR 0.655).