Stretching (Before or After Exercise) Does Not Meaningfully Reduce DOMS

Muscle soreness — specifically DOMS, the diffuse ache that peaks 24-72 hours after unaccustomed or eccentrically-loaded exercise (see Muscle Soreness — Acute vs. DOMS) — is mechanically distinct from the reason people reach for a stretch. DOMS is now understood as microtrauma to muscle/connective tissue plus an inflammatory cascade (Three Hypotheses of Muscle Soreness), not a state of "tightness" that lengthening the muscle would relieve. That mechanistic mismatch is why the null result below isn't surprising once you see it, even though "stretch it out" is one of the most common pieces of post-workout advice given.

The foundational evidence (Herbert, de Noronha & Kamper, Cochrane 2011 update, the definitive systematic review on this specific question): pooled across 12 randomized trials (2,597 total participants, dominated by one large field trial of 2,377 military recruits), stretching — whether performed before exercise, after exercise, or both — did not produce a clinically important reduction in DOMS in healthy adults. The actual numbers: post-exercise stretching alone reduced next-day soreness by about 1 point on a 0-100 pain scale (mean difference -1.04, 95% CI -6.88 to 4.79) — statistically indistinguishable from zero. Pre-exercise stretching reduced it by about 0.5 points. The one condition that reached statistical significance — stretching both before AND after, in the large recruit trial — still only produced a 4-point reduction in peak soreness over a full week (mean difference -3.80, 95% CI -5.17 to -2.43), which the review's own authors characterized as real but "very small," not clinically worthwhile. This held consistent whether soreness was measured half a day to three days post-exercise.

Confirmed, not weakened, by more current evidence. A 2021 systematic review and meta-analysis (Afonso et al., Frontiers in Physiology, 11 RCTs, 229 participants after exclusions) found no significant effect of post-exercise stretching vs. passive recovery on DOMS at 24, 48, or 72 hours (effect sizes ranging -0.09 to -0.24, all with confidence intervals crossing zero, p = 0.19-0.63) or on strength recovery (ES = -0.08, p = 0.75) — heterogeneity was near zero (I² = 0.0%) across studies, meaning this wasn't a case of conflicting trials averaging out to nothing; the individual trials themselves consistently found nothing. The review's own conclusion: there wasn't sufficient evidence to reject the hypothesis that stretching and simply doing nothing (passive recovery) have equivalent effects on recovery.

2025 currency check (most current available): Zhang, Chen & Xing, a 2025 meta-analysis of post-exercise stretching (static, dynamic, or PNF) vs. no stretching on lower-limb recovery — 15 studies, 465 participants — found statistically non-significant effects on muscle soreness (SMD = -0.06, 95% CI -0.32 to 0.19, p = 0.63), strength (SMD = 0.27, 95% CI -0.14 to 0.68, p = 0.19), and performance (SMD = 0.18, 95% CI -0.11 to 0.46, p = 0.22). This is the most recent dedicated meta-analysis on the question and it reaches the same conclusion as the 2011 Cochrane review 14 years earlier, across a broader set of stretching modalities (not just static) — the null finding has not been overturned by newer or larger evidence; if anything it has been reinforced by every subsequent synthesis.

What stretching does still do: static stretching genuinely improves range of motion and can reduce the sensation of joint/muscle stiffness (see Static Stretching), and some evidence suggests it modestly reduces central inhibition around a stiff joint — but stiffness and DOMS are not the same symptom, and relief of one doesn't imply relief of the other. This module is specifically about the DOMS-reduction claim, not a statement that stretching has no value at all.

Distinct from a different, already-covered claim: this is not about static stretching's effect on power/strength output performed immediately before a lift (see Static Stretching Before Lifting — Acute Power/Strength Impairment for that separate, dose-dependent, real effect) — that's a same-session performance-timing question. This module is about whether stretching (at any timing) changes how sore you feel in the following days, which is a different claim with different evidence behind it.

• Cochrane 2011 (Herbert, de Noronha & Kamper — update of a 2007 original review): 12 RCTs, 2,597 participants. Stretching before, after, or before-and-after exercise does not produce a clinically important reduction in DOMS in healthy adults.
• Post-exercise stretching alone: ~1-point reduction on a 0-100 soreness scale at one day post-exercise (mean difference -1.04, 95% CI -6.88 to 4.79) — not statistically distinguishable from zero.
• Pre-exercise stretching alone: ~0.5-point reduction — smaller still.
• Stretching before AND after (the one statistically significant result, from a single large trial, n=2,377): ~4-point reduction in peak weekly soreness (mean difference -3.80, 95% CI -5.17 to -2.43) — real but the review's own authors call it "very small," not clinically worthwhile.
• 2021 meta-analysis (Afonso et al., 11 RCTs, 229 participants): no significant effect of post-exercise stretching on DOMS at 24h/48h/72h or on strength recovery; near-zero heterogeneity across trials (I² = 0.0%) — a consistent null, not conflicting evidence averaging out.
• 2025 meta-analysis (Zhang, Chen & Xing, 15 studies, 465 participants, covering static/ dynamic/PNF stretching): non-significant effect on soreness (SMD = -0.06), strength (SMD = 0.27), and performance (SMD = 0.18) — the null finding holds as of the most current available synthesis, across stretching modalities beyond just static holds.
• Stretching still has real, separate benefits (ROM, perceived stiffness) — this module addresses the DOMS-reduction claim specifically, not stretching's value overall.
• Distinct from the separate, real, dose-dependent pre-lift power/strength deficit covered in Static Stretching Before Lifting — Acute Power/Strength Impairment — different claim, different timing, different evidence.