The fitness influencer speaks with the certainty of the converted. If you are running and not stretching, you are asking for injury. Dynamic before. Static after. Mobility work on your rest days. The prescription arrives without citation, without nuance, without the faintest acknowledgment that an entire body of research has spent two decades arriving at the opposite conclusion. Stretching does not prevent running injuries. It has never prevented running injuries. And the evidence is not close.
Somewhere on social media, a verified account is dispensing commandments. Dynamic plus static stretching and mobility work, or you are courting disaster. The post collects thousands of likes from runners who nod along, who have internalized this catechism so thoroughly that questioning it feels like questioning gravity. They stretch before runs and after runs and on rest days, tithing minutes of their finite lives to a ritual that science cannot justify.
The belief is ancient by internet standards and universal by running standards. Stretch or break. Limber or lame. It persists not because anyone has proven it true but because no one has bothered to check whether it is.
We checked.
The Trials That Settled It
The question of whether stretching prevents injury is not open. It has been answered, repeatedly, by large randomized controlled trials spanning military recruits, recreational athletes, and distance runners. The results do not vary. The signal does not waver.
Pope et al. (2000) randomized 1,538 male army recruits into two groups. One performed 20-second static stretches for six leg muscle groups during every warm-up. The other skipped stretching entirely. After 12 weeks: hazard ratio 0.95 (95% CI 0.77–1.18). No significant difference.
The authors calculated that 337 people would need to stretch to prevent a single injury. Three hundred and thirty-seven. At that number, the intervention is indistinguishable from doing nothing at all.
— Pope et al. (2000), Medicine & Science in Sports & Exercise
Jamtvedt et al. (2010) conducted the largest trial directly testing stretching for injury prevention: a pragmatic RCT with 2,377 physically active adults who stretched seven lower-limb and trunk muscle groups before and after exercise for 12 weeks. The primary outcome: HR = 0.97 (95% CI 0.84–1.13). Stretching reduced nothing.
— Jamtvedt et al. (2010), BMJ
In recreational runners, Van Mechelen et al. (1993), assigned 421 male runners to a 16-week program of warm-up, stretching, and cool-down. The intervention group experienced slightly more injuries (5.5 vs 4.9 per 1,000 running hours), though the difference was not significant. The stretchers did not fare better. They fared, if anything, marginally worse.
— Van Mechelen et al. (1993), American Journal of Sports Medicine
Three large trials. Three null results. But perhaps individual studies can mislead. Let us consult the meta-analyses, where the data of thousands are pooled and the noise of chance is filtered away.
The Meta-Analyses Agree. All of Them.
What makes the case against stretching particularly damning is not any single study but the convergence. Independent research teams, working in different decades, using different methods, analyzing different populations, arrive at the same conclusion with almost eerie consistency. The effect of stretching on injury rates falls between a 4% and 8% reduction across every major meta-analysis ever conducted. None reaches statistical significance. None approaches clinical relevance.
The numbers speak with one voice:
Herbert & Gabriel (2002): Stretching reduces injury risk by approximately 5%. Not clinically meaningful.
Thacker et al. (2004): Pooled OR = 0.93 (95% CI 0.78–1.11). Not significant. Concluded there was "not sufficient evidence to endorse or discontinue routine stretching."
Lauersen, Bertelsen & Andersen (2014): 25 RCTs, 26,610 participants, 3,464 injuries. Stretching RR = 0.963 (95% CI 0.846–1.095). The only intervention category that failed to reach significance.
Leppänen et al. (2014): 68 RCTs. Stretching OR = 0.92 (95% CI 0.80–1.06). Not significant.
Afonso et al. (2025): Screened over 300,000 records, included 19 studies and 9,002 participants. OR = 0.945 (p = 0.396). No benefit detected across 19 sensitivity analyses.
— Herbert & Gabriel (2002); Thacker et al. (2004); Lauersen et al. (2014); Leppänen et al. (2014); Afonso et al. (2025)
Five independent meta-analyses. Over 300,000 records screened. Tens of thousands of participants. The effect sizes cluster around 1.0 with the discipline of a metronome. Stretching does not prevent injury. The question is settled. It has been settled for years. The fitness industry simply has not received the memo, or has chosen not to open it.
The Muscle Strain Exception (and Its Catch)
In the interest of doctrinal honesty, the Cult acknowledges one wrinkle. A small subset of research suggests stretching may reduce acute muscle strains specifically. Takeuchi et al. (2024) found a 63% reduction in muscle-specific injuries across four RCTs. This sounds impressive until you examine the fine print.
The 2025 Delphi consensus panel of 20 leading international researchers flagged a critical finding: in studies showing reduced muscle injuries, bone and joint injury rates increased. Stretching did not reduce the total injury burden. It redistributed it. You traded a pulled hamstring for a stress fracture. The body's accounting is ruthless.
Moreover, the muscle-strain benefit appears limited to sports involving explosive stretch-shortening cycles: sprinting, football, soccer. For distance runners, whose injuries are overwhelmingly overuse-driven—patellofemoral pain, IT band syndrome, tibial stress—the trade is especially poor.
— Takeuchi et al. (2024); Warneke et al. (2025), Journal of Sport and Health Science
This is the nuance the social media post does not contain. Stretching may help with one category of injury while worsening another. The net effect remains zero. The body is not fooled by the intervention. It simply moves the damage elsewhere.
The Mobility Myth
When the case for stretching collapses, the faithful retreat to a second fortification: mobility work. Foam rolling. Banded distractions. The assumption is intuitive and wrong: more range of motion means fewer injuries.
No randomized controlled trial has ever demonstrated that foam rolling reduces running injury incidence. The theoretical chain from rolling to range of motion to fewer injuries has a broken middle link.
Christopher, McCullough & Snodgrass (2019) systematically reviewed clinical predictors of running injury and found "very low quality of evidence" for all assessment categories. The results defied the flexibility gospel entirely: decreased hip internal rotation was protective against injury, and increased hip abduction strength predicted higher injury risk. The tighter runners fared better.
A 2022 meta-analysis of biomechanical and musculoskeletal measures in 3,404 runners testing over 100 risk factors concluded that available literature "does not generally support biomechanical or musculoskeletal measures as risk factors" for running-related injury in non-elite runners.
— Christopher et al. (2019); Baumbach et al. (2022), Sports Medicine – Open
Foam rolling improves range of motion. This is documented. What is not documented is that improved range of motion prevents injury in runners. The logical leap from one to the other has been assumed so broadly and for so long that most runners do not realize it was never proven. It was simply repeated until it felt true.
The Expert Consensus of 2025
In January 2025, twenty of the world's leading stretching researchers published a Delphi consensus statement after three structured rounds of expert deliberation. These were not fringe contrarians. The panel included David Behm, Anthony Blazevich, Andreas Kay, and other figures who have collectively spent centuries studying human flexibility. They required 80% agreement for any recommendation to stand.
On injury prevention, the panel reached 85% consensus on a single, unambiguous recommendation: "The panel does not recommend stretching for injury prevention in general."
The statement went further. Stretching does not contribute to muscle growth. Does not improve posture. Does not enhance post-exercise recovery. It reliably improves range of motion and reduces muscle stiffness, which may serve specific clinical or performance goals. Injury prevention is not among them.
— Warneke et al. (2025), Journal of Sport and Health Science
When the people who have devoted their careers to studying stretching tell you it does not prevent injuries, perhaps it is time to listen.
What the Evidence Actually Supports
If stretching does not protect the runner, what does? Here the research offers genuine, actionable clarity.
Strength Training
The contrast between stretching and strength training is not subtle. It is a chasm.
Lauersen et al. (2014) found that strength training reduced sports injuries to less than one-third of baseline rates: RR = 0.315 (95% CI 0.207–0.480). A follow-up analysis confirmed the effect is dose-dependent—every 10% increase in strength training volume reduced injury risk by more than 4 percentage points.
Stretching: 4% reduction, not significant. Strength training: 69% reduction, highly significant. The evidence could not be more clear about where a runner's non-running time should go.
— Lauersen et al. (2014); Lauersen, Andersen & Andersen (2018), British Journal of Sports Medicine
Training Load Management
The strongest predictor of running injury remains what it has always been: doing too much, too soon, without adequate recovery. Gradual progression, recovery weeks, and attention to single-run spikes matter infinitely more than whether you touched your toes beforehand.
Sleep and Recovery
The time you spend foam rolling before bed would serve you better as sleep. Recovery is not a product you perform. It is a biological process that occurs when you stop performing and allow the body to rebuild.
The Liberation
You have been told, by coaches and influencers and well-meaning friends, that failure to stretch is negligence. That the ten minutes of hamstring holds and hip openers and calf stretches are the price of admission to a running life free from injury. The tithe has been collected for decades.
The tithe purchased nothing.
Over 300,000 research records have been screened. Tens of thousands of participants have been randomized. Twenty of the world's foremost experts on stretching have conferred and voted. The answer has not changed since Herbert and Gabriel first quantified it in 2002: stretching does not meaningfully reduce injury risk for runners. The effect is a rounding error dressed in athletic clothing.
The body does not require limbering. It requires loading. Strength, not suppleness, stands between you and injury. The minutes you surrender to the stretching mat are minutes stolen from the squat rack, the single-leg deadlift, the Nordic curl—interventions that reduce injury by two-thirds rather than by a statistically imaginary four percent. Redirect the offering.
If you enjoy stretching, stretch. It will not harm you. It may improve your range of motion where deficits exist. It may feel pleasant in the way that any ritualized movement can feel pleasant. But do not stretch because you believe it will keep you healthy. That belief, however deeply held, is not supported by the accumulated weight of human inquiry into the matter.
The influencer's post will collect its likes. The comments will fill with runners confirming what they have always believed. The algorithm will amplify the signal. And somewhere, a runner who could have spent twenty minutes getting stronger will instead spend twenty minutes getting more flexible, mistaking one for the other, trading protection for performance theater.
You know better now. The evidence has spoken. It spoke clearly, and it spoke decades ago.
Go lift something heavy. Then go run.
