Does stretching prevent running injuries? What the evidence shows

Part two of the Ironman series. Keeping range is worth doing, and every good reason for doing it differs from the reason nearly everybody gives.

Someone I trained with for years warned me, in a car park after a Sunday long run, that my body would eventually come apart. He'd watched me cross the line, drop onto the kerb, unlace my shoes and drive off without touching a single muscle. He was kind about it and he had thirty years of coaching orthodoxy behind him.

Holding a muscle at length has a decent case for range, and a decent case for settling you down after hard work. The case it has never made is the injury one, which happens to be the job almost everybody adopted it to do. That sounds academic until you count your hours. Six a week, a job, children, and the question becomes what earns the slot you can actually protect. Part one of this series took apart the number everyone quotes before a long race. This one is about what keeps you training at all.

This is an educational and strategic perspective, not personal medical advice. The views are the author's own and not statements by Atlas Cove Lda.

Thirty years of advice on a thin file

The rationale is intuitive, which explains most of its staying power. Tightness drags a joint off its line, something downstream takes force at an angle it was never built for, and restoring length restores the alignment. The joint really does travel further afterwards, and you can measure it. The reasoning breaks the moment your outcome stops being degrees of movement and becomes counted injuries.

Thacker and colleagues reviewed the published work linking stretching to injury risk and found no consistent protective effect, before a session or after it, in any group they could assess (Thacker et al., 2004). Behm and colleagues reached the question from another side twelve years on, and their answer had three parts. A static hold buys extra range for a period. That range fails to convert into fewer injuries among healthy, active people. And a lengthy passive hold, taken right before a demanding effort, can leave force and power temporarily reduced (Behm et al., 2016).

Read both carefully, because the crude version of this argument has damaged how it lands. Neither review says stretching does you harm. Neither says flexibility is worthless. Neither has a word to say about the quiet ten minutes some people spend on a mat before bed. One idea fails, which is that lengthening a muscle protects an otherwise healthy athlete, and the useful judgement lives between that modest finding and the sweeping one people believe they heard.

The limitations belong in the open rather than in a footnote. Thacker dates from 2004 and it is old. Its trials vary widely in design, in how an injury was defined and in whom they recruited, and a substantial share of the strongest data came from military recruits rather than from runners. A recruit trains to somebody else's schedule, under a load he didn't choose, often short of sleep. An adult fitting a marathon build around work and family is a different case. So the fair statement is this: careful work looked for protection wherever it could and kept coming back with nothing. That falls short of proving absence, and it sits well above the confidence with which the opposite gets repeated.

Static, dynamic, and the distinction that keeps getting lost

Two different practices share one word. Static stretching takes a muscle to the end of its range and parks it there, passively, usually thirty to sixty seconds. A dynamic warm-up takes a joint repeatedly through the range it's about to work in, intensity climbing, in shapes resembling the session ahead. When the literature counsels caution before hard efforts, the passive hold is what it means. Nobody is telling you to head out cold. (Where the foam roller sits between the two, I honestly don't know, and I've stopped pretending otherwise.)

Some reasons for working on range need no defence at all. One is a real deficit in something your sport demands. The runner whose ankle refuses to dorsiflex, which is the toes-toward-shin motion letting the body travel over a planted foot. The swimmer whose shoulder won't reach the arc the stroke wants. That gap sits in the path of an action repeated thousands of times, so closing it removes an actual ceiling.

Another is compensation. A stiff joint gets covered by a neighbour poorly built for the job, so limited hip extension in running often gets paid for further up, and the complaint surfaces well away from its cause. Giving that hip its range back isn't about suppleness. It switches off a repeated overload sitting upstream of something fragile. Nine hours in a chair then an hour of running fits that better than anything about racing does. What has no defence is the undirected hour closing no deficit you could name.

The Achilles that grumbles for ten minutes

To understand why loading beats lengthening, get precise about how these injuries occur. Load tolerance is what nearly all of them come down to. Tissue was asked, over and over, to take more mechanical stress than its structure could manage, until the demand found the weakest link. Running is a sequence of ground reaction forces, and every footstrike asks the leg to soak up energy and hand it back. The stretch-shortening cycle makes that cheap, and it relies on muscle and tendon acting as a spring several thousand times an hour. Run downhill and the eccentric portion rises. Eccentric loading means a muscle producing force while lengthening under load, which is the pattern tendons accumulate stress under.

Seen from there, the grumbles of every running club read differently. An Achilles that hurts for ten minutes and then goes quiet. A patellar tendon that flares after a heavy block. Neither is a stiffness problem. Both describe tissue asked to carry more than it had been prepared for.

If that reading holds, the answer lifts the ceiling on stress a tissue absorbs rather than adding length to it, and the evidence separates cleanly here. Lauersen and colleagues gathered randomised trials of exercise-based injury prevention. The strength programmes cut injury rates substantially. The stretching programmes achieved nothing of the kind (Lauersen et al., 2014). That gap is hard to account for unless the thing being moved is load tolerance.

Name the limits of that paper too. It combined interventions that differ enormously, since proprioceptive and multi-component programmes went into the same pool as strength training, and "strength training" ran from a gym block to a circuit on grass. Heterogeneity of that order buys a direction you can trust and a dose you cannot, and injury trials are close to impossible to blind.

The mechanism underneath is well enough described. Load a tendon progressively and the structure changes, gaining tendon stiffness, so it returns more force before damage accrues (Bohm et al., 2015). Here is the caveat that always gets skipped: Bohm measured tendon properties, mechanical and morphological, and not injuries. Joining that to a lower injury count is my inference rather than the paper's result. Eccentric work specifically, the controlled lowering a heavy squat or a Romanian deadlift hands you for free, is the foundation of loading-based rehabilitation for Achilles and patellar tendinopathy, the load-driven degenerative tendon condition (Malliaras et al., 2013). Progressive loading is already standard care for the injuries endurance athletes most want to avoid.

Stretching alters how loose a tissue feels this evening. Loading alters what it can survive next spring.

Static stretching and progressive loading, on the same axes

Both have a defensible case. They are not cases about the same thing, which three shared axes show faster than prose does.

What the evidence supports

Static stretching: a genuine short-term gain in range, plus a drop in how tight you feel (Behm et al., 2016). No consistent reduction in injury rates among healthy active people (Thacker et al., 2004; Behm et al., 2016).
Progressive loading: substantially lower injury rates across pooled trials (Lauersen et al., 2014), better running economy (Beattie et al., 2014; Balsalobre-Fernández et al., 2016), and status as the loading basis of tendinopathy rehabilitation (Malliaras et al., 2013).

What changes in the tissue

Static stretching: tolerance to the stretched position shifts and the joint travels further. How much stress the tissue absorbs is not the variable being moved.
Progressive loading: the tendon adapts as a structure, gaining stiffness and returning more force before damage accrues (Bohm et al., 2015). The muscle makes the same force at a smaller fraction of its ceiling.

Over what timescale

Static stretching: minutes to hours for the acute gain, and it fades. Range built through consistent work holds while the work continues, which is an argument for keeping it and not an argument about injuries.
Progressive loading: weeks for early neural gains, months for structural tendon adaptation, and the economy benefit in trained runners appeared across roughly two months (Balsalobre-Fernández et al., 2016).

Carrying more muscle up the hill

If loading only kept you healthy it would already deserve protected space in a full week. It also makes most people faster. The objection is reasonable enough: muscle weighs something, and endurance performance runs on power to weight. I accepted that for years and used it to skip the gym. Beattie and colleagues pooled what exists on strength training in endurance athletes, and the result points the other way. Resistance training improved running economy and the measures around it, while maximal oxygen uptake barely shifted and neither did body composition (Beattie et al., 2014).

Running economy carries that result and deserves understanding rather than repetition. It is the oxygen you burn to sustain a given pace, and two connected things improve it. A stiffer tendon hands back a greater share of the elastic energy each footstrike stores, rather than losing it as heat, and a stronger muscle needs a smaller slice of its ceiling for the same force. A narrower meta-analysis confined to highly trained runners reported a large economy benefit from a couple of months of lifting and plyometric work (Balsalobre-Fernández et al., 2016).

Confined is the word to hold onto. Highly trained runners have less room left to improve, so a large effect among them is impressive. They also bring years of movement competence most readers of this page don't have, so that effect size is not something anybody should be promised. Better economy in the closing hour of a long race is time and fuel returned to you at the only moment you cannot buy either. If racing isn't your thing, the same adaptation turns up as a Tuesday that leaves something in the tank by six.

What gets cut first when the week goes wrong

For nearly everybody reading this, the binding constraint is not knowledge and it is not motivation. It is hours. A build for a long event already swallows the spare time around a job and a family. Stack two or three gym sessions on that, then several more hours of undirected mobility work, and something gets dropped.

I can tell you precisely what gets dropped, because I dropped it. Across most of the decade I spent working toward long-distance racing, the gym went first whenever a week fell apart. My training log never turned red for a missed session, so the missing kept happening, and I filed it as a calendar problem when it was really a choice. What changed my mind was reading the literature properly, which is duller than a dramatic breakdown and happens to be what occurred. For years I had been rigorous about the hours with the least evidence behind them and sloppy about the one hour with the best. That still annoys me.

The part I would defend is what came next. With loading no longer negotiable, the two Ironman builds that followed had a different shape: three lifting sessions weekly, sets in the 8 to 12 range, kept in place through the heaviest endurance blocks, no static stretching at all, and neither build finished with an overuse injury. By that point I had trained consistently for twelve years, with six marathons and a pair of Ironman finishes on the record.

Now the careful part, about my own story. One body, one history, one set of habits. It settles nothing. It leans the way the pooled evidence leans, which pleases me and which nobody should rest a decision on, and treating an anecdote as evidence would repeat the exact error this habit was built from. An anecdote is how the belief got here in the first place.

Why the pleasant option keeps winning the last free hour is human enough. A long hold drops felt tightness immediately, while heavy loading leaves you tired and pulls on recovery your endurance work has already claimed. Something that feels safe stays reassuring until you work out what it returns.

Range of motion, defended on its own terms

Everything so far answers one question, which is how to avoid getting hurt while training hard on very few hours. That shouldn't be inflated into a judgement on mobility generally. Range has value with no bearing on injuries at all.

It falls away across adult life, unevenly across joints and consistently enough to quantify. Araújo examined around 6,000 Flexitest results. Flexibility declined at somewhere near one percent of range annually, with hips and ankles among the earliest joints to stiffen (Araújo, 2013). Handle that number with care. It is cross-sectional Flexitest data, so it describes how a population looks across ages rather than a path any individual was followed along. Your own rate might be quicker or slower. What it establishes is the direction.

None of it appears in a race result. It appears in whether you can sit on a floor with your children and stand back up without your back objecting, or rise from the ground unaided at seventy. Those matter more to me than any finishing time I own. Keeping a working range is a goal that stands up by itself, and it sits beside the argument for building strength as the decades stack up rather than against it.

Apply the same honesty to the recovery case. Gentle stretching does take the edge off felt tightness, and a few slow, well-breathed minutes at the end of the day shift things toward parasympathetic activity, which fits with treating recovery as part of the training. Defending that as injury prevention is the mistake. Or rather, it needs no defence at all, since helping you settle after a hard day justifies itself. A body planning to train for another twenty years wants range and load tolerance both, and they compete only when the hours run out.

Where the scarce hours should go

The division follows without much strain.

Keep the dynamic warm-up. Preparing tissue for the work ahead was never the disputed part.
Keep targeted mobility work, for a genuine sport-specific deficit or a compensation you can point at.
Untargeted mobility work can stay too. Keep it for the range and for winding down at night, and be honest that this is the job it's doing.
Two or three strength sessions a week of compound movements, load creeping up over months. That one is not available for trading.

Strength work has the best evidence on both counts, staying in one piece and going quicker, and it is reliably the first thing to disappear when the calendar tightens. I know. I spent a decade doing exactly that.

The man in the car park had one thing wrong, and it happened to be his firmest conviction. Stretching isn't a waste, and fifteen quiet minutes on the floor after a long ride does you no damage. The real correction is narrower than the shouted version. Mobility work has good justifications, injury prevention is not among them, and when the hours are short the hour doing most to keep you in one piece is the one you spend loading tissue.

Sources

Thacker, S. B., Gilchrist, J., Stroup, D. F., & Kimsey, C. D. (2004). The impact of stretching on sports injury risk: a systematic review of the literature. Medicine & Science in Sports & Exercise, 36(3), 371-378. DOI: 10.1249/01.MSS.0000117134.83018.F7
Behm, D. G., Blazevich, A. J., Kay, A. D., & McHugh, M. (2016). Acute effects of muscle stretching on physical performance, range of motion, and injury incidence in healthy active individuals: a systematic review. Applied Physiology, Nutrition, and Metabolism, 41(1), 1-11. DOI: 10.1139/apnm-2015-0235
Lauersen, J. B., Bertelsen, D. M., & Andersen, L. B. (2014). The effectiveness of exercise interventions to prevent sports injuries: a systematic review and meta-analysis of randomised controlled trials. British Journal of Sports Medicine, 48(11), 871-877. DOI: 10.1136/bjsports-2013-092538
Bohm, S., Mersmann, F., & Arampatzis, A. (2015). Human tendon adaptation in response to mechanical loading: a systematic review and meta-analysis of exercise intervention studies on healthy adults. Sports Medicine - Open, 1(1), 7. DOI: 10.1186/s40798-015-0009-9
Malliaras, P., Barton, C. J., Reeves, N. D., & Langberg, H. (2013). Achilles and patellar tendinopathy loading programmes: a systematic review comparing clinical outcomes and identifying potential mechanisms for effectiveness. Sports Medicine, 43(4), 267-286. DOI: 10.1007/s40279-013-0019-z
Beattie, K., Kenny, I. C., Lyons, M., & Carson, B. P. (2014). The effect of strength training on performance in endurance athletes. Sports Medicine, 44(6), 845-865. DOI: 10.1007/s40279-014-0157-y
Balsalobre-Fernández, C., Santos-Concejero, J., & Grivas, G. V. (2016). Effects of strength training on running economy in highly trained runners: a systematic review with meta-analysis of controlled trials. Journal of Strength and Conditioning Research, 30(8), 2361-2368. DOI: 10.1519/JSC.0000000000001316
Araújo, C. G. S. (2013). Age-related mobility loss is joint-specific: an analysis from 6,000 Flexitest results. GeroScience (AGE), 35(6), 2399-2407. DOI: 10.1007/s11357-013-9525-z

Questions people actually ask

Should I stretch before running?

Warm up, certainly. Hold a static stretch for a minute beforehand, probably not. Take the joints through the range the run will use, at rising intensity, and you've done the part nobody disputes.

The caution applies to long passive holds right before hard efforts, which can temporarily reduce force and power (Behm et al., 2016). Before an easy hour, ignore it. Before intervals, put the holds elsewhere in your day.

Does stretching prevent injury?

On the current evidence, no, not in healthy athletes as a general measure. Thacker and colleagues looked for lower injury rates from stretching before or after exercise and did not find them consistently (Thacker et al., 2004). Behm and colleagues arrived at the same place from a different starting point (Behm et al., 2016).

Be precise about what that doesn't mean. It doesn't mean stretching causes injuries, and it says nothing about rehabilitating a problem under guidance or about a sport that demands a range you lack. It means the general habit, performed as insurance by people who are currently fine, hasn't been shown to do the job it was adopted for.

Will lifting make me heavier and slower?

That was my assumption for years and the pooled evidence went the other way. Beattie and colleagues found resistance training improved running economy and related performance measures in endurance athletes, with maximal oxygen uptake and body composition staying where they were (Beattie et al., 2014).

The mechanism isn't mysterious. A stiffer tendon returns more of the energy each footstrike stores, and a stronger muscle spends less of its ceiling on the same force, so the oxygen cost of your pace falls. I put on about two kilograms across a twelve-week build and detected no cost on the road.

How much strength work does an endurance athlete need?

Two or three sessions a week of compound movements, load progressing over months, sets of eight to twelve reps. That is the shape the evidence supports and the shape I use.

The numbers matter less than whether the sessions survive a bad week. Almost nobody is undone by the wrong rep scheme. They're undone by the gym going first when the calendar tightens.

Where Atlas Cove fits

An Atlas Cove week gets built on this same division of hours, so what a particular body is short of is established before the scarce time is assigned. A habit can't do that reading for you. How the week is sequenced around it is set out in the method.

See how we sequence the work

Tom Wuerden · Co-Founder

Engineer turned Ironman