Stretching before or after a workout: the answer depends on what you want from it

Few training questions have swung this hard in twenty years. There was a time when the whole football team sat on the grass stretching before kick-off. Then the reversal arrived: never stretch beforehand, it makes you weaker. Now you are standing in the hallway at half past four in the morning with no idea whether those two minutes help or hurt.

The good news is that the evidence here is unusually clear. The less good news is that it does not answer the question people usually ask. "Before or after" assumes stretching is one thing with one best slot. It is not. This guide starts by splitting the question apart, then goes through what each moment actually does, and ends with a dose that has a source.

The question contains three different goals

When someone asks whether to stretch before or after a session, they are really after one of three quite different things.

• Perform better in the session. Run faster, lift heavier, jump higher.
• Avoid injuries. Not pull a hamstring, not wake up hurting.
• Become more mobile. Reach your shoes, sit comfortably on the floor, turn your head to check the mirror.

Those three goals have three different answers, which is why the advice online seems to contradict itself. An answer that is right for the first goal is irrelevant to the third. Keep them apart and the rest resolves.

Before the session: warming up is movement, not stretching

The clearest signal does not come from a study but from a health service's own routine. The NHS has two separate pages, one on warming up before exercise and one on stretching after it. The warm-up routine contains no stretches at all. It is marching on the spot for three minutes, sixty heel digs in sixty seconds, thirty knee lifts in thirty seconds, two sets of shoulder rolls and ten knee bends. The NHS says it should take at least six minutes, and states plainly that the purpose is to prevent injury and make your workouts more effective.

The stretches only turn up on the second page, the one about after the session. Two pages from the same organisation, two entirely different sets of content. That is effectively the answer to the question in the headline.

Why long holds before a session cost you

Behm and colleagues surveyed the evidence in a 2016 systematic review in Applied Physiology, Nutrition, and Metabolism. The averages, when performance was tested immediately after stretching, looked like this:

But the average hides the important part. The review found a clear dose-response: static stretches held for under 60 seconds per muscle group produced −1.1 per cent, while 60 seconds or more produced −4.6 per cent. That is a fourfold difference.

So the problem is not the time of day. The problem is somebody holding each stretch for a minute and a half. An ordinary set of holds lasting 15 to 20 seconds sits well below the point where the effect becomes measurable in practice.

The trial that tested it properly

In 2018 the same research group ran the experiment those conditions called for. Writing in Medicine & Science in Sports & Exercise, Blazevich and colleagues had twenty male team sport athletes complete a comprehensive warm-up four times, in random order. After a low-intensity start they did either a 5-second static stretch, a 30-second static stretch split into three holds of ten, five repetitions of dynamic stretching per muscle group, or no stretching at all. The stretches covered seven lower body regions and two upper body regions. The warm-up then continued with test-specific practice building to maximum intensity, after which they were tested on sprinting, vertical jumping and change of direction.

Result: no effect of stretch condition on any of the tests.

The study's second finding is at least as useful. Before the trial, 18 of the 20 participants picked dynamic stretching as the condition most likely to improve performance, and 15 of 20 picked no stretching as least likely. Immediately before testing they rated the no-stretch day as less effective, 4.0 on a ten-point scale, against 5.3 to 6.4 for the three stretching conditions. Those ratings bore no relationship to how they actually performed.

It is an honest finding about all of us. The stretching felt like part of the preparation. The measurable part of the preparation was the warm-up.

What to do instead

Warm up with movement in the range you are about to use. Five to ten minutes of brisk walking or very easy jogging, then leg swings, walking lunges and a few strides if the session includes speed. There is a ready-made version in the guide to starting running after 40. If a stiff calf or hip needs loosening before you begin, loosen it. Just not in holds lasting a minute.

After the session: this is where stretching belongs

The NHS cool-down routine is five stretches held for 15 to 20 seconds each, taking about five minutes in total: buttock, hamstring, inner thigh, calf and front of thigh. The stated purpose is to gradually relax, improve flexibility and slow your heart rate.

Two things make the post-session slot suitable. The muscles are warm, which for most people is the difference between an unpleasant stretch and a pleasant one. And the performance cost no longer exists, because the session is over. The one rule that applies before a workout is simply irrelevant afterwards.

What it does not do matters just as much. Stretching after a session is not a cure for muscle soreness. That question is covered in detail in the guide to muscle soreness, which sets out the Cochrane numbers. The short version: half a point on a 100-point scale for stretching before, about one point for stretching after. The authors' own conclusion is that none of it is clinically important.

What stretching actually gives you

After everything that does not work, it is easy to conclude stretching is pointless. That would be wrong. It does exactly one thing, it does it reliably, and that one thing is worth something on its own.

In the short term: mobility that lasts under half an hour

Behm and colleagues reviewed 47 studies with 110 effect sizes in 2023 in Sports Medicine Open. A single bout of stretching produced a small improvement in range of motion. The subgroup findings are the interesting part: no measurable difference between stretching techniques, between intensities, between trained and untrained participants, or between women and men. The effect was robust but modest. According to the 2016 review, it typically lasts less than thirty minutes.

That explains a common disappointment. Stretch once and measure the next day and everything is back. Mobility behaves like fitness: what counts is what you do repeatedly.

In the long term: this is where the real gain sits

One of the largest syntheses arrived in 2026 in Sports Medicine Open. Oba and colleagues reviewed 79 randomised trials with 3,287 participants in total, comparing regular static stretching against not stretching. The result was a moderate improvement in lower-limb flexibility, Hedges' g 0.851 with a confidence interval of 0.710 to 0.991.

The finding that matters most for anyone reading this page is in the subgroups. The authors tested whether the effect was moderated by age, sex, training status, how flexible participants were to begin with, and which muscle group was stretched. None of them made a measurable difference. The age groups compared ranged from children and adolescents to older adults, with middle-aged participants as their own category.

Translated: you have roughly as much to gain from mobility work at 55 as you did at 25. That is unusual in this kind of research, where the response to training often dulls with the years.

What stretching does not do

Three claims that survive despite the evidence saying otherwise. Two of them belong to other pages in the guide, so they are summarised here with a link to the numbers.

The 2016 review is more ambivalent on this point, which is worth stating plainly. Its results section says that static stretching and PNF showed no clear effect on either all-cause or overuse injuries, and that there was no data at all for dynamic stretching. Its own conclusion nonetheless points the other way: the authors recommend stretching inside a warm-up that also contains dynamic activity, explicitly for reducing muscle injuries. That recommendation therefore does not rest on the injury numbers in the same abstract. The Lauersen meta-analysis is the one carrying the figures.

This is the single most important practical consequence of the whole page. If you want to reduce your risk of getting hurt, stretching is the wrong tool. Strength training is the one with numbers behind it, and a sensible spread of load across the week comes next.

The dose that actually has a source

Most stretching advice gives you a number without saying where it came from. Here are the two that can be traced.

The two fit together. Sixty seconds in total can be three holds of twenty, or two of thirty. The dose is the sum per muscle group, not the individual hold. And note the contrast with the warm-up section: the same sixty seconds that is the long-term target is the line you do not want to cross immediately before a session where you need to perform. After the session there is no such line.

Mobility is not in the recommendation for adults aged 18 to 64

Here is something rarely said out loud. Sweden's national public health agency recommends three things for adults aged 18 to 64: aerobic activity of at least 150 to 300 minutes a week at moderate intensity or 75 to 150 minutes at vigorous intensity, muscle-strengthening activity on at least two days a week covering all the body's major muscle groups, and limiting time spent sitting. Mobility is not mentioned.

The word only appears in the recommendation for those aged 65 and over, and in a different context: to preserve physical function and prevent falls, older adults should supplement everyday movement with physical activity that combines balance, strength and mobility on three or more days a week.

The conclusion to draw is not that mobility work is worthless. It is that mobility work does not compete for the same slot in the week. If you have three hours a week to move, most of it should go to cardio and strength, because that is where the recommendation sits. Stretching is the five minutes afterwards, not a separate session that pushes something else out.

How to do it in a busy week

This is the version for someone who trains early, commutes and does not have a spare half hour anywhere.

• Before the session: move for five to ten minutes. Brisk walking, easy jogging, leg swings, lunges. No holds longer than half a minute.
• After the session: five minutes, three to five stretches. Calf, hamstring, quadriceps, hip flexor and glute cover most of what a runner or a commuter needs.
• Aim for 60 seconds total per muscle group, two days a week. Not every session. Two days is the dose.
• Breathe out into the stretch and stay there. Do not bounce. It should feel like a pull, never like pain.
• Stretch what is stiff, not what is easy. Most people stretch their hamstrings because that stretch is comfortable, and skip the hip flexor that is actually short after a day at a desk.
• If you do not have five minutes, take two. Two stretches done consistently beat five stretches done once a month. Mobility works like everything else: the dose that happens is the one that wins.
• If you are stiff after a day at a desk, the answer is usually to stand up more often, not to stretch longer in the evening. Swedish national guidance says long periods of sitting should be broken up and replaced with some form of physical activity.

If you have a specific stiff calf or pain under the foot, there are dedicated and more detailed stretching protocols. They sit in the guide to plantar fasciitis, which covers the plantar fascia and calf stretches as treatment rather than as general mobility.

When to seek medical advice

Stretching is self-care for a normally stiff body, not a treatment for aching joints. On its own stretching page, the NHS notes that anyone who has an injury or a medical condition, or who is pregnant, may need to adapt some of the stretches, and should talk to a physiotherapist or healthcare professional about which ones are right for them. In the safety notes to its exercise videos, on the warm-up page, the NHS adds that you should stop immediately if you feel any pain or become unwell.

Swedish health service guidance on joint pain says the following. Contact a health centre if you have recurring or long-lasting joint pain. Contact a health centre or an on-call clinic immediately if any of these apply:

• You develop a fever or feel unwell at the same time as a joint suddenly swells.
• You get severe pain, swelling and redness in a larger joint, such as the knee, ankle, shoulder or elbow.
• Your child or someone close to you will not put weight on a leg, or will not move an arm or a leg.

If the health centre or on-call clinic is closed, the guidance is to go to an emergency department. In the UK, contact your GP, or use 111 online or by phone if you are unsure how urgent something is.

Beyond that: a stretch that produces pain radiating down the leg, numbness or pins and needles is not a stretching sensation. That is something else and should be assessed rather than stretched further.

This text provides general health information. It does not replace an assessment by a healthcare professional. It does not diagnose anything.

Frequently asked questions

Should you stretch before or after a workout?

Before the session, warm up, which means move. The NHS warm-up routine contains no stretches at all: it is marching on the spot, heel digs, knee lifts, shoulder rolls and knee bends, and it is meant to take at least six minutes. The stretches appear in the NHS cool-down routine instead, held for 15 to 20 seconds each over about five minutes. If you are only going to do one of the two, do it after the session. That is where it costs you nothing and where you are already warm.

Does static stretching before a workout ruin your strength?

Only if it is long. In the systematic review by Behm and colleagues in 2016, the average performance change after static stretching was minus 3.7 per cent, but the spread is the interesting part: stretches held under 60 seconds per muscle group produced minus 1.1 per cent, while 60 seconds or more produced minus 4.6 per cent. So the rule is about duration, not about the time of day. Testing also happened on average 3 to 5 minutes after stretching, and in the studies that added dynamic activity afterwards, no clear effect was seen at all.

How long should you hold a stretch?

The NHS cool-down routine says 15 to 20 seconds per stretch. The American College of Sports Medicine recommends a total of 60 seconds per exercise for each major muscle-tendon group, on at least two days a week, which can be three holds of 20 seconds. Stay under 60 seconds per muscle group if you are stretching immediately before a session where you need to perform. After the session, duration no longer matters for performance, because the session is over.

Does stretching prevent injuries?

No, not according to the evidence. In a meta-analysis of 25 randomised trials with 26,610 participants, stretching came out at a risk ratio of 0.963 with a confidence interval crossing 1, meaning no demonstrated effect. Strength training in the same analysis sat at 0.315. The 2016 review by Behm and colleagues similarly found no clear effect of static stretching on all-cause or overuse injuries. If you want to lower your injury risk, strength training is the intervention with numbers behind it.

Does stretching help with muscle soreness?

Barely measurably. The Cochrane review found that stretching before exercise reduced next-day soreness by an average of half a point on a 100-point scale. Stretching afterwards gave about one point. The two estimates rest on three and four studies respectively. The authors conclude that stretching, whether before, after, or both, does not produce clinically important reductions in delayed onset muscle soreness in healthy adults. Stretch if you enjoy it. Just do not treat it as a cure. The full walkthrough is in the guide to muscle soreness.

Do you get stiffer with age, and do you need to stretch more after 40?

Joint cartilage does become stiffer with age, which Swedish health service guidance describes as part of ageing. The good news is that your response to stretching does not appear to fade. In a meta-analysis of 79 randomised trials with 3,287 participants, regular static stretching produced a moderate improvement in lower-limb flexibility (Hedges' g 0.851) and the effect was not measurably moderated by age, sex, training status, starting flexibility or which muscle group was stretched. So you have roughly as much to gain at 55 as you did at 25. The authors also note that variation between the studies was substantial.

Sources for this page: NHS How to warm up before exercising and How to stretch after exercising for the warm-up and cool-down routines and for the safety note; Behm, Blazevich, Kay and McHugh 2016 in Applied Physiology, Nutrition, and Metabolism (PMID 26642915) for the acute effects and the dose-response; Blazevich and colleagues 2018 in Medicine & Science in Sports & Exercise (PMID 29300214) for the trial of stretching inside a full warm-up; Behm and colleagues 2023 in Sports Medicine Open (PMID 37962709) for the acute range of motion effect; Oba and colleagues 2026 in Sports Medicine Open (PMID 42435098) for the long-term flexibility effect and the age analysis; Lauersen, Bertelsen and Andersen 2014 in the British Journal of Sports Medicine (PMID 24100287) for the injury numbers; Herbert, de Noronha and Kamper 2011 in the Cochrane Database of Systematic Reviews (PMID 21735398) for muscle soreness; the American College of Sports Medicine position stand 2011 in Medicine & Science in Sports & Exercise (PMID 21694556) for the dose; the Public Health Agency of Sweden's physical activity guidelines for the recommendations for adults and for those aged 65 and over; 1177, the Swedish national health guide, on joint pain for the care advice and for the age-related change in joint cartilage.