Deep divefrom Episode 11Training8 November 202617 min

Does Stretching Before Sport Really Slow You Down?

A deep dive into one paper: what was known before, what the researchers did, what the numbers say and what they don't.

The paper: Acute effects of six real-world warm-up and stretching strategies on strength, power and sprinting (SportRxiv preprint)

MP3

Chapters

Also linked

Source doctoral thesis (Charles Darwin University)

Transcript

The script the two AI voices read from. The times are worked out from the script, not heard, so a line can start a few seconds early or late.

Read the whole episode

Cold open

Host 1You heard the trailer — this is the whole story. Welcome to the Hastrup Review of Science, where artificial intelligence reads real science, so you don't have to pretend you understood the paper.

Host 2Today is one paper, a preprint posted in October on the sports science server SportRxiv, by a single author, Anthony Boutagy. It asks whether the stretching people actually do before sport, short holds inside a normal warm-up, makes them weaker, slower or less springy.

Host 1Before we start, what does preprint mean, and should that change how much I trust it?

Host 2A preprint is the author's own manuscript, shared before other scientists have reviewed it for a journal. It has been submitted, but nobody independent has checked it yet, so treat everything today as a careful first draft.

Host 1And if you had to give me the whole paper in one honest sentence, what would it be?

Host 2Sixty active young adults tried six common ways of getting ready, and almost nothing changed their jump, strength or sprint. The one signal worth a second look is that one kind of stretching on its own made ten metre sprints a touch slower, and adding a proper warm-up after it seemed to cancel that out.

Why this question exists

Host 1Why did anyone start studying stretching before sport in the first place?

Host 2For most of the twentieth century, stretching was simply part of getting ready, in school gyms and professional dressing rooms alike. As the paper puts it, people believed it lowered injury risk, increased range of motion, which is how far a joint can move, and improved performance.

Host 1So when did scientists start to doubt it?

Host 2From the late nineteen nineties, a wave of laboratory studies reported that a bout of stretching, especially static stretching and a type called contract-relax stretching, reduced maximal force, power and jump height, with effects lasting up to two hours. On the strength of that, influential position statements, for example from the American College of Sports Medicine in twenty eleven, took static stretching out of the warm-up whenever strength or power mattered.

Host 1You said static and contract-relax, so what is the actual difference between the kinds of stretching?

Host 2Static stretching is the familiar one: you ease a muscle to a long position and hold it still. Contract-relax stretching, which scientists call proprioceptive neuromuscular facilitation, adds a twist, you stretch, push against a partner for a few seconds, relax, and then go a little further.

Host 1And then there's dynamic stretching, which I think is the swinging-your-legs kind?

Host 2Close enough. In this paper it is called active mobility, controlled movements like walking lunges that take the muscles through their range rather than holding them still.

What we already knew

Host 1So the advice changed, but what was wrong with the studies that changed it?

Host 2The doses were far beyond real life. Many of the studies that found harm used fifteen to sixty minutes of stretching on a single muscle group, while surveys of strength coaches put a typical hold at only twelve to eighteen seconds.

Host 1Fifteen minutes on one muscle sounds absurd, so how did the testing itself differ from real sport?

Host 2Two ways. People were usually tested straight after stretching, and the stretching was done on its own, not inside a full warm-up with jogging and practice of the actual movement.

Host 1And once people tested realistic amounts, what did they find?

Host 2Mostly small, inconsistent or absent effects. In the wider picture of the field, the reviews since then point to a rough line at about sixty seconds per muscle, below which static stretching seems to do little harm, especially when it sits inside a full warm-up.

Host 1Give me a feel for the size of the harm in those older studies, is it large or small?

Host 2Small on average. The paper cites one pooling of one hundred and four studies that found strength down about five percent and power down about two percent, but with stretches lasting roughly one and a half to five minutes per muscle.

The open question

Host 1If the field mostly agrees already, what was still left to find out?

Host 2Few studies had compared several warm-up strategies side by side on the same set of tests, and combining contract-relax stretching with practice sets had not been tested directly. Samples were also usually under twenty people and mostly men.

Host 1Why would it matter that most of the volunteers were men?

Host 2Women's muscles and tendons are on average more compliant, more stretchy, and seem to need a bigger dose before they loosen. So in principle stretching could affect men and women differently, and you can't see that if almost everyone is male.

Host 1What did the author actually expect to find going in?

Host 2The opposite of what he found. He predicted that static and contract-relax stretching would make people weaker and slower, and that mobility drills and warm-up sets would make them better.

What the study did

Host 1Who were the volunteers in this study?

Host 2Sixty healthy adults, thirty two men and twenty eight women, mostly exercise science students and personal trainers, all training four to six days a week. They were randomly split into six groups of ten.

Host 1And what did each of the six groups do to get ready?

Host 2One group did nothing, they simply tested, rested three to five minutes and tested again. The static group held seven stretches for hips, thighs and calves, fifteen seconds at a time, four times, so one minute per muscle and about twelve minutes in all, always just short of discomfort.

Host 1How did the contract-relax group's routine work in practice?

Host 2Same muscles, but at the end of each stretch they pushed against resistance at about a fifth of their maximum for seven seconds, relaxed, and moved further, four rounds per muscle, roughly ten minutes. The mobility group did five dynamic drills with names like the inchworm and the scorpion, five slow, controlled repetitions each.

Host 1And what is a specific warm-up set, in plain words?

Host 2It means practising the test itself at lower effort: three rounds at roughly forty, sixty and eighty percent of full effort, then two minutes of rest. The sixth group did the contract-relax stretches, two minutes rest, then those same practice sets, then two more minutes before testing.

Host 1What exactly were they being tested on, before and after?

Host 2Three things. A ten metre sprint from a standing start, timed with light gates, a jump straight up from standing, and a strength machine called an isokinetic dynamometer.

Host 1That last one is a mouthful, so what does that machine actually do?

Host 2You sit strapped in and kick your lower leg out against an arm that moves at a fixed speed, no matter how hard you push. It records the peak turning force at the knee, called torque, the total work across the repetitions, and how much you tire, at three speeds, slow, medium and fast.

Host 1How soon after warming up did the second round of testing start?

Host 2About two minutes after, and the whole battery took about ten minutes in a random order. So depending on the order, the sprint came anywhere from roughly two to twelve minutes after the warm-up ended, which matters, because a small effect could fade in that time.

Host 1How did they turn all of that into a verdict?

Host 2For each person they took the after score minus the before score, the change, and compared those changes between groups. They did that for twelve measures one at a time, and then once more with all measures together while also accounting for sex.

Host 1Can ten people per group really detect anything at all?

Host 2Only big effects. The author's own calculation says the design could reliably spot differences that explain about a fifth of the variation between people, so small but real effects could easily slip through unnoticed.

What they found

Host 1So what happened to jumping after all six routines?

Host 2Nothing. Jump height, time in the air and time on the ground did not differ between groups, including after static and contract-relax stretching.

Host 1And on the strength machine, did any of the routines change it?

Host 2Not at the slow or the fast speed. At the medium speed, two measures came out significant, but in odd directions: peak torque rose after contract-relax stretching by about twelve newton metres, rose after static stretching by about six newton metres, and fell slightly after the combined routine.

Host 1Wait, so stretching made people stronger at that one speed?

Host 2Not in a way the author believes. The pattern only appeared at one of three speeds, it pointed different ways for different measures, and no routine differed from doing nothing, so he reads it as chance among many comparisons.

Host 1Then the sprint must be the interesting part, what changed there?

Host 2Sprint time was the clearest result out of the twelve. On average, ten metre times got slower by about five hundredths of a second after contract-relax stretching alone and six hundredths after the mobility drills, and faster by fourteen hundredths after contract-relax plus practice sets.

Host 1Is five hundredths of a second over ten metres a lot?

Host 2It is small. For scale, the paper says this sprint test's ordinary trial to trial wobble is about four hundredths of a second, so the contract-relax slowing is roughly the size of the noise for one person, though here it is an average over ten.

Host 1And the fourteen hundredths improvement, is that solid?

Host 2It is the biggest average change, but the spread in that group was large, more than twenty hundredths either way. So some people improved a lot and others very little.

Host 1So if I have this right, contract-relax stretching alone made you slower than doing nothing?

Host 2Not quite, and this is the subtle part. Two standard ways of checking which pairs of groups differ disagreed: one said the combined routine beat contract-relax and mobility, the other said contract-relax was slower than doing nothing and slower than practice sets.

Host 1Why would two statistical checks on the same numbers disagree like that?

Host 2Because the groups varied very differently in how spread out their results were, and the two methods handle that differently. The upshot, in the paper's own words, is that no strategy differed from doing nothing under both methods.

Host 1With twelve separate tests, isn't one significant result almost bound to turn up by luck?

Host 2That is exactly the worry, and the author addresses it. With a stricter cut-off that corrects for twelve tests, only the sprint result survives, the medium speed total work sits on the edge, and the peak torque result drops out.

Host 1What did the analysis with everything together show?

Host 2Once sex was in the model, the warm-up routine had no significant overall effect, while sex did. A further analysis that tried to tell the six groups apart from their results found them bunched closely together.

Host 1Does that mean men and women responded differently to stretching?

Host 2No, there was no sign of that. Men and women differed in their changes overall, but the routines did not work differently by sex, and with groups this lopsided that is hard to judge anyway.

A word from our sponsor

Host 2This episode is brought to you by Stiffwell, the spray starch for people. One coat before training, and nothing on you will stretch for up to two hours.

Host 1I use it before every sprint. My hamstrings have not moved since the spring.

Host 2Stiffwell was tested on ten volunteers, randomised with a table. Change score: zero, plus or minus nothing.

Host 1I spray my evenings now as well. They no longer stretch into the night, they simply end.

Host 2Not suitable for elastic bands, accordions or the summer. Stiffwell has held the horizon exactly where it was since two thousand and seven.

Why it matters

Host 1So what is the main lesson for someone who stretches before a game?

Host 2Short stretching, about a minute or less per muscle and held short of discomfort, inside a warm-up, did not meaningfully hurt strength, power or sprinting here. That fits what the wider field has concluded over the past decade or so.

Host 1And what is the principle underneath that, why would a short stretch be harmless?

Host 2The paper's reading is that brief, gentle stretching changes the muscle and tendon very little, and that the extra range people feel comes mostly from tolerating the stretch better. But the author is clear that he did not measure that, so it is an explanation, not a finding.

Host 1Who should change what they do because of this?

Host 2Mainly coaches and people who play sport. The paper's advice is to keep practising the real movement, run-throughs, practice jumps, building sets, and layer any stretching around that rather than replacing it with stretching.

Host 1And the contract-relax result, is there anything practical in it?

Host 2A cautious one. If you do partner contract-relax stretching right before maximal sprinting, follow it with progressively faster practice, because on its own it was the least favourable option here.

Host 1What about long stretching sessions, should those be dropped?

Host 2The paper suggests keeping long holds, two minutes or more per muscle, for separate flexibility sessions rather than the minutes before a big effort. That comes from the wider literature it cites, not from this study, which never tested holds that long.

Host 1Does any of this say whether stretching prevents injuries?

Host 2No, injuries were not measured at all. This paper is only about performance in the minutes after warming up.

Caveats

Host 1What are the biggest weaknesses I should keep in mind?

Host 2The groups were not balanced. The static stretching group averaged almost thirty three years old against nineteen to twenty four elsewhere, and was nine men and one woman, while the practice set group was eight women and two men.

Host 1How much could that imbalance really distort the results?

Host 2It could matter, because sex had a clear overall effect in the analysis. Comparing changes rather than raw scores helps, but the author says confounding by age and sex cannot be ruled out.

Host 1You mentioned the data are old, how old are we talking?

Host 2Collected in the mid two thousands, with ethics approval in twenty oh seven, and first written up in the author's doctoral thesis. The study was not registered in advance, nobody was blinded, and the volunteers and tester knew who did what.

Host 1And how were the statistics redone so many years later?

Host 2The individual raw data were lost, so only group summaries from the original statistics files survived. The pair-by-pair comparisons were rebuilt from those, with help from an artificial intelligence tool, the author says it was Claude from Anthropic, and he reviewed the computations and takes responsibility for them.

Host 1Is there anything else that limits who this applies to?

Host 2The volunteers were fit trainers and sport students who may already have had views about stretching, and the stretch may have been a weaker stimulus for them. Range of motion and muscle stiffness were never measured, and the practice set groups rehearsed the actual tests, so some of their benefit is simply practice.

Host 1So what can't we conclude from this, put plainly?

Host 2We can't say any routine reliably helps or hurts, that the contract-relax sprint effect is real, or that it applies to older adults, beginners, or dancers and gymnasts who stretch far harder. The author himself calls the sprint result hypothesis generating, a lead to test, not a result.

What to watch next

Host 1What would a study that settles the sprint question look like?

Host 2The paper asks for a properly sized trial with equal numbers of men and women, focused on contract-relax stretching before sprinting with and without practice afterwards. It also asks for measures of muscle stiffness and range of motion alongside the performance tests.

Host 1Is there a bigger trend in the field worth following?

Host 2Yes, in the wider field the open questions have moved from whether short stretching hurts to dose and intensity: how long, how hard, and what comes after. The recent reviews this paper cites keep landing on the same rough answer, about a minute per muscle, inside a warm-up.

Host 1So the short version for my Sunday game is what?

Host 2Warm up by doing a gentle version of the thing you are about to do, and a minute of easy stretching per muscle along the way appears harmless. If you do partner stretching, don't make it the last thing before you sprint.

Sign-off

Host 1That's the deep dive for this week. Every weekend this same feed carries the roundup, the week's new papers in one short episode. Thanks for listening to the Hastrup Review of Science.