Hamstring stretches: the one settled dose

Hamstring stretches are the most studied stretch in the literature. Not because the hamstring matters more than the rest of you, but because it is easy to measure: one joint, one plane, a person lying on a plinth and a protractor. Decades of that have left a dose which can be stated here with more confidence than anywhere else in the body.

So this page gives the numbers. It also prints, at full strength and before any caveat, the two findings that argue the whole exercise is aimed at the wrong variable.

The short answer

• Thirty seconds a hold. Sixty did not beat it in the trial that tested sixty.
• One hold a day matched three at the same length.
• Across a week, total time is what tracks the gain — more sessions rather than longer holds.
• Six weeks of it raises range with no measurable change in the muscle or the tendon.
• For hamstring injury, the modifiable factors in the best prospective work were strength and muscle architecture, not flexibility.
• Limber holds its hamstring positions for 40 to 45 seconds: above the tested floor, inside the window the trials covered.

Hamstring stretches, by the dose

Bandy and Irion, in Physical Therapy in 1994, took 57 people with limited hamstring range and gave them 15, 30 or 60 seconds of static stretching, five days a week, for six weeks. Thirty seconds beat fifteen. Sixty seconds did not beat thirty.

Three years later, in the same journal, Bandy, Irion and Briggler put the frequency question to 93 people: one 30-second hold a day against three a day. One matched three.

Thomas and colleagues pooled the chronic stretching trials in the International Journal of Sports Medicine in 2018 and found that what tracked the gain was total weekly stretching time, not the shape of any single session. That reconciles the other two. If you cannot get to a stretch every day, the days you do get to it are where a longer or repeated hold earns its place.

Grade: solid, and narrow. Young healthy adults, one muscle group, six-week windows, range read as a knee-extension angle. It is the best-supported stretching dose in existence, and it was established on the easiest case anyone could have picked. Do not carry that confidence across to a shoulder or a spine.

The working reading: 30 seconds is the floor, most days beats one long session, and past about a minute in a position you are spending time rather than buying range.

The nine positions in the app

Limber's catalogue holds 56 moves across ten body areas. Nine are filed to the hamstrings as their primary area — one short of the hips, which have ten, and level with the shoulders and the upper back.

Three things are worth reading off that list.

Four of the nine are dynamic. In this app a dynamic move is one repetition per breath cycle, six to twelve of them, and it belongs at the front of a session as a warm-up. The long holds belong at the back. Sitting down cold and hauling on a strap is the version most people do and the version the trials did not test.

Four are taken per side — M008, M025, M026 and M029. A fold gives one number for both legs and hides any difference between them. If your two hamstrings disagree, only the single-leg positions will show you.

Two carry the PNF flag — M008 and M029, both strap positions, because a strap gives you something to push against. That protocol is a hold, then six seconds of contraction at about 40 per cent effort, two seconds of release, then 15 to 20 seconds deeper, for two or three rounds. It never belongs in a warm-up.

One caveat on that, because the flag is not the whole gate. The engine also requires the move itself to sit at level 2 or above, and M008 is a level 1 move, so it is flagged and can never actually receive the protocol. Of the two hamstring positions here, only M029 can. Rest day mobility prints the full gate.

The rest is ordinary. A mild pull along the back of the thigh, never a sharp one behind the knee, and a breath out longer than the breath in.

What changes, and what does not

Konrad and Tilp, in Clinical Biomechanics in 2014, ran six weeks of stretching, confirmed the range gain, and then measured the muscle and the tendon. Neither had changed in any way they could detect.

That result is repeated often enough to be the default explanation. What improves first is not the length of the tissue but how much stretch your nervous system will permit before it stops you. Magnusson and colleagues framed it that way in the Journal of Physiology in 1996, and Weppler and Magnusson set out the full case in Physical Therapy in 2010.

Grade: reasonably solid. It is not a reason to stop stretching. It is a reason to stop believing that a longer muscle is what you bought.

Against us, at full strength

Two findings sit badly with a page about hamstring flexibility. They go here rather than in a footnote.

Timmins and colleagues, in the British Journal of Sports Medicine in 2016, followed elite footballers prospectively and found the modifiable hamstring-injury risk factors were short biceps femoris fascicles and weak eccentric knee flexors — not flexibility. The levers were muscle architecture and strength. How far the hamstring stretched did not appear.

Lauersen, Bertelsen and Andersen, in the same journal in 2014, pooled the randomised injury-prevention trials and found that strength training cut injuries to under a third, while stretching showed no significant effect. That is the comparison in one sentence, and it does not favour stretching.

So if injury is what brought you here, the evidence points at loaded eccentric work, not at holding a fold. Stretch because you want the range, because a position is somewhere you would like to be able to get to, or because it is pleasant. Those are honest reasons. Injury prevention is not one this literature will support.

Hamstring pain that arrived suddenly during a sprint, pain that persists, that wakes you at night, that radiates into a limb, that comes with numbness or weakness, or that followed a fall is a matter for a clinician who can examine you. Nothing here diagnoses anything and nothing here is treatment.

What the toe touch can say

The toe touch test is the reading. Five bands, one number: more than 20 cm from the floor reads 10, fingertips to the shins reads 35, fingertips to the floor 60, knuckles to the floor 75, palms flat on the floor 95.

In the app that single reading feeds the hamstring area in full, and half of the lower-back area, with the deep squat supplying the other half. The hamstring score you see is the toe touch, restated. Nothing else in the battery reads a hamstring at all.

Two limits follow, and they are worth more to you than another position.

The bands are 15 to 25 points apart. The smallest change the test can express is a whole band. Six weeks of good work that carries you from the middle of a band to the top of it shows up as nothing.

A fold is not a hamstring test. It is the whole back of the body and the spine at once, and how to tell if your hamstrings are tight is the page that pulls them apart.

The Range Score files that hamstring reading into the posterior group, alongside the glutes and the lower back, and averages five groups. Four of the five can be surveyed today. The fifth, the spine, is fed only by a rotation test that is not built in the app or here, and the score reports the gap rather than filling it.

Where to start

Take the toe touch, write the date beside it, and stretch for six weeks before taking it again. Thirty seconds or more per position, most days, sides recorded separately where the position allows it. Then you are comparing two numbers of your own, which is the only comparison worth anything.

If the fold itself is the goal, how to touch your toes is the specific progression. If the front split is what the hamstring is being trained for, an honest splits timeline answers in six-week training blocks and a wide range of months rather than a date, because no trial has followed adults to a split. If the hamstring turns out not to be your limit, hip mobility exercises covers the joint above it, and the rest of the cluster sits under hips and legs.

Questions

How long should I hold a hamstring stretch?

Thirty seconds is the supported floor. Bandy and Irion found that 30 seconds beat 15 and matched 60 over six weeks in people with limited hamstring range, and nothing since has overturned it. Limber's hamstring holds default to 40 or 45 seconds, which sits above the tested minimum without pretending that longer buys more.

How often should I stretch my hamstrings?

Most days, at whatever hold length you will actually do. One 30-second hold a day matched three a day in Bandy, Irion and Briggler's follow-up, and the pooled evidence says total weekly time is what tracks the gain. Frequency is the lever you have, not duration.

Do hamstring stretches prevent injury?

The evidence says no. Pooled randomised trials found stretching had no significant effect on injury rates while strength training cut them to under a third, and prospective work in elite footballers found the modifiable hamstring risk factors were muscle architecture and eccentric strength rather than flexibility. Stretch for the range if the range is what you want.

Does stretching make a hamstring longer?

Not in any way six weeks of it has been shown to produce. When researchers stretched people for six weeks and then examined the muscle and the tendon, the range had risen and the tissue structure had not changed detectably. What moves first appears to be how much stretch you will tolerate.

Which hamstring stretch is best?

No trial ranks one against another, so the honest answer is the one you will do and can hold without straining. Limber carries nine, from a standing fold to a supine strap pull, and the more useful split is not which position but whether it is dynamic — those go at the start of a session — or a long hold, which goes at the end.

Why has my toe touch score not moved?

Possibly because the test cannot see the change. The five bands are 15 to 25 points apart, so real progress inside a band reads as nothing, and going from the middle of a band to its top is invisible. It is also possible the hamstring was never your limit, since a fold reads the lower back, the calves and the pelvis at the same time.