Mobility for runners is a smaller subject than the search volume suggests. Running does not ask for much range. It asks for a modest range, produced correctly, several thousand times an hour, on legs that are getting tired. That is a different problem from the one a stretching routine is built to solve, and most of the advice written for runners never notices the difference.
This page sets out what the stride costs at each joint, what happened when stretching was tested against injury in large randomised trials, and the well-known result that argues the more flexible runners are the less economical ones.
The short answer
- Running is small range, enormous repetition. Nothing in the stride approaches end range.
- The joints that matter are the ankle, the hip in extension, and the calf and hamstring tolerating volume.
- Stretching has not been shown to prevent running injuries. Strength training has, for injury in general.
- One cross-sectional study found the less flexible distance runners were the more economical.
- Two of Limber's five instruments read what running uses: the knee-to-wall and the toe touch.
- The app has no runner programme and cannot see that you ran.
What running actually asks
A running stride moves each joint through a fraction of its available travel. The hip swings perhaps 40 degrees front to back at an easy pace, less than half of what a hip can do. The knee never locks and never folds. The ankle passes through maybe 10 to 15 degrees of dorsiflexion as the shin travels over the planted foot, against the 38 degrees a full squat has been measured to take.
So the range demand is low. What is high is the count. At 170 steps a minute an hour of running is roughly ten thousand landings, each one loading the calf and the hamstring at speed, each one asking the hip to extend behind you and the ankle to give way and then give it back.
That reframes the problem. A runner rarely runs out of range in the way a lifter does at the bottom of a squat. A runner runs out of tolerance: the tissue that has to absorb ten thousand landings stops absorbing them comfortably. Range work does not obviously address that, and the trials below say so.
Mobility for runners, by joint
Four readings are worth having, and only three of them can be taken honestly at home.
The ankle. Dorsiflexion is the one range in running with a plausible mechanical case behind it, because a short ankle changes what everything above it does. The knee-to-wall test is the measurement: big toe at the wall, heel down, drive the knee forward until it touches, then walk the foot back until the knee will not reach without the heel lifting. Measure wall to toe, both sides. On Limber's scale 2 cm reads 15, 6 cm reads 55, 10 cm reads 85, with straight-line interpolation between.
The calf. The same test reads it, because a calf that will not lengthen is what usually stops the shin travelling. Calf tightness covers the distinction between the two heads.
The hamstring. The toe touch reads it in five bands: 10, 35, 60, 75, 95. A fold reads the whole back of the body rather than one muscle, and the page says so.
The hip in extension. This is the range running most obviously uses and the one hardest to measure. The hip flexor test is a Thomas-style reading, and Vigotsky and colleagues showed in PeerJ in 2016 that it is not a valid measure of hip extension unless pelvic tilt is controlled. Grade: solid, and inconvenient. Take it if you like, on the same surface each time, and treat a small change as noise.
The finding that goes against us
Craib and colleagues, in Medicine & Science in Sports & Exercise in 1996, measured flexibility and running economy in sub-elite male distance runners. The less flexible runners were the more economical ones. Tighter hips and tighter calves went with using less oxygen at a given pace. The proposed explanation is elastic return: a stiffer tendon stores and returns more of the energy of each landing, so less has to be paid for by muscle.
That result is printed here at full strength because it is inconvenient for a mobility app and it is the first thing a runner asking about stretching deserves to know.
Grade: weak. It is cross-sectional, the sample is small, it is one population of trained men, and it shows an association rather than the effect of a training intervention. Nobody has taken a group of runners, stretched them for a year and measured their economy going down.
The correct reading is narrow. There is no evidence that becoming more flexible makes you a faster runner, and there is one credible finding pointing the other way. Range you need for the movement is worth having. Range beyond that is a goal, not a performance lever.
Injury: strength, not stretching
The prevention evidence has been tested at a scale this subject rarely reaches, and it did not go the way the category sells it.
Lauersen, Bertelsen and Andersen pooled the randomised injury-prevention trials in the British Journal of Sports Medicine in 2014. Strength training reduced sports injuries to under a third and overuse injuries to about half. Stretching showed no significant protective effect. Pope and colleagues had randomised 1,538 army recruits to stretch or not stretch before exercise, in Medicine & Science in Sports & Exercise in 2000, and found no reduction in injury. Herbert and Gabriel reached the same conclusion in the BMJ in 2002.
Grade: reasonably solid and consistent. Large randomised trials over two decades, pointing one way.
Timmins and colleagues add the specific case. Following elite footballers prospectively in the British Journal of Sports Medicine in 2016, they found the modifiable hamstring-injury risk factors were short biceps femoris fascicles and weak eccentric knee flexors — not flexibility. The lever was architecture and strength.
Two honest caveats, after the finding and not before it. Most of those trials tested stretching immediately around a session rather than months of sustained range work, which is a different intervention nobody has properly run. And "injury" pools a hamstring tear with shin pain, which blunts everything. Does stretching prevent injury is the long version.
If you are here because running keeps hurting you, the evidence points at loading, not lengthening. Pain that persists, wakes you at night, radiates down a limb, comes with numbness or weakness, or followed a specific incident is a matter for a clinician who can examine you rather than for a routine — see when to see a physio. Nothing here diagnoses anything.
Does the ankle reading predict anything
Slightly, in a neighbouring sport, and not in the way you would hope.
Backman and Danielson followed junior elite basketball players for a year in the American Journal of Sports Medicine in 2011 and found low ankle dorsiflexion associated with developing patellar tendinopathy. Malliaras, Cook and Kent had reported reduced dorsiflexion in volleyball players with patellar tendinopathy in the Journal of Science and Medicine in Sport in 2006. Both are about a low absolute reading, not about a left-right gap, and neither is running.
Grade: moderate for the association, absent for the intervention. No trial has taken restricted ankles, raised the range and shown fewer injuries afterwards. Radford and colleagues' review in the British Journal of Sports Medicine in 2006 found static calf stretching buys a few degrees of dorsiflexion, which the review itself called of questionable clinical importance.
And screening does not rescue it. Moran and colleagues found in the same journal in 2017 that movement-screen composites predict injury poorly, and Bahr argued in 2016 that this is a general property of screening rather than a fixable flaw.
What Limber does, and does not
Thirty-four of the 56 moves in the catalogue carry a running tag. Nothing in the build reads that tag. It is parsed into the engine's model and never consulted during selection. That is a criticism of our own data rather than a feature, and it is printed here so that nobody infers a runner mode from the library filter.
There is no runner programme. Ten tracks are drawn — Full-Body Foundations, Posture and Desk-Undo, Hips Unlocked, Shoulders and Overhead, Lower Back Care, Hamstrings, Neck and Upper Back, Ankles and Feet, Morning Wake-Up, Evening Downshift. None of them is for runners, and there are no sport packs.
Limber cannot see your run. Health data goes one way: the app writes a finished session and never reads one back. The engine holds a designed mapping — a run points at hamstrings, calves and ankles, and hip flexors, inside an 18-hour window — and that mapping has never run on a real workout, because nothing feeds it. It is a drawn rule with no input, not a capability.
What works today: you pick the areas and the minutes, and the engine builds from your readings, your equipment and any movements you have flagged, naming the reason for each pick. Post-run stretches is the routine, and the whole cluster sits under mobility for training.
Where to start
Take the knee-to-wall on both sides and the toe touch once. Write the date beside them. Then do the loading work the injury literature actually supports, keep the range work for after the run or its own slot, and re-measure in six weeks.
One warning about the ankle number. One centimetre is worth 5 to 10 points on Limber's scale, and Powden, Hoch and Hoch reported in Manual Therapy in 2015 that the test's minimal detectable change is around one to two centimetres. A small difference between two readings is the instrument, not you.
Questions
Do runners need to stretch?
Not for the reasons usually given. Stretching has not been shown to reduce injuries in randomised trials, including one of 1,538 army recruits, and it has never been shown to make anyone faster. It is worth doing if you want more range, if a position is somewhere you would like to be able to get to, or because you find it pleasant after a run. Those are honest reasons and they are enough.
Should I stretch before a run?
Not with long static holds immediately before you start. The measured acute cost to strength and power appears around 60 seconds of holding per muscle group and is trivial below it, so a brief hold is harmless and a long one before a session of intervals is not ideal. Move through the ranges you are about to use instead, and save the holds for afterwards.
Does being flexible make me a better runner?
There is no evidence that it does, and one finding points the other way: Craib and colleagues reported in 1996 that less flexible sub-elite distance runners were more economical, probably through elastic return. That study is small, cross-sectional and one population, so it does not prove the reverse either. Range for the movement, yes. Range beyond it is a goal in its own right.
Which ankle range do I need for running?
A stride uses roughly 10 to 15 degrees of dorsiflexion, far less than a deep squat. The knee-to-wall gives you a centimetre reading rather than a degree one, and the useful thing is comparing two of your own readings over time rather than hitting a number. Remember that a centimetre moves the score 5 to 10 points while the test's own detectable change is a centimetre or two.
Can the app build a session around my run?
No. Health is write-only in Limber: it records a finished session and never reads a workout back, so the app has no way to know you ran. The engine contains a designed rule that would lead with hamstrings, calves and hip flexors after a run, and that rule has never had an input. You choose the areas and the time, and the engine gives its reason for each move.
My calves hurt after every run. What should I do?
Not treat this page as the answer. Calf pain that persists, that arrived suddenly, that comes with swelling or with numbness, or that is worse at rest belongs with a clinician who can examine you. If it is ordinary post-run tightness, the honest expectation from the literature is that stretching will do very little for the soreness and a few degrees for the range, and that stronger calves are better supported than longer ones.