Gait Analysis for Running Shoes: What the Shop Test Sees
The free gait analysis in a running shop is a shoe-fitting aid, not a screening. What those 30 seconds on the treadmill see, and how to get the useful version.
Almost everyone’s first experience of gait analysis for running shoes is the same: you walk into a running shop, someone points at a treadmill, you jog for thirty to sixty seconds while a camera films your ankles from behind, and a few minutes later you are looking at a slow-motion clip of your own heels with a member of staff drawing a line up the back of your leg. Then you are handed two or three boxes.
It feels like a diagnostic test. It is not, quite. It is a shoe-fitting aid with a diagnostic aesthetic, and understanding the difference tells you exactly how much weight to put on the result.
None of what follows is an attack on running shops. Good ones do something genuinely valuable, which is getting you into a shoe that fits your foot, in the right size, that you have actually run in before paying for it. That is worth the trip. The problem is the gap between what the treadmill test measures and what people walk out believing it measured.
What the shop test actually measures
Almost always one thing: rearfoot motion in the frontal plane. A camera behind you, at ankle height, watching how much your heel rolls inward after it lands. Some shops now run software that draws the angle automatically; most still eyeball it. Either way, the output is a category, usually neutral, mild overpronation, or severe overpronation, and the category maps onto a shelf.
Everything else about how you run is out of frame. Your hips, your pelvis, your trunk, your knee, what happens to any of it once you are tired, and, critically, whether your two legs are doing the same thing as each other.
Is a free gait analysis worth it?
It is worth it as a fitting service and not much as a screening. Go for the fit, the sizing and the chance to run in the shoe before buying. Do not go expecting to learn why your knee hurts, and do not treat the pronation verdict as a finding about your body.
There are five specific reasons the result carries less than it appears to.
1. It measures a weak predictor. The whole test rests on the idea that pronation predicts injury. A one-year prospective study of 927 novice runners, all in the same neutral shoe, found no association between foot pronation and injury risk. We go through that study in the truth about overpronation.
2. Sorting shoes by that measurement has been tested and did not work. Military trials that assigned footwear by foot shape and arch height across thousands of recruits found little difference in injury risk compared with giving everyone the same shoe. A review of the practice concluded that prescribing running shoes this way is not evidence based.
3. Sixty seconds on a treadmill is not your run. Treadmill running differs from overground running, the shop belt is usually short and set to a pace you did not choose, and you are wearing shoes you have never run in while a stranger watches. Most importantly you are fresh. Almost every meaningful breakdown in running form appears with fatigue, at mile eight or ten, not in the first minute.
4. The rear view is the least informative angle. The foot is the smallest, fastest-moving part of the body and it is partly hidden inside a shoe. It is the hardest thing in the frame to measure accurately and the least connected to anything you can train. Meanwhile the hip, which is large, slow, clearly visible and highly trainable, is not being filmed at all.
5. The output space is the inventory. This is not dishonesty, it is just structural. A test whose possible answers are all shoes will produce a shoe. If the true answer to your problem is “your right glute medius is weaker than your left” or “you added 40% to your mileage last month,” there is no shelf for that.
The shop test can only end in a shoe. That is not a criticism of shops. It is a reason not to ask them a question whose real answer is a strength plan.
What a gait analysis is genuinely good for
Take the same idea, point it at the parts of the body that matter, and it becomes useful. The signals that survive a phone camera and lead somewhere trainable are:
- Left versus right asymmetry. The most valuable output of any home analysis, for a technical reason: both your legs appear in the same clip at the same camera angle, so comparing them cancels most of the measurement error that makes absolute angles unreliable. See muscle imbalances in runners.
- Contralateral pelvic drop. The opposite hip dipping during stance. A direct read on lateral hip control. See pelvic drop and glute medius.
- Dynamic knee valgus. The knee travelling toward the midline under load. See dynamic knee valgus.
- Hip extension at toe-off. How much you are actually driving back. See reduced hip extension.
- Cadence and foot placement. The two things that genuinely respond to a deliberate change. See running cadence.
Every item on that list resolves to a muscle group you can train, which is the test of whether an analysis was worth doing.
Shop fitting versus gait screening
| Shop treadmill test | Gait screening | |
|---|---|---|
| What it measures | Rearfoot roll, one angle | Hip, pelvis, knee, cadence, left vs right |
| View | Behind, ankle height | Multiple angles including side-on |
| State | Fresh, 30 to 60 seconds | Ideally at the end of a real run |
| Output | A shoe category | A specific muscle group and a strength block |
| Repeatable | Once per purchase | Retest every 6 to 8 weeks |
| Cost | Free | Free at home, £60 to £150+ in a clinic |
The two are not competitors. Get your shoes fitted by people who fit shoes. Just do the screening separately, and do not confuse one for the other.
Getting the useful version at home
You need a phone and somewhere flat to run about 20 metres.
- Film at the end of a run, not the start. Twenty minutes in minimum. Fatigue is where the information lives.
- Shoot at 60fps or higher, phone braced or propped, camera at roughly hip height, landscape.
- Get two angles: directly behind or in front for the frontal plane signs, and side-on for hip extension and foot placement.
- Wear fitted clothing. Loose shorts hide the exact joints you are trying to see.
- Run past the camera at your normal easy pace. Do not perform. A stride you are consciously producing is not the stride you run 40 miles a week on.
- Read differences, not degrees. If your right hip drops noticeably more than your left, that is real. If a number says 7.4 degrees, that number carries five to fifteen degrees of camera error and is not worth the decimal point.
The full protocol is in running gait analysis at home.
When to pay for the clinical version
A proper lab or physio-led assessment is worth real money in three situations: you have pain that keeps coming back in the same place, you are returning from a significant injury and need to know if you are ready, or you are chasing a performance ceiling and have already done the obvious training work. In those cases you are buying a clinician’s judgement, not a camera, and the judgement is the expensive part.
For everything else, a phone and a habit of retesting gets you most of the way there for nothing.
Keep it honest
Gait analysis is screening, not diagnosis. It can tell you which side of your body is working harder and which muscle group is a plausible cause. It cannot tell you what is wrong with a painful knee, and anything that offers to is overreaching. If you have pain, see a physio.
What it can do, if you point it at the right joints, is answer a question the shop treadmill was never set up to ask: not which shoe should I buy, but which muscle should I train first. That is the question strength training for runners is built around, and it is what RunGait produces from a short phone video, on-device, with a confidence rating on every signal and a strength block instead of a shoe recommendation at the end.