Do Running Shoes Prevent Injuries? What the Research Says

Running shoes are the most heavily marketed injury-prevention device in sport. Every year brings new foams, new plates, new geometries, and a fresh set of claims about what each one will save you from. Over the same forty years, running injury rates have stayed stubbornly flat at somewhere between a third and a half of runners per year. Those two facts sit together awkwardly, and they are the honest starting point for any answer to do running shoes prevent injuries.

The short version: shoes change how running feels, they demonstrably change how fast you can race, and they have never been shown to reliably change whether you get hurt. That is not a reason to run in bad shoes. It is a reason to stop treating the shoe wall as your injury plan.

Do running shoes prevent injuries?

On the current evidence, no, not in the way the marketing implies. No shoe category has been shown to reduce overall running injury risk across runners in a way that replicates. Shoes are worth choosing well for comfort and for performance, but the injury-prevention claim attached to shoe categories is not supported.

The clearest test of that claim was the foot-type prescription model, the one every running shop still uses: read the arch, sort the runner into neutral, stability or motion control, hand over the matching shoe. It is intuitive, it is decades old, and it has been tested properly.

The military ran the cleanest version of the experiment, because they can randomise thousands of recruits and then count exactly who breaks. Across US Army, Air Force and Marine Corps basic training, recruits had their foot shape assessed and were either given shoes matched to their arch type or given the same shoe as everybody else. Assigning shoes by foot shape made little difference to injury risk in men or women. A separate Air Force trial reached the same conclusion.

A British Journal of Sports Medicine review had asked the underlying question a few years earlier and answered it in the title: is your prescription of distance running shoes evidence-based? The finding was that it was not.

And the sign the whole system is built on turns out to be a weak predictor in its own right. A one-year prospective study of 927 novice runners, all in the same neutral shoe, found that foot pronation was not associated with higher injury risk. We unpack that one in the truth about overpronation.

The honest counter-evidence

Here is the part most articles on this topic leave out, because it complicates a clean story.

A well-designed Luxembourg trial gave 372 recreational runners either a standard shoe or the motion control version of the same model, blinded both participants and assessors, and followed them for six months. Overall injury risk was lower in the motion control group, with a hazard ratio of 0.55. A secondary analysis found the effect concentrated in runners with pronated feet, who had fewer pronation-related injuries in the motion control shoe. The original trial is here.

So the picture is not “shoes never matter.” It is that a single good trial found a benefit that the larger body of work, including much bigger military randomisations, has not reproduced. One study is a reason to keep an open mind, not a reason to buy. If the effect were as large and as general as shoe marketing implies, forty years of trials would not be this equivocal.

What about cushioning?

Also unclear, and interestingly conditional. A double-blind trial of 247 runners in soft versus hard midsoles found no difference in overall injury risk between them. A much larger follow-up in 848 recreational runners found the answer depended on body mass, with the cushioning effect running in different directions for lighter and heavier runners.

Read those two together and the message is not “cushioning is useless.” It is that the effect of cushioning is small, conditional on the individual, and nowhere near the size that a wall of shoes sorted by stack height implies.

Why shoes change less than you would expect

There is a mechanical explanation for all of this, and it is the most useful idea in the whole field.

Nigg and colleagues proposed the preferred movement path: across study after study, changing someone’s shoe, insert or orthotic changes their skeletal motion only slightly. The nervous system appears to defend an individual movement pattern and will work around whatever you put under the foot to stay on it. Put a wedge under someone’s arch and their muscles quietly adjust until the leg is doing roughly what it was doing before.

Their second idea is the comfort filter: runners are reasonably good at intuitively selecting a shoe that suits them, using perceived comfort as the signal. Which produces a genuinely practical rule, and it is close to the opposite of how shoes are usually sold.

Your feet are not a problem the shoe has to solve. Comfort is not a soft, unscientific criterion. It is the best single signal you have.

Where shoes really do matter

Two places, and they are worth being clear about because the sceptical case gets overstated in the other direction.

Race performance. This one is unambiguous. When researchers tested a carbon-plated prototype against established marathon racing shoes, it lowered the metabolic cost of running by about 4% on average, and every one of the eighteen runners tested improved. Super shoes are real, they are large, and they are the single best-evidenced piece of running equipment ever made. Note what that is evidence of: it is a performance effect, not an injury effect.

Comfort and tolerance. A shoe you find comfortable is a shoe you will run in consistently, and consistency is what actually builds durability. If a shoe rubs, pinches, or leaves your calves wrecked, that is real information. Change it.

One more practical finding worth knowing: runners who rotated between multiple shoe models over a 22-week period had a lower injury risk than single-shoe runners in an observational cohort. It is not a randomised trial, so hold it loosely, but varying the load a little is at least a plausible mechanism and rotation costs you nothing beyond owning two pairs.

How to actually choose a running shoe

1. Run in it. Not a treadmill jog in the shop, ideally. Comfort under fatigue is different from comfort in the first ninety seconds.
2. Pick the comfortable one. Not the one that matches your arch print. Not the one the video analysis recommended. The comfortable one.
3. Ignore the category names. Neutral, stability and motion control are marketing tiers, not clinical classifications, and they are not standardised between brands.
4. Rotate two pairs if your budget allows, and keep a heavier trainer for easy volume.
5. Change nothing dramatically at once. A big drop in heel-to-toe offset, or a jump into minimalist shoes, moves load to the calf and Achilles quickly. Transition over weeks.
6. If you have pain, see a physio, not a shoe wall. Persistent pain is a diagnosis question and no shoe answers it.

What to spend the attention on instead

If shoes are a comfort decision rather than an injury decision, the injury decision has to be made somewhere else. The two places with real evidence behind them are load and strength.

Load. The single most consistent predictor of running injury is a sudden increase in training. Novice runners who progressed weekly distance by more than 30% in a week took on more distance-related injuries than those who kept increases under 10%.

Strength. Pooled randomised trials show strength training cutting overuse injuries roughly in half, with a dose-response relationship. Nothing in footwear research comes close to that effect size. Strength training for runners covers the specifics.

The frustrating thing about this comparison is that shoes are easy and pleasant to buy, while a strength habit is neither. That asymmetry, not the evidence, is why the shoe conversation dominates.

Keep it honest

None of this means your shoes are wrong or that you have been conned. It means the shoe is doing a smaller and more comfort-shaped job than the marketing says, and that the effort you have been putting into choosing one has a much higher-yield home.

If you want to know where that effort should go, the useful question is not what your feet do inside your shoes. It is which side of your body is giving out first, and which muscle group is behind it. That is what running gait analysis is actually for, and it is what RunGait does from a short phone video: track your joints on-device, compare your left side to your right, report each signal at the confidence it deserves, and turn the strong ones into strength work. No shoe recommendation at the end of it, because we do not have one worth giving.