Choosing the Right Running Shoes

The standard advice, and its problem

Walk into most running stores and you’ll be put on a treadmill, filmed from behind, and told whether you overpronate. From that, you’ll be steered toward a shoe category: neutral, stability, or motion control.

It’s a coherent, plausible system, but the evidence hasn’t supported it for some time.

Nigg and colleagues reviewed four decades of running shoe research¹ and concluded that the two pillars of the traditional model, impact force modification and pronation control, had failed to produce consistent evidence of injury prevention. A focused review in Frontiers in Sports and Active Living² put it plainly: in the absence of strong evidence for specific shoe features, the practice of prescribing footwear based on foot type has been challenged for over a decade, though it remains common practice.

That last clause matters most. The advice has lasted because nothing replaced it, not because anyone validated it.

Pronation is normal

Pronation is the inward roll of the foot after landing. It’s a normal, necessary shock-absorbing mechanism that everyone has. The idea that pronating “too much” causes injury, and that a shoe correcting it prevents injury, is a two-step argument where neither step has held up well under testing.

Studies assigning runners to shoes by foot type have repeatedly failed to show the injury reduction the model predicts. In some cases, runners assigned shoes according to the conventional rules did worse than those assigned shoes at random.

This does not mean stability shoes are bad; plenty of people run happily in them. It means the diagnostic step, inferring your correct shoe category from your arch or your gait video, doesn’t do what it claims.

What the research suggests instead

Nigg’s group proposed two replacement ideas, and the practical one is the comfort filter.

The proposal is that a runner intuitively selects a comfortable shoe, and in doing so selects one that lets them stay in their preferred movement path, the way their limbs naturally want to move. A shoe that fights that path requires extra muscular work to resist, which plausibly increases fatigue and injury risk.

The practical upshot is unusually simple: comfort, assessed by you, at the point of trying them on, is the best available selection criterion. Arch height, a treadmill video and the category label on the box all rank below it.

A focused review for running medicine clinicians² landed in the same place. In the absence of a clearly supported paradigm, it recommends footwear that is lightweight, comfortable, and has minimal pronation control technology.

How to buy shoes

A short checklist for the store:

• Go late in the day. Feet swell over the course of a day and after running, so a shoe fitted to a morning foot can be tight by evening.
• Leave a thumb’s width at the front. Feet lengthen under load, and toes hitting the end on downhills produce bruised toenails within weeks. Most people need a running shoe half a size to a full size larger than their everyday shoe.
• Run in them rather than standing in them. Any decent running store will let you run, and standing tells you very little.
• Trust immediate comfort. If a shoe needs “breaking in”, it’s the wrong shoe. The comfort filter only works if you listen to it.
• Ignore the category label. Try neutral and stability shoes and see which feels better; the label is a marketing taxonomy, not a prescription.
• Buy the boring pair. Carbon-plated racing shoes are designed for racing at speed. For easy running as a beginner they’re expensive, less durable, and offer nothing you need.

What about drop, cushioning and minimalism?

Heel-to-toe drop, the height difference between heel and forefoot, has no established optimum. Lower drop shifts load toward the calf and Achilles, and higher drop shifts it toward the knee. Neither is better in general, but changing drop substantially changes where load goes, so switch gradually.

Cushioning has surprisingly weak evidence in either direction. Maximal cushioning has not been shown to reduce injury, and neither has minimal, so choose what feels good.

Minimalist and barefoot-style shoes do change loading, mostly moving stress from the knee to the foot and calf. The transition period is where people get hurt: several documented cases of metatarsal stress fractures came from switching too quickly. If you want to try them, take months, not weeks.

When to replace them

The usual guidance is 300–500 miles (500–800 km), which is a reasonable planning figure but not a rule. Midsole foam loses resilience with compression cycles, and shoes and runners vary a lot. In our own running, shoes have gone comfortably past the top of that range, including for heavier runners, where you’d expect the opposite. So treat the mileage as a cue to start paying attention rather than a date to replace.

More useful signals:

• New aches appearing with no change in training, which is often the first sign.
• Visible compression creasing through the midsole, not just outsole tread wear.
• The shoe feeling noticeably “dead” underfoot compared to when new.

Given what a decent pair costs, get as much out of them as you can. Retire them when those signals show up, not when the mileage total says you’re due.

Once you’re running regularly, it’s worth rotating two pairs. Foam recovers between runs, both pairs last longer in total, and varying the shoe slightly varies the loading pattern, which some evidence suggests is protective.

Do you need trail shoes?

Only if you’re running on soft or technical ground. Road shoes handle gravel paths, canal towpaths and packed trails perfectly well.

Trail shoes add grip on mud and wet rock, and some protection from sharp stones. In exchange they’re less comfortable on asphalt: aggressive lugs on hard ground feel like running on studs, and they wear down quickly.

If your running is a mix, buy road shoes first. Trail shoes become worth owning once you’re regularly on ground where you’re actively worrying about slipping.

Socks, and the unglamorous causes of misery

Two things cause far more early running discomfort than shoe choice, and both are cheap to fix.

The first is cotton socks. They hold moisture against the skin, and wet skin blisters. Any synthetic or merino running sock solves this, and it’s the single highest-return $12 in running.

The second is lacing. If your heel slips, use the extra eyelet at the top: thread the lace through to make a loop on each side, then cross the laces through the opposite loop before tying. This “heel lock” or “runner’s loop” holds the heel without needing to over-tighten across the top of the foot, which is what causes numbness and pins and needles in the toes.

If your toes go numb on runs, the answer is almost always lacing or shoe volume, not a foot problem.

One thing that matters more than the shoes

Worn-out or badly fitting shoes appear in the risk-factor lists for injuries like medial tibial stress syndrome, but their effect is small compared with training load.

No shoe compensates for increasing your mileage too fast. If you’re weighing an extra $75 on shoes against taking an extra month over your progression, the month is worth far more. See building weekly mileage safely.

Before you buy your next pair

Pronation-based shoe prescription is not supported by the evidence, even though it’s still standard practice in running stores. The best-supported criterion is your own comfort, assessed by running in the shoe. Shop late in the day, leave a thumb’s width at the toe, and ignore the category label.

Drop, cushioning and minimalism have no established optimum, but changing any of them substantially calls for a gradual transition. Replace shoes when new aches appear or the midsole is visibly compressed, not on a fixed schedule. And keep in mind that shoes matter far less than load progression, so spend your attention accordingly.

This article is general information, not medical advice.

References

1. Nigg BM et al. Running shoes and running injuries: mythbusting and a proposal for two new paradigms: ‘preferred movement path’ and ‘comfort filter’. British Journal of Sports Medicine, 2015. pubmed.ncbi.nlm.nih.gov
2. Agresta C et al. Running injury paradigms and their influence on footwear design features and runner assessment methods: a focused review to advance evidence-based practice for running medicine clinicians. Frontiers in Sports and Active Living, 2022. ncbi.nlm.nih.gov