How to choose running shoes: by use, by terrain and by the evidence
Why foot type should not drive your purchase, what cushioning, drop and weight mean according to the trials, which shoe suits each use and terrain, and how much carbon plates really give you.
Shoes are the only essential purchase for running and the one that carries the most myths: the gait analysis, the "overpronator" shoe, the plate that makes you a minute faster. This chapter separates what has trials behind it from what is shop habit, and ends with a short list of what to check before you pay.
Choose by comfort, not by foot type
Choosing shoes by foot type or pronation does not reduce injuries
In the largest trial, with more than 3,000 US Army recruits, half were given motion-control, stability or cushioned shoes according to the shape of their footprint and the other half all got a stability shoe: injury risk was practically the same (risk ratio 1.01 in men and 1.07 in women). In a cohort of 927 novice runners in a neutral shoe followed for a year, pronated feet were not injured more than neutral ones; if anything, they had slightly fewer injuries per 1,000 km. And in a trial with women runners, the motion-control shoe produced the most pain and the most missed days, pronated feet included.
Comfort is the best filter we have
After reviewing the evidence on impact and pronation, Benno Nigg’s group proposed two ideas to replace the old ones: every runner has a preferred movement path that the shoe should respect rather than correct, and perceived comfort is the practical way to find the shoe that respects it. It is a reasonable proposal, not a proven law, but it is a better criterion than the footprint test.
- 1
Try them in the afternoon or after a run
Feet swell over the day and during a run. Buying with cold morning feet is the most common way to end up a size too small.
- 2
Leave a thumb’s width at the front
About a centimetre between your longest toe and the end of the shoe. On descents and late in a long run the foot slides forward, and that is where toenails are lost.
- 3
Run in them, do not just walk
A couple of minutes on a treadmill or outside tell you more than ten laps of the shop. If you can, compare three or four models back to back.
- 4
Keep the one you do not notice
The right one disappears on your foot: no rubbing, no pressure points and no feeling that you will have to get used to it. "They will break in" is a poor reason to buy.
Cushioning, drop and weight: what the trials say
- Cushioning
- How much the midsole deforms on landing
- Drop
- Height difference between heel and forefoot
- Stack height
- Total thickness under the foot
- Weight
- Grams per shoe, in a reference size
- Stability
- Reinforcements to limit pronation
- Plate
- Stiff carbon or plastic sheet in the midsole
Softer is not always better: it depends on your body mass.
0 to 12 mm. It shifts load around, it does not change the total.
More stack, more protection and less ground feel.
It matters on race day; much less on easy runs.
Not shown to prevent injuries.
Improves economy; it is for racing.
More cushioning is associated with fewer injuries, especially in lighter runners
A randomised trial with 848 recreational runners handed out two versions of the same model, one soft and one hard. In the hard version injury risk was 52% higher. When split by body mass, the benefit of cushioning appeared only in lighter runners; in heavier runners there was no clear difference.
Drop does not change risk overall, but it does depend on how much you run
Another trial from the same group compared 10, 6 and 0 mm drops in otherwise identical shoes. Overall, drop did not change injury risk. Split by regularity, low drop was associated with fewer injuries in occasional runners and more in regular runners. In practice: if you already run often in a high-drop shoe, do not switch overnight.
Every 100 g per shoe costs about 1% more energy
Adding mass to the foot raises oxygen uptake by around 1% per 100 g per foot, a figure repeated across several studies. But running barefoot was not more economical than running in a light, cushioned shoe: cushioning also saves muscular work. Weight matters on race day; for easy running, comfort matters more.
Which shoe for which use
| Use | What to look for | What you do not need |
|---|---|---|
| Just starting | A comfortable daily trainer with medium or high cushioning | A plate, "just in case" stability, top-of-the-range |
| Daily easy runs | Durability, cushioning and not noticing it at 10 km | Lightness at any cost |
| Intervals and tempo | Something lighter and snappier than your daily trainer | A carbon plate for training |
| Road racing | Light, with a responsive foam; a plate if you are chasing a time | Wearing it for the first time on race day |
| Track | Spikes or a light racing shoe | A lot of stack height |
| Trail on dirt roads | Moderate lugs and some underfoot protection | Mud outsoles or a carbon plate |
| Technical trail or mud | Deep lugs, wet grip, a secure foothold | Very tall, unstable foams |
| Treadmill | Your usual daily trainer | A dedicated shoe |
If you are only going to own one pair
A comfortable, versatile daily trainer covers 90% of what a recreational runner does, intervals and races included. The second purchase, if it comes, is a faster shoe for racing or a trail shoe if you head for the hills.
Carbon plates: how much they give and to whom
They improve running economy, but with large variation between runners
The study that made them famous measured 4% lower energetic cost with Nike’s prototype versus two other racing shoes; it was funded by the brand. The later meta-analysis puts the average improvement at 2–3%. And in highly trained runners, the individual gain ranged from practically nothing to more than 5%: for some it changes the race, for others it adds almost nothing.
- They make sense if you race on the road and are chasing a time, from 10 km upwards.
- They do not make sense for easy running: they are expensive, lose performance sooner and take away work your foot does in a normal shoe.
- Try them in training before racing in them: individual variation is large, and how they feel late in a race cannot be guessed in the shop.
Trail: grip and protection
In the mountains the priorities change: first do not slip, second do not feel every stone. There are no controlled trials comparing trail shoes with one another, so what follows is practical judgement, not evidence. More on the terrain in trail and mountain.
- Lugs: 3 to 4 mm works on dirt roads and dry trails; mud and wet grass need 5 mm or more, spaced wider so they do not clog.
- Rubber: on wet rock the outsole compound matters more than the lug shape.
- Rock plate: a sheet under the forefoot that protects on stony ground; it reduces ground feel.
- Foothold: the foot must not slide around on descents; that is where ankles roll and toenails go.
- Drop and stack: lower gives stability on technical ground; higher gives protection over long distances.
Minimalist and barefoot: only with a slow transition
Switching quickly to a minimalist shoe increases stress on the bones of the foot
In a small trial, after a 10-week transition to five-toed shoes, 10 of 19 runners showed increased bone marrow oedema on MRI, significantly more than in the group that stayed in their usual shoes. And at equal mass, running barefoot saved no energy compared with a light, cushioned shoe.
If you want to try them, start with short stretches inside normal runs and build up over months, not weeks. Less shoe asks more of the foot and calf, just like the foot-strike change explained in technique and cadence.
How long they last and why to own two pairs
Rotating several pairs is associated with fewer injuries
In a follow-up of 264 runners over 22 weeks, those who alternated more than one pair had around 39% lower injury risk. It is an observational study —people with two pairs may also train differently— but the proposed explanation is reasonable: varying the shoe varies the load.
The 500 to 800 km lifespan figure is common practice, not a trial result. Better than the odometer are the signs: a creased or flattened midsole, an outsole worn down to the foam, or new niggles that disappear when you start a fresh pair. Injuries have more to do with load than with shoes, so before blaming them, review how you have increased your volume.
What to check before buying running shoes
Brand and model matter less than fit. This is what is worth checking before you pay.
- Comfortable from the first minuteThe criterion with the most support on the whole list.
- A thumb’s width at the frontBetter half a size up than half a size down, especially for trail and long distances.
- Last year’s versionModels are updated every year with minor changes, and the previous one usually drops a lot in price.
- A plate only if you raceFor easy running, a daily trainer lasts longer and costs half as much.
- Return policySome shops accept returns if you have only tried them at home or on a treadmill: check first.
Takeaways
- Choosing by foot type has not been shown to prevent injuries; comfort is the best filter.
- More cushioning is associated with fewer injuries, especially if you are light.
- If you run often in a high-drop shoe, do not jump straight to a low drop.
- Carbon plates give 2–3% on average, with large variation: for racing, not for easy runs.
- Rotate two pairs if you can, and replace shoes by the signs, not only by the mileage.
Sources for this chapter
- [1]Knapik JJ, Swedler DI, Grier TL, et al. (2009). Injury reduction effectiveness of selecting running shoes based on plantar shape. Journal of Strength and Conditioning Research. trial
- [2]Nielsen RO, Buist I, Parner ET, et al. (2014). Foot pronation is not associated with increased injury risk in novice runners wearing a neutral shoe: a 1-year prospective cohort study. British Journal of Sports Medicine. trial
- [3]Ryan MB, Valiant GA, McDonald K, Taunton JE (2011). The effect of three different levels of footwear stability on pain outcomes in women runners: a randomised control trial. British Journal of Sports Medicine. trial
- [4]Richards CE, Magin PJ, Callister R (2009). Is your prescription of distance running shoes evidence-based?. British Journal of Sports Medicine. review
- [5]Nigg BM, Baltich J, Hoerzer S, Enders H (2015). Running shoes and running injuries: mythbusting and a proposal for two new paradigms: 'preferred movement path' and 'comfort filter'. British Journal of Sports Medicine. review
- [6]Malisoux L, Delattre N, Urhausen A, Theisen D (2020). Shoe cushioning influences the running injury risk according to body mass: a randomized controlled trial involving 848 recreational runners. The American Journal of Sports Medicine. trial
- [7]Malisoux L, Chambon N, Urhausen A, Theisen D (2016). Influence of the heel-to-toe drop of standard cushioned running shoes on injury risk in leisure-time runners: a randomized controlled trial with 6-month follow-up. The American Journal of Sports Medicine. trial
- [8]Franz JR, Wierzbinski CM, Kram R (2012). Metabolic cost of running barefoot versus shod: is lighter better?. Medicine & Science in Sports & Exercise. trial
- [9]Hoogkamer W, Kipp S, Frank JH, et al. (2018). A comparison of the energetic cost of running in marathon racing shoes. Sports Medicine. trial
- [10]Kobayashi EN, Toledo RRF, Almeida MO, et al. (2026). Metabolic effects of carbon-plated running shoes: a systematic review and meta-analysis. Frontiers in Sports and Active Living. meta-analysis
- [11]Barnes KR, Kilding AE (2019). A randomized crossover study investigating the running economy of highly-trained male and female distance runners in marathon racing shoes versus track spikes. Sports Medicine. trial
- [12]Ridge ST, Johnson AW, Mitchell UH, et al. (2013). Foot bone marrow edema after a 10-wk transition to minimalist running shoes. Medicine & Science in Sports & Exercise. trial
- [13]Malisoux L, Ramesh J, Mann R, et al. (2015). Can parallel use of different running shoes decrease running-related injury risk?. Scandinavian Journal of Medicine & Science in Sports. trial