John Davis

Written by John Davis


How to find the right running shoe

Unless you’ve been living under a rock the past few years, you’ve probably heard something about the barefoot running/“minimalism” debate.  Spearheaded by Christopher McDougall’s bestselling Born to Run, the barefoot enthusiasts contend that modern running shoes are ineffective at preventing injuries and furthermore, are harmful to your feet and lower legs.  This claim has attracted a significant amount of pushback from podiatrists, sports orthopedists, and of course, the running shoe industry.  While claims of a massive conspiracy are downright loony, like many popular movements, there is a grain of truth to the claims made by McDougall in Born to Run and by other barefoot enthusiasts: running shoes don’t function like they’re supposed to.

Getting started

If you’ve ever been to a running store, the sales representative probably examined your feet, watched the way you walked, and recommended a particular shoe based on those factors.  Some studies (but not all!) have connected biomechanical factors like arch height and running gait with injury risk.  The different varieties of running shoes were designed to correct these issues.  The logic is as follows: people with low arches need supportive shoes, and people with high arches need cushioned shoes.  Supportive shoes are also purported to work best for people who pronate—that is, people whose ankles roll inward as they walk or run, as the stiff wedge of foam on the inside of the arch counteracts this inward rolling.  A proper shoe, in theory, would correct your gait and prevent injury.  But did this pan out in practice?

The research on finding the right running shoe

Until recently, we had to rely on the anecdotal evidence of runners, doctors, and podiatrists.  But how could they reconcile their model of supportive and cushioned shoes with the rising tide of “minimalists” who found that they were able to run injury-free in thin, unsupportive shoes? And the many examples of people who were not any better off in the “right” shoe?  Recently, two large-scale, randomized studies were undertaken to examine whether the “right” shoe has any effect on injury risk.  A randomized study is the “gold standard” of research: it’s the best tool we have for examining, for example, whether a certain drug treats a disease or not.  It removes all of the problems of anecdotal evidence.

Pronation and motion control shoes

A 2011 study1 at the University of British Columbia tested whether assigning running shoes based on pronation had any effect on injury risk in female runners.  The 81 subjects were classified by the scientists as having “neutral,” “pronated,” or “highly pronated” feet.  Within each group, one-third received a cushioning shoe (intended for “neutral” feet), one-third received a support shoe (intended for “pronated” feet), and one-third received a motion control shoe (intended for “highly pronated” feet).  So within each group, some people go the “right” shoe for their foot type, and others got the “wrong” one.  Then, all the subjects trained for a half-marathon, and the researchers monitored how many within each group and sub-group got injured.  The results were a bit puzzling: people with “pronated” and “neutral” feet tended to do slightly better with the wrong shoe, but in all cases the shoe with the worst record was the motion control shoe! These findings are a bit weak, as a group of 81 subjects is not quite large enough to draw strong conclusions.

Injury risk and shoe type

Perhaps assigning shoes based on arch height would provide better results.  A 2010 study 2 by the US Marine Corps did exactly this: a group of 1,400 recruits was divided into an experimental group and a control group.  All Marines in the control group received a “support” type shoe regardless of their foot type, while those in the experimental group were assigned shoes based on their arch height.  As in the previous study, three types of shoes were used: a motion control shoe, a stability shoe, and a cushioning shoe.  These shoes were assigned to subjects with low, medium, and high arches, respectively.  The recruits were followed for their 12-week basic training program.  This study found no significant difference in injury risk based on shoe type, even when other factors like previous physical activity were controlled.

But how can this be? Do all shoes have the same effects on your feet? Intuition tells us “no,” and it turns out that indeed different shoe conditions affect your feet in different ways.

Shoe inserts

This has actually been established for a while in the field of biomechanics; we have to look back to 2001 to find a study on shoe comfort and injuries. This study, conducted by esteemed University of Calgary researcher Benno Nigg and coworkers3, investigated whether various shoe inserts could reduce injuries and (more relevant to our interests) whether inserts that accomplished this were subjectively more comfortable to the recruits.  Some 206 military personnel were presented with a variety of shoe inserts and were instructed to choose the most comfortable one.  The subjects had very different feet, which responded differently to the various shoe inserts, and this resulted in the subjects selecting different inserts as the “most comfortable.”  Furthermore, the group as a whole experienced fewer injuries than a control group given a standard flat insert.  If comfortable shoe inserts reduce injury, it shouldn’t be too much of a stretch to assume that comfortable shoes reduce injury as well.

Shoe comfort and performance

What’s more, it appears that comfortable shoes may increase performance too! A study in 2009, again by Benno Nigg’s research group4, investigated whether more comfortable shoes led to better running economy—a measure of how much energy a particular runner needs to maintain a given pace.  Ten male runners evaluated five different pairs of shoes, ranking them from “most comfortable” to “least comfortable.”  The men subsequently underwent a treadmill test to determine their running economy in the most and least-comfortable shoes.  The result? A 0.7% increase in running economy in the most comfortable shoe vs. the least comfortable shoe.  This might not sound like much, but that’s well over a minute off your marathon—just by swapping out shoes! What’s more, this study also controlled for the weight of the shoes, which has been shown to affect performance, and it even obscured the appearance of the shoes from the subjects so flashy designs wouldn’t deceive them.  However, this study was quite small compared to the others we’ve seen, and all ten subjects happened to rank the same shoe as the “most comfortable,” so this result isn’t a sure bet.


Now, all of these studies have their limitations.  The first study we saw was only done on women.  The second and third study was done on soldiers and marines, who, though they run a lot, also do plenty of crawling, marching, climbing, and so on.  And the last study was rather small.  No good scientist bases his conclusions on just one or two studies, but the plurality of evidence available points towards comfort as the best metric for what shoe really is the “right” one for you.

How can you use this research to find the right running shoe

All of this leads us to a profound but rather unsurprising conclusion: wearing comfortable shoes are your best bet.  For some reason, nobody has bothered to replicate Benno Nigg’s study of shoe insert comfort and injury risk with shoes themselves, but I’d be floored if such a study found contradicting results.  From the studies we’ve reviewed today, it’s pretty clear that the “traditional” methods of assigning shoe types based on foot shape or walking gait are outdated, and that comfort is the best method to pick out shoes as of today.

So where does this leave you, the aspiring runner, when you are staring at dozens upon dozens of shoes mounted on the wall at your local running store?

Well, there’s no better way of finding the right shoe than to try them out! I recommend trying out a lot of shoes—perhaps six or eight pairs or more if you can’t find a pair that feels great on your feet.  Lace up a pair, walk around in them, and take them for a spin in the parking lot (any reputable running store should let you do this).  If they don’t feel phenomenal, try on another!

Much of the common wisdom about running shoes still holds: heavier runners ought to try on larger, more cushioned models first, whereas lightweights might consider opting for a thinner, sleeker model, and it’s never a bad idea to get a new pair of shoes after you’ve logged 300 or 400 miles on your previous pair.  But beware of sales representatives telling you that you’ll have to “break in” your shoes—your new shoes should feel like a second skin from the get-go.  If not, do not buy them!  And finally, if you’ve been running injury-free in the same shoe model for a long time, you’ve found a keeper.  Stick with what works!

Free Injury Prevention Course

Learn What is Causing Your Running Injuries and How to Fix It

Here’s what we’ve got for you

The scientific, underlying cause of your injuries and how to fix them.

The mistakes you're making in training that are leading to your injuries.

Case studies with specific examples of runners exactly like you and the strategy they implemented to get healthy


1. Ryan, M. B.; Valiant, G. A.; McDonald, K.; Taunton, J. E., The effect of three different levels of footwear stability on pain outcomes in women runners: a randomised control trial. British Journal of Sports Medicine 2011, 45 (9), 715-721.
2. Knapik, J. J.; Trone, D. W.; Swedler, D. I.; Villasenor, A.; Bullock, S. H.; Schmied, E.; Bockelman, T.; Han, P.; Jones, B. H., Injury Reduction Effectiveness of Assigning Running Shoes Based on Plantar Shape in Marine Corps Basic Training. The American Journal of Sports Medicine 2010, 38 (9), 1759-1767.
3. Mündermann, A.; Stefanyshyn, D.; Nigg, B., Relationship between footwear comfort of shoe inserts and anthropometric and sensory factors. Medicine & Science in Sports & Exercise 2001, 33 (11), 1939-1945.
4. Luo, G.; Stergiou, P.; Worobets, J.; Nigg, B.; Stefanyshyn, D., Improved footwear comfort reduces oxygen consumption during running. Footwear Science 2009, 1 (1), 25-29.

Connect with Jeff Gaudette on Google+

14 Responses on “How to find the right running shoe

  1. I disagree with this. I find a neutral cushioning shoe to be the most comfortable out of the box. Unfortunately the lack of stability created by being propped up on an inch of cushioning has my knees pissed off by mile 4. I think comfort is important, but it’s not the only factor which is how this article reads.

  2. Thanks for the feedback, Jason. I should point out that I don’t think picking the most comfortable shoe is 100% guaranteed to get you the best result—indeed, your experience proves otherwise! But if you aren’t sure what try first, the research indicates that comfort is your best bet for finding out the right shoe. You may need to adjust later, as you did, to how your body reacts to the shoe.

  3. I have to agree with Jason – the most comfortable shoes at the running store tend to be the ones that feel like I am walking on pillows. However, I’ve found that overly cushioned shoes have a negative affect on my knees and ankles.

  4. I went to the local running store, and they did the whole gait analysis thing, and gave me about 5 pairs of shoes to try on that were all comparable from the big shoe companies. I went with the most comfortable of the bunch (Saucony ProGrid Omni) and wore them for about 400 miles. When it came time to retire them, I bought the exact same paid on amazon for about 60% of the store price (I know you’re supposed to support the local store!). 400 miles later I went to reorder them online, and realized if I buy the same shoe, but last year’s model (or the year before), I can get them for about $30. That’s hard to argue with. But I always stay with the same shoe now that I’ve found it.

  5. On a separate note, do you have any suggestions for what shoes to wear when running in sand? I want to run my first Ultra in 2013, and there’s a local 50K, but it’s on the beach, so this year I’m going to start running on the beach. Any suggestions?

  6. two things to consider about sand:
    1) it is unstable
    2) it is softer than the road
    on account of both of these, you might consider a shoe somewhat thinner or firmer than you usually wear, since that will put you closer to the ground and you might be less affected by the instability on the sand. The cushiness of the sand, combined with a shoe that’d feel good on the road might feel TOO soft and too unstable.

    In any case I’d make an effort to go run on the sand in your normal shoes and see how they feel—too soft, too high off the ground, or just fine. Then go from there.

    (of course I’m imagining the kind of “soft” sand on the dry part of the beach—if you’ll be on the very hard and smooth “wet” sand on an ocean beach it might be different!

  7. Awesome… thanks! I watched a documentary on the Marathon De Sables, but they all wore covers on their shoes (gaiters?) to keep the sand out, so I couldn’t tell what they were running in.

  8. Thanks for taking the time to write this article and sharing it with us. Though I agree that the running shoe industry since the 1970’s has run a “prescription” system that has yet to be backed by any evidence, I see the jump to “forget all that and just go for the most comfortable shoe you find in the shop” a little too much of a swing in the opposite direction. I appreciate it is easy to grow tired of unbacked claims made by both running shoe companies and barefoot diehards, but I am not convinced the quoted trials provide any evidence that we should disregard analysis before selecting a trainer and just go for “comfort”. The problem is the quality of analysis and more often than not a decision made by just looking at what happens at foot level, which as far as I can see is what happened in the stated trials (unless any other informafion says otherwise).
    Full body gait analysis often reveals a correlation between internal or external rotation at hip or knee with level of pronation. Use of orthoses or a certainbbuild of trainer can modify level or pronation and in turn reduce inefficient rotation higher up in the chain. Though no evidence exists showing that “overpronation” is bad for you (nor that an ideal level of pronation even exists), our bodies are subject to the laws of physics. Given that running is all about moving forwards and working as much as possible with gravity as opposed to against it, I think we can agree excessive rotation in either direction is undesirable. If rotation is reduced by either trainer type or corrective exercise, there is potential for an increase in running efficiency and/or a decrease in injury rate. Just basing trainer choice on level of pronation makes little sense but taking movement of the rest of the body into consideration can help an educated decision on suitable trainer can be reached, a decision that may not be the same as the shoe that feels the most ¨comfortable”. The variance in the results seen in the trials reflects this.
    I am also not sure that suggesting a heavier runner will always need a more heavyweight cushioned shoe is entirely correct. Do most injuries occur in midstance or initial contact? The level of cushioning plays no effect in midstance. I suggest translation of impact force is more related to running form & efficiency as opposed to bodyweight.
    I look forward to any comments.

  9. Midstance is when the foot of the weight bearing leg is fully planted, and when the opposite knee of the leg in swing phase passes. It is at this point that peak joint torques & stresses are at their maximum and when I suggest most injuries occur, as opposed to at the moment of initial contact. And let’s remember, landing on your heel is fine as long as your knee is not fully extended. It’s not a question of what part of your foot strikes the ground first (heel/mid/forefoot), it’s a question of where it lands in relation to the hips.

    • I think it’s over simplistic to say that injuries occur only through mid stance. The stress is not the the same in all tissues uniformly at any given moment. For example, in a foot that has prolonged pronation into late stance will have greater loading forces on the tibialis post tendon because of its lengthened position and big eccentric forces placed upon it.
      I have to say that landing on different densities/sizes will change GRF patterns, but adaptations are complex and I think research has shown unpredictability in how these accommodations take place.
      I agree with your landing pattern comments however.
      Personally, I think comfort is important in shoe prescription and I tend to be less fussy these days about shoe suggestions unless the current shoe is obviously contributing to the issue. Ie I am individualistic in prescription rather than categorical.

    • So if I am ‘midstance’ on 3 inches of eva you believe it it has aboslutely no difference to ‘midstance’ on 3mm of eva. Seriously?
      You can be injured in any part of the gait cycle.
      I don’t see the relevance to your additional comments on heelstriking, you statement was about midstance.

  10. So if I am ‘midstance’ on 3 inches of eva you believe it it has aboslutely no difference to ‘midstance’ on 3mm of eva. Seriously?
    You can be injured in any part of the gait cycle.
    I don’t see the relevance to your additional comments on heelstriking, you statement was about midstance.

  11. Cushioning shoes are shoes that have little to no lateral support. These shoes are good for runners who do not need this support, and have neutral feet. Generally this type of shoe will be for the runner with a high arch. Instances where this type of shoe is not right is in a case where you are a pronator or an overpronator.-

Leave a Reply

Your email address will not be published. Required fields are marked *

Adding new comments is only available for RunnersConnect Insider members.

Already a member? Login here

Want to become an Insider for free? Register here