Cadence and injuries: Once more, with science

Posted by Dave on July 12, 2011 | 9 Comments

ResearchBlogging.orgThe plan for this site has always been to supplement discussion of peer-reviewed science with other reasoned discussions based on anecdotes and experience. When I’m discussing the peer-reviewed research, I always include the Research Blogging icon you see at the top of this post. When I’m discussing books or other non-peer-reviewed sources, I don’t. That doesn’t mean the conclusions are wrong, just that they haven’t been systematically tested. But since some commenters expressed doubt about the conclusions of my post last week on cadence (based on the advice of legendary running coach Jack Daniels), I thought I’d see what I could find in the scientific literature on the subject.

So, is there scientific research to support the conclusion from the post that “you might be able to reduce your risk of injury associated with landing shock by increasing your cadence”—that is, the number of steps you take per minute?

I found a 2004 review article on the subject by Alan Hreljac. Hreljac examined all the work he could find on impact injuries to see if he could identify the most important risk factors. The key insight is this: Three separate studies found that runners on treadmills who landed hardest were more susceptible to injuries than those who generated less force when they land. A runner can incur up to five times her body weight in force each time her foot hits the ground. But runners vary in many different ways, and those whose running form limits this vertical force are significantly less likely to get injured.

How can you limit this force? A more recent study suggests that one way this can be achieved is by reducing stride length while maintaining the same speed. The only way to reduce stride length at a constant speed is to increase cadence, which is exactly what Daniels suggests. The researchers analyzed the stride of runners who decreased stride length—and increased their cadence—by ten percent and found a significantly decreased chance of injury. This isn’t a true empirical test, but the model the researchers used did find that likelihood of injury is reduced with a shorter stride. The other studies Hreljac mentions in his review support this model: They find fewer real-world injuries in runners with less vertical impact in each stride.

Is there a better way to decrease your risk of running injuries? Absolutely: Run less. But since this site is “Science-Based Running” and not “Science-Based Sitting Around,” that’s not exactly an option we take very seriously, is it?

Actually, it is: One key to effective training is to minimize the types of training that may lead to injury, and in future posts I’ll be discussing some ways to do that.

HRELJAC, A. (2004). Impact and Overuse Injuries in Runners Medicine & Science in Sports & Exercise, 845-849 DOI: 10.1249/01.MSS.0000126803.66636.DD


9 Responses to “Cadence and injuries: Once more, with science”

  1. Alex Hutchinson
    July 13th, 2011 @ 6:19 am

    Great stuff, Dave — really enjoying the blog. And glad to see you holding yourself (or being held by your commenters!) to high standards of scientific rigour. The Daniels 180 strides-per-minute claim is the epitome of anecdotal wisdom that has been assimilated by runners without critical analysis. It may be true, but it hasn’t been peer reviewed!

    That being said, there’s plenty of science in this area, stretching back decades. First of all, given your previous post on running economy: it’s pretty widely accepted that running economy has a U-shaped curve if you vary stride frequency, with the minimum at whatever frequency you’d “naturally” select at a given speed. Most people (including me) suspect that you can gradually change that set point by learning to run with a difference cadence — but initially, at least, you’ll be less efficient if you speed up or slow down your cadence at a given speed. The classic reference is Cavanagh and Williams from 1982 (

    But you’re more interested in cadence and injury. There have been lots of studies about stride length and shock attenuation (e.g. this one from 2002: Most recent is one published earlier this year by Bryan Heiderscheit at Wisconsin: He concluded that shortening stride length (which is equivalent to quickening cadence at any given speed) by 5-10% produces biomechanical changes equivalent to running barefoot — which isn’t surprising, given that one of the key benefits of running barefoot is that the lack of heel cushioning prevents you from overstriding. I asked Heiderscheit about 180 as a magic number, and he didn’t buy it: he suggests that beginners in particular usually overstride, but doesn’t see why people with different heights, limb lengths and joint ratios should all magically gravitate to the same cadence.

    One caveat on all of these biomechanical studies: showing that joint forces change is NOT the same as showing that injury rates will improve. That study hasn’t been done (and, for the most part, hasn’t been done for almost any intervention discussed in running, because of the difficulties in organizing and controlling a prospective trial with sufficient statistical power). That’s why the barefoot debate won’t be ending anytime soon: all the sides have evidence, but none of it measures what we’re actually interested in.

  2. Robin Judice
    July 13th, 2011 @ 11:07 am

    Your last sentence made me laugh out loud! Thanks for the post.

  3. Dave
    July 13th, 2011 @ 4:32 pm


    Thanks for the comment, and for the compliment on the blog. Anyone who hasn’t already added your blog Sweat Science to their reading list should do so right away!

    Your caveat is well taken, but some of the studies that Hreljac cites in his review article do attempt to go beyond merely measuring the forces involved and look at actual injury rates among people whose stride involves more or less vertical force, and the finding is that people with more vertical force are more likely to get injured.

    But there are still horrible confounds in all of this — it’s possible to reduce the vertical forces involved without shortening your stride.

    I agree that 180 probably isn’t a “magic number,” but it might be a minimum threshold for many people, and Daniels has actually looked at a lot of elite runners and rarely finds anyone with a lower cadence than 180 (though plenty have much faster cadences).

    One thing I wonder about, as a larger runner (6′ 1″ and 187 pounds), is whether it’s possible for someone my height to run efficiently at that cadence—after all, it’s pretty rare for an elite runner to be as big as I am.

  4. Alex Hutchinson
    July 13th, 2011 @ 11:22 pm

    “Daniels has actually looked at a lot of elite runners and rarely finds anyone with a lower cadence than 180…”

    Elite runners have a lot of traits in common beyond even body size — most fundamentally, they run very, very fast. Cadence increases with speed (see, e.g.,, so taking a specific cadence number from elite athletes and trying to apply it to someone running slower than, say, 6:00 miles is problematic.

  5. Alex Hutchinson
    July 13th, 2011 @ 11:31 pm

    “…the finding is that people with more vertical force are more likely to get injured.”

    Sorry to quibble endlessly here — its just an interesting topic that I find hard to resist! I’d just amend your statement above to: “the finding is that people who have been injured in the past tend to have more vertical forces.” None of those studies were (a) prospective or (b) intervention trials.

    The reason I think that’s relevant is because it’s the same logical leap that running shoe designers made in the 1980s. Retrospective studies showed that excessive pronation was linked to higher injury rates; biomechanical studies showed than motion control shoes reduced excessive pronation; ergo, motion control shoes must reduce injury. But now that we’re FINALLY doing prospective intervention studies (, it looks like (a) shoes that reduce pronation DON’T reduce injuries, and (b) they may even increase injury rates in some circumstances.

  6. Greg Lehman
    July 21st, 2011 @ 1:36 am


    Daniel’s 180 frequency rule is based on elite runners running relatively fast paces. I think that much of the magic of this number is strongly pace driven. If you look at Pavel Komi’s work you will see that sprinters exceed 240 steps/minute.

    I have filmed a number of elite and sub-elite runners (e.g. sub 2:33 marathon females) and do stride rate counts.

    For example, one elite female runner tested at the following paces and cadences:

    3:30 min/km pace = 190 steps,
    4:00 min/km pace = 182 steps
    5:00 min/km pace = 172 steps
    6:00 min/km pace = 168 steps

    The vast majority of runners (more than 80% if you look at average race times for the Goodlife Marathon 5 km results) don’t run 5 min/km.

    I think pace is a big factor and shoe horning all your recreational runners into 180 magic may not be wise.

    Great website by the way.

    All the best,


  7. MaiLinda
    April 9th, 2012 @ 4:05 pm

    Forget research studies. I have already found, as a newbie runner, that the HIGHER the cadence, the LESS my injuries hurt.

    Am recovering from shin splints (too much too soon – yaddie ya…. all that beginner stuff)….went on a run today (okay – more like a job) and thought my shins were gonna kill me. Then, I decided to pick up my cadence.

    Actually, I didn’t decide, Madonna decided it for me in a tune (LOL) and wouldn’t ya know – the pain not only subsided, but I found the running MORE COMFORTABLE.

    Less time having my feet / ankles support my body weight really helped me to “move” along.

    I was moving between 170-180 bpm versus my usual 155-165.

    Thanks for letting me know I”m not nuts :)

  8. Joe Warne
    August 22nd, 2012 @ 7:09 am

    Hi Guys,

    Alex, the U curve for cadence may not be quite as straightforward as we think. Take the example of a lot of barefoot/minimalist runners becoming more economical as an acute effect to changing footwear. While ofcourse this is also related to factors such as elastic restitution and decreased braking impulses, we commonly see that the immediate effects of increasing cadence without footwear can lead to improved economy, this definately needs more research, and I am currently looking into investigating self selected shod cadence vs self selected VFF cadence and the effects of each in the opposite shoe condition (using a metronome),to determine what kind of an influence cadence is actually having.

  9. cadence | 100 days 100 ways
    March 12th, 2013 @ 8:01 pm

    […] For more information on the topic of cadence I suggest you read this article. […]