Sunday, July 8, 2012

Do All Elites Run at a 180 Cadence: None of the 5K Finalists at the US Olympic Trials Did

One of the most common myths I encounter in discussions about running form is that 180 steps per minute is some kind of magic number (Google “180 cadence” to see what I mean). It’s not, and I’ll be happy when this one gets put to rest. Much of the mythology about this number stems from a misinterpretation of what Jack Daniels reported in his book “Daniels’ Running Formula” regarding the cadence of elite runners at the 1984 Olympics. He did not observe that they all ran at a 180 cadence, he observed that they ran at a cadence of 180 steps/min or more. My emphasis on the “or more.”

We cover the topic of optimal running cadence for efficiency and reduction of joint loading extensively in Chapter 8 of Tread Lightly, but here I wanted to share some data that I compiled from a set of slow motion videos taken by physical therapist Jeff Moreno at the 2012 US Olympic Trials. Jeff obtained footage filmed at 210 frames per second from both the Men’s and Women’s 5000 meter finals (as well as the 1500, which I hope to examine in a separate post).

Here are the two videos:

Women's 5000 Meter Final

Men's 5000 Meter Final

Because I know the frame rates at which these videos were captured (210 fps), I can simply count the number of frames between various stride events (e.g., initial contact, toe off, etc) by importing the videos into Quicktime. A few simple calculations in Excel allows me to produce a bunch of data on stride kinematics, which are presented below. Runner order is the order in which they pass the camera at this point of the race, not the final finishing order – If I’ve made any ID errors please let me know! Contact time is for the first footstrike observed, swing time is the time from toe off of the first contacting foot to the next contact of the same foot. Cadence is determined by determining stride time and using that to extrapolate the number of strides that would be taken in 60 seconds, then multiplying that value times two to yield steps/minute (a stride is from the contact of one foot to the next contact of the same foot). All times are in fractions of a second.

Here are the data (Update 7/10/2012 – I initially made a dumb Excel error in my calculation of flight time – doesn’t affect any of the other numbers though – data are now correct):

Order Name Cadence Contact Time 1 Swing Time Flight Time
1 Bernard Lagat 195 0.143 0.471 0.167
2 Galen Rupp 187 0.138 0.505 0.176
3 Lopez Lomong 192 0.157 0.467 0.157
4 Andrew Bumbalough 202 0.148 0.448 0.143
5 Mo Trafeh 191 0.171 0.457 0.143
6 Benjamin True 192 0.162 0.462 0.148
7 Elliott Heath 205 0.162 0.424 0.138
8 Hassan Mead 191 0.171 0.457 0.143
9 Scott Bauhs 188 0.157 na 0.162
10 Ryan Hill 191 0.162 0.467 0.152
11 Trevor Dunbar 197 0.176 0.433 0.129
           
           
           
Order Name Cadence Contact Time 1 Swing Time Flight Time
1 Julia Lucas 202 0.152 0.443 0.152
2 Molly Huddle  197 0.167 0.443 0.138
3 Julie Culley  195 0.171 0.443 0.133
4 Abbey D'Agostino 189 0.167 0.467 0.143
5 Emily Infeld 192 0.162 0.462 0.157
6 Elizabeth Maloy 188 0.176 na 0.143
7 Kim Conley  197 0.171 0.438 0.124
8 Lisa Uhl 210 0.157 0.414 0.124

What you’ll note is that at least at this point of the race, none of these elites were running with a cadence of 180. In fact, the vast majority of them were running with a cadence above 190, with some pushing higher than 200! My hope is that data like this will put the 180 cadence myth to rest for good. I’ll say it again: 180 is not an “optimal number”!

[Note: A few people have commented to me that cadence may be on the high end for these runners because they are nearing the end of the race. Even if this is true, which we can’t tell from limited video footage, it further makes my point – cadence varies with speed (I will say that I have video from the Boston Marathon of elites with equally high cadence numbers). It’s quite possible that Lisa Uhl has a high turnover here because she’s making a move to pass. I’ve done a bit of experimentation with how my own cadence changes with speed here: http://www.runblogger.com/2011/09/running-speed-human-variability-and.html]

One interesting thing to note about these data is the overall similarity of the numbers between runners. Contact and swing times, for example, seem to fall in a fairly narrow range for most of the competitors, though there are a few notable individuals. Galen Rupp, who won both the 5K and 10K at the trials has the shortest contact time and longest swing time of any of the competitors. Unfortunately, it’s hard to know if these gait characteristics contributed to his success. Among the women, Julia Lucas had the shortest contact time at that point in the race – she was in the lead, but ran out of steam in the final lap. Lisa Uhl had the highest cadence of any runner, male or female. All interesting observations, but what they mean from a practical standpoint is hard for me to know given my lack of a coaching background or any attempt to research factors that contribute to elite performance (my interest is in trying to understand how shoes work and why recreational runners get hurt).

I’ll finish by saying that the value of the 180 number is that it’s higher than the cadence of most recreational runners, and a cadence significantly lower than 180 can indicate a problem with overstriding. But, holding fast to a single number for an optimal cadence makes no sense to me, and I prefer to think in terms of a range (say 170-190), or even just aiming to increase you cadence by 5-10% from baseline if you believe you are an overstrider.

24 comments:

  1. Great stuff here, Pete. I remember counting out my strides (not too few, not too many!). Hopefully, this will soon become a thing of the past.

    ReplyDelete
  2. I'm wondering about wether pace effects cadence? Without having checked it out myself it doesn't seem unreasonable to me. If so, your ordinary amateur runs nowhere near as fast as these guys and girls do on an ordinary run. If they were to slow down and run as fast as the average amateur then perhaps they would run round about 180 steps per minute, give or take?
    Or do we keep pretty much the same cadence regardless of the pace?
    I'm not trying to suggest that your wrong in saying that 180/min is not necessarily optimal, but I'm wondering how relevant the figures above are.

    What say you?

    ReplyDelete
    Replies
    1. Cadence definitely changes with speed - here are some data taken from myself: http://www.runblogger.com/2011/09/running-speed-human-variability-and.html

      The point I was making here is that people often say that 180 is optimal because that's what elites do based on what Daniels reported in his book based on his observations at the 1984 Olympics. I actually was interviewed by a major magazine and asked to confirm that 180 is optimal just last week! It's a widespread belief for some reason. The issue is that Daniels never said that all elites run at 180 exactly, and in fact most don't when you look at them in a race. And for the very reason that cadence can change significantly with pace, telling people to aim for 180 makes absolutely no sense.

      Delete
  3. Pete, great videos and data - thanks. I was going to mention what you've already noted. The videos are shot with 600m to run. From the split data off the results website we can see that the leading females are running at approximately 2:53/km pace past this point whereas they were running at 3:05/km pace earlier in the race. The men are running at about 2:35/km pace compared to 2:45/km pace mid-race.

    I'd like to know what the cadences are earlier in the race... whether their speed is adjusted by stride-length or cadence or a combination of both.

    I'd also like to know the cadences of these runners when they're running at 'mere mortal speeds' - for instance between 4:00 and 6:00 per km - the speeds that most of us run at. Do they 'lope' at all when jogging at these speeds or do they keep a relatively high cadence and just shorten their strides?

    ReplyDelete
    Replies
    1. All great questions - wish I had the same video from every lap!

      Unfortunately I wasn't there, just using the videos that Jeff Moreno shot, so working with what I have available.

      If I had to guess, there would be some variation in how each runner relies on stride length vs. cadence modification to increase speed.

      Delete
  4. Just ordered a metronome to ensure at least at 180. Now I'm wondering if I shouldn't bother? Thoughts?

    ReplyDelete
    Replies
    1. Depends entirely on what your current stride looks like. Cadence will vary by person and with speed. If you aren't a big overstrider I'd be hesitant to mess with it. What's your current cadence?

      Delete
  5. A cadence of 180 may not be the "optimal number" but I don't believe the majority of the general public has that fast of a cadence so in that regard it may be the minimal optimal number to strive for when you are a casual runner not an elite athlete like an olympian.

    ReplyDelete
    Replies
    1. I agree, but prefer to work up from the baseline exhibited by each individual runner rather than pointing to a specific number. Start by upping 5% and see if that helps, then maybe go to 10% if needed. taking someone with a cadence of 155 and getting them to jump right to 180 is a pretty big change.

      Delete
  6. My interest lies in injury prevention for the "standard" runner. I do have elite athletes I see and they tend to maintain a higher cadence. I don't manipulate it, but I do investigate the biomechanics.
    I find most of my clientele run between 6-12 miles/week. The others vary around 50-100 miles. These guys are required to run weekly for their jobs. The cadence for the majority are ranging from 155-170 with many complaints of injuries. This is when I work with the "magical number" of 180, granted I place a range on it and increase by 10% . Yes, they over stride, but it's difficult to get them to increase step rate... I utilize the metronome so they can hear it while running to set up a shorter stride. I also work on their deficits of strength.....

    ReplyDelete
    Replies
    1. "I place a range on it and increase by 10%"

      I think this is the wise approach - a fixed number with no real backing makes little sense. The goal is to reduce overstriding, and the amount of stride manipulation it takes to do that probably varies from person to person. I've listened to Blaise Dubois talk about this a few times and he recommends looking at baseline cadence and aiming for a target range from around 170-190 depending on the individual.

      Delete
  7. I use a GPS with cadence on all my runs. The faster I run, the higher my cadence. Most training runs it's 160-170, for races it's 180-190.

    ReplyDelete
  8. Speaking for me personally, and I am by no means elite, I average around 182 over the course of most runs. The key here being average. My actual cadence will vary a good deal between 175 and 205 depending on how fast I am running at any given moment. As a Good Form Running Instructor at the running store where I work, I bring up 180 in regards to the topic of cadence as a general guideline. I always tell them to work towards it by stepping up from their known baseline in increments. I also tell them that 180 should be seen more as an average and that their cadence will vary based on speed.

    ReplyDelete
    Replies
    1. I would also like to add that most people I work with are to become comfortable with a cadence above 170 within a matter of a couple of weeks.

      Delete
  9. Well, I've been timed at 180, so...

    I just recently saw a pic of me running the Naked Foot 5k in bare feet. I still looked like I was over striding a bit, which is why I ordered the meter. However, Barefoot Josh suggested that since I am still a newbie barefooter, my feet are reaching out a bit to test the waters of the terrain and that with time, that will change.

    ReplyDelete
    Replies
    1. I'd give it some time if you've just started out.

      Delete
  10. I realize all the natural and barefoot running "coaches" out there are trying to get their students to run at 180, but I don't think they say that is the fastest rate they should run do they? I know Ken Bob says 180 is his minimal - and usually goes about 200 spm. I always recommend the Joyo Mini Pulsating Metronome because it is cheap, vibrates so I don't annoy other runners, and small enough to hold in my hand (can also clip on my shorts but I can't feel the vibration. I like 180 as a warm up pace and if I feel good I push it up a few notches.

    ReplyDelete
  11. It's interesting that the women average a higher cadence (196.3 steps/min) than the men (193.7 steps/min) despite running at a slightly slower pace.

    Your cadence versus running speed data suggests that cadence increases about 5 steps/min per mph of running speed. Extrapolating the cadence of these elites down to a more recreational runner pace (8min/mile or 7.5mph) puts their cadence in the 160 steps/min range...

    I wonder how the cadence of these elites correlates to other parameters such as their height, weight, leg length, or VO2Max. I wonder if there is an "optimal cadence" for running at certain percentages of VO2Max pace.

    ReplyDelete
  12. I'd love to have video of all of them on an easy run. I'd bet the relationship of cadence to speed is not linear, and that they would be well above 160.

    ReplyDelete
  13. Pete you are correct. you pointed out in our video in the cadence section that my cadence was faster than 180 when at the faster pace. You can see the side by side about 4:20 in
    http://youtu.be/zSIDRHUWlVo

    when you drive your foot to the ground faster it pops off the ground faster. its all about optimizing elastic recoil and the cadence will naturally occur depending on your spring and foot strike pattern. think pogo stick. great post and data crunch!

    Mark

    ReplyDelete
  14. You guys are right in a sense. 180 is a great tool for young runners who are still developing (or really anyone learning to run) and working on their own bio mechanics. I have found that several, even very talented, high school kids who have been struggling with some injuries have been doing recovery and even threshold runs between 160 and 170. 180 can be a good tool just to measure against as a starting point with younger runners as they learn their own bodies and how to run and train themselves. The event and pace you are running can make big differences bio mechanically as well, but I think that 180 is a great place to start. I 100% agree that it should not be seen as an optimal value that everyone should strive for. Also it would be interesting to compare how these values compare to say the 800m finalists. While racing and then even while just jogging, I'm sure you will find that elite half milers will have a much lower frequency on the easy runs, as they are much less efficient at the slower runs. I'm sure the majority here are all more distance oriented but I thought the differences would be worth noting.

    ReplyDelete
  15. Pete,

    These runners all have highly trained elasticity in their strides. Very short stance phases as they "pop" off the ground.

    Do you have data on stance phase time of some of the videos where you show the heel strike overstriders?


    Mark

    ReplyDelete
  16. Mark - yes, have lots of it from a student project last year. Need to publish it.

    ReplyDelete
  17. I got the latest Garmin watch that has cadence data from the heart rate monitor, along with ground contact time and oscillation. I think more good data will become available.

    ReplyDelete

Related Posts Plugin for WordPress, Blogger...