If you’ve watched the Tour de France in the last decade, you’ve probably noticed some riders staring at their handlebars on the biggest climbs, looking at a little screen as if they are in the middle of a Netflix binge. Most of the time, they are looking at the readout on their power meters.
Four-time winner Chris Froome is particularly known for this practice. Gone are the days of staring in the eyes of competitors to decide when it’s time to attack. In their place, riders can plug a number into an algorithm and meter out their effort accordingly. Critics would say it’s a transformation from art to math, human to machine.
The reason most cycling training programs use power meters is that they provide an effective way to plan and implement training. There’s a lot of variation with how different training systems work, but the basic premise is that if you know your functional threshold power (FTP), or the power you can sustain for about an hour, you can extrapolate or interpolate the power you should be putting out in interval workouts or races (with lots of variation for different training methods). That’s why the cycling peloton has largely replaced sexy stories of personal grit with what often seems like a transcription of old dial-up internet connection sounds (beep-bop-beeeeeep-bop Froome wins again).
A similar device would be the holy grail of running training metrics, right? It’s a bit more complicated than that in running. Let’s ask some straight questions and get some answers. (Note: the terminology below sometimes varies depending on context, and many of these points are hotly debated.)
What does a running power meter measure?
Those bike power meters measure external mechanical power (usually with a sensor in the crank arm or elsewhere): how much work is transferred from your body to the machine. To be more precise, they often measure the magnitude and direction of the applied force and the angular velocity of the crank to spit out a power number (with variation for different designs).
As outlined in a white paper, run power meters like Stryd try to measure mechanical power similarly by combining two primary variables: forward force (calculated with an accelerometer that compares the max and min velocities during takeoff and landing) and vertical force (calculated from vertical displacement of the center of mass). From those variables and others, Stryd claims to estimate power accurately compared to controlled lab tests conducted by outside researchers.
That was boring. Why might that number be helpful?
Scientist and author of Endure (which makes a great Christmas present), Alex Hutchinson had a wonderful breakdown in a January 2018 article in Outside. Basically, mechanical power on its own isn’t too helpful. Instead, what is helpful is how it correlates to metabolic demand, or how much energy the body actually uses. In cycling, the mechanical power number usually has a steady relationship to metabolic demand (many studies show ~20- to 25-percent efficiency, with variance for the type of activity), allowing it to be used for pacing and fueling.
That relationship is not as clear in running. It gets back to how the body actually generates force for forward propulsion. In biking, whether you are going uphill or into the wind or anything else, you’re still just putting power into the pedals in a nearly uniform way. In running, there are more factors at play, from braking force when you land to the internal work of arms pumping. Stryd calibrated their algorithm so it’s not only measuring mechanical power, but also incorporating other factors like gradient. Essentially, running power meters estimate metabolic demand from a mix of mechanical power proxies and other variables, which change based on grade, among other things. If that estimation is correct and repeatable, then it can be used a lot like bike power meters for pacing and fueling.
Still boring. What do the recent studies say?
An August 2018 study in the Journal of Strength and Conditioning Research sought to answer the question about running power meter estimates and their relationship to metabolic demand. In that study, 13 recreational runners and 11 elite runners had their VO2 assessed at three different paces, both indoors on a treadmill and outdoors on a track. The results showed a significant relationship between running power and metabolic demand, but it was weak, which the authors said suggested that running power “is not be a great reflection of metabolic demand of running in a mixed ability population of runners.”
The offset gets back to how the number is derived, as discussed above. Cycling power meters are measuring a comparatively simple human-to-machine force transfer. Running power meters estimate the number through a more complex range of variables and corrections, and “variables that may have a direct impact on the efficiency of running (including running surface, coefficient of drag, or wind) cannot be captured or accounted for at present.” Theoretically, that could be more impactful on trails that have varying surfaces and technical terrain.
However, the study found, the power meter did show significant relationships between ground contact time, vertical oscillation and cadence when compared to running economy in recreational runners. Elite runners, who had many years/coaches/competitions to optimize efficiency, did not show the same relationship. The takeaway was that running power meters may be able to help non-elite runners improve running mechanics over time by making changes in form and monitoring the power relationship.
An article in the December 2018 issue of the Journal of Strength and Conditioning Research from Stryd researchers refutes the study’s findings. For the purposes of our discussion, all you really need to know is the title: “Methodological Flaws in Aubry, RL, Power, GA, and Burr, JF. An Assessment of Running Power as a Training Metric for Elite and Recreational Runners.” Essentially, the argument is that the data is valid, but the normalization of running power and metabolic demand over speed is problematic, and that the underlying data actually supports the effectiveness of running power meters. Stryd cites a number of other studies to support their claims. It’s the scientific version of a rap-beef.
This is more exciting. So what do you think?
I have seen some athletes use running power meters to pace negative split marathons on hilly courses, evaluate performance improvements over time on variable terrain and alter running mechanics to improve efficiency. Others have disputed the effectiveness of power meters. As a coach, I have not used it personally. However, renowned mountain runner Joe Gray uses it, and he may be the best trail runner out there. Whether those evaluations of effectiveness are indicative of causation, the placebo effect, personal bias or incidental observations is uncertain.
In that way, it’s a lot like a GPS watch, a heart-rate monitor, heart-rate variability tracker or other device. It could be useful for some athletes to calibrate perceived exertion, especially those that like to look at data and view that analytical approach as part of the fun of training. It might not be useful for others, especially if thinking of the power number as a 1:1 relationship with metabolic numbers across all circumstances.
And as the technology improves, power meters may end up becoming indispensable, like their cycling counterparts. We’re probably not there yet, but it’s exciting to see companies taking a risk with novel technologies. Only time will tell whether we’re looking at a technology like that developed by the Wright brothers (the plane) or Dean Kamen (the Segway).
David Roche partners with runners of all abilities through his coaching service, Some Work, All Play. With Megan Roche, M.D., he hosts the Some Work, All Play podcast on running (and other things), and they wrote a book called The Happy Runner.