Thank you for sharing. I'm awaiting to see what some of the big players in our sport do with a running power meter. Pondered jumping in with the Stryd but havn't heard a lot of feedback.

---

Arete Endurance

I'll have an article up on the main page in the next week or two on Stryd. Interesting device.

These metrics break down for hills, rough terrain and significant wind conditions right as none of the units on the market take into account outside factors in their modeling.

Stryd accounts for work done against or by gravity.

The reported power has been shown to correlate closely with VO2.

Wind has a much, much smaller effect on runners than on cyclists.

Most of the metrics in the charts aren't based on power in the first place.

Just wanted to preserve these nuggets (for engineers/scientist's giggles).

Somebody has to try to make a buck by bringing this to the market even if the technology is still unreliable and sketchy.
Triathletes are the best target market/audience for half-baked products.
Credulous and easily impressed by pseudo-science.
Perfect!

You do know that the Stryd was developed by two well-funded (primarily by NSF) PhD faculty in electrical engineering? They may not be experts in exercise physiology and biomechanics, but they know they're way around instrumentation.

Reliability (reproducibility) of the data is actually adequate. In my tests, for example, the CV across days is <5%. The results also correlate with force plate data, and are sensitive to individual differences that you would expect to find based on the literature (e.g., children need to generate more power than adults to run at a given speed).

I do agree that triathletes are a perfect target market for such devices - just look at the success of things like PowerCranks. Unlike that product, however, the Stryd appears capable of doing what it claims to be able to do, i.e., quantity running power (and other parameters) with adequate accuracy and precision.

The question is therefore not "Is it any good?" but rather, "How do you best utilize the data?" That's what the charts I have developed are intended to help with, and what we* hope to explore over the next few years with support from USATF and the USOC.

*Steve McGregor is the PI.

Thank you for sharing, WRT "WKO4 therefore employs a validated mathematical model drawn from the scientific literature to calculate Kleg", Does the reference would be following one?
http://journals.humankinetics.com/...ocumentItem/4734.pdf

---

Ale Martinez
www.amtriathlon.com

"That is, individuals with stiffer “springs” rebound off the ground more readily, minimizing the need for their muscles to generate force and the time during which they must do so."

Serious question..., how do i get stiff?

---

"Good genes are not a requirement, just the obsession to beat ones brains out daily"...the Griz

I bought a Stryd on impulse on eBay. Early adopter. I had an Ergomo. In fact I had two.

I've used the Stryd once, so very early impressions and a single data point. I was doing 1k intervals at 3m50s/km pace +/- 2 seconds (for once I was pretty consistent. So that's about 3m 50s of effort per int. Power came out consistently at 403-406W ave per int. So first thing is is that I was impressed that power was as consistent as pace was.

Secondly, I don't find intervals at that pace especially enjoyable - that's pretty much my 5k PB pace. My bike equivalent is 5 min ints at 300-320W, which is slightly over my CP20 wattage. So I was quite surprised to see such a large delta between similar duration ints at my 20min run pace and my 20mins bike pace. Admittedly the run ints are 1+min shorter, but even so. Maybe I am slacking on the bike; but the bike stuff has all be done on a turbo so I assume wind resistance isnt factoring in.

Third, at around 5m20s /km pace, I was surprised to see running power around 270w, because my RPE at that running pace is lower than my RPE would be biking at 270w.

So, where does that leave me after one outing? Provisionally thinking that ftp for a bike is not the same umber as for a run. Perhaps obviously, but my feel is that they are more different that I'd expect even with the starting point that they will be different. Also I realised that Stryd is much more like iBike than (say) an SRM or a Quarq. It's a power calculator rather than a power meter. That's not to say it's not accurate, but it's obviously not measuring power in the same way that a PM does; it is using some no doubt impressive algorithms to calculate power from the data it can grab. Whether in the long term it will be more useful than running to a pace (which I find extremely useful), I'm not sure. Early days.

Finally, as someone who doesn't routinely wear a HR belt, it's easy to forget to put it on. Like today, when I forgot all about it until my Suunto told me it was looking for it. Need to make a habit...

That's the $64,000 question, but of course, in order to answer it for yourself, you first need to be able to quantify it. Based on the literature, though, the answers would seem to be 1) run lots, 2) run on soft surfaces, and 3) perform resistance training (especially plyometrics).

At the same metabolic rate (and hence approximately the same perceived effort), power (I.e., rate of work performed against the external environment) will be higher when running vs. cycling. This is due to energy recycling via the series-elastic elements of the musculoskeletal system. As an example, a comfortably hard effort for me while running is ~4 W/kg, whereas during cycling it is ~3.3 W/kg.

That's interesting. Do you happen to have any links to papers that discuss this? Thanks.

Well known phenomenon. I suggest starting your reading with Cavagna's classic papers in the 1970s and working backwards from there (that's what I did).

Yes, that's the one. I chose it because it has been found to be more accurate than other models in the literature:

http://www.ncbi.nlm.nih.gov/pubmed/22682258

Let me know if you need help implementing it in GoldenCheetahR - we wouldn't want the programs to grow too far apart. ;)

Here is a figure I put together a few months ago to try to illustrate the phenomenon:

a couple of things. you made a very compelling case for why a running power meter is, well, to quote an eminent exercise physiologist: "I just don't see the point."

then you became a consultant for stryd. now you've changed your mind.

now, to be very clear, you were very clear. you disclosed your consultancy relationship with stryd, which was proper. still, were i to make the very case, point by point, you made a year and a half ago against the relevance of this metric for running (and why it is relevant for cycling but not for running) would those points be wrong today?

i'm not arguing against the metric. i have two people on my team investigating pursuant to reporting on this: one on the device, the other on jim vance's book. i have no opinion. i don't have enough facts to feel qualified to have an opinion.

just, there are companies out there pushing HR, power, ventilation, hemoglobin, cellular hydration, HR variability, blood lactate, economy of motion, as the key metric during running training. the very first question i ask these companies is among the very first questions you asked about this metric: how does it impact training; and why does it deserve to replace metrics we now use?

---

Dan Empfield
aka Slowman

First, did you even bother to read the article I posted? If you did, you apparently missed the fact that the vast majority of the analyses that I described are NOT based on running power. As such, they are equally applicable to and accessible by, e.g., Garmin footpod users.

Second, as I've described it to my wife, I personally wouldn't have invested significant time and energy developing a running powermeter, because the use cases seem somewhat limited.* Now that somebody else has, however, why not try to figure out the best way to use the information provided? I do think, though, that greatest potential lies in the fact that such devices "can make your body a mobile biomechanics laboratory" (to paraphrase SRM, and as pointed out by Steve McGregor). IOW, it's not power alone that intrigues me, it is the whole package.

*I have changed my opinion on a couple of things since starting to work with Stryd: 1) trail running is more popular than I realized, and 2) I underestimated the value of power as a real-time pacing tool, which accounts for changes in elevation and is not prone to the sorts of errors that can limit the utility of footpods and GPS.

Final comment: if you bother to wade back through the original thread, you'll find that I never questioned the accuracy of running power measurements - only the potential impact in the sport. This is the opposite of, e.g. left-side-only cycling powermeters, where accuracy itself is the issue.

ETA one more comment: I am frequently given devices by various companies marketing wearable tech and asked to try them, presumably in hopes that I will get excited about them and help them become popular (and their producers profitable). Of the various offerings, only Stryd intrigued me enough to give things more than a cursory look.** So, while only time will tell whether power will have as big if animpact on running as it has on cycling, I do think it has more potential than other measures.

**I did spend a fair bit of time testing the Moxy, but that was really more of a professional interest, I.e., to see if I could use it in research.

"I personally wouldn't have invested significant time and energy developing a running powermeter... Now that somebody else has, however, why not try to figure out the best way to use the information provided?"

that's fine. no problem. just, the posts you wrote questioning the utility of a running power meter were based on this very meter, rather than a hypothetical discussion of a running power meter were somebody to eventually make one.

"if you bother to wade back through the original thread, you'll find that I never questioned the accuracy of running power measurements"

right. and i didn't say you did. you questioned the usefulness of the metric, the impact of wind on running versus cycling, and as i recall something about the greater variability in running versus cycling power or effort.

look. i'm agnostic. and maybe my memory is wrong, but when that power meter first came out it seems to me your posts all leaned toward why this won't be effective versus why power is effective in cycling.

but opinions can change. i think your opinion changed. nothing wrong with that.

---

Dan Empfield
aka Slowman

As I just said, I am still not entirely convinced that running power, per se, will have a huge impact on how runners train and perform.* Now that such devices exist, though, I find it an interesting intellectual challenge to see if I can leverage the data they provide to maximum benefit. Moreover, as I have come to realize, the true strength of Stryd is not their current device, but the brains behind it.

*Outside of the pacing benefits I mentioned, that is.

"Outside of the pacing benefits I mentioned, that is."

this brings us back to whether the power meter or the HR monitor is the better gauge of pace. i think you make a winning case for power in cycling. but i think the questions you raised originally about using this metric for running rise up and present themselves again when again choosing how best to gauge running effort and pace.

i'm not arguing with you, or at least i don't think i am. i'm presenting the two andy coggans, one of whom argued one way and the other of whom is arguing another way. if we could square the circle then there would be no argument.

---

Dan Empfield
aka Slowman

Given the way heart rate always lags behind changes in exercise intensity, I don't think there is any question that power is better than heart rate (both, of course, being interpreted in the context of perceived exertion).


I think that I have made my positions and opinions perfectly clear, and that you're seeing contradictions that don't exist.

Andrew Coggan
Feb 2, 15 10:23
Post #47 of 207 (3092 views)
A few thoughts...

Knowledge of a person's actual power output while cycling is immensely valuable, because:

1) mechanical power is what moves the bike down the road, and

2) cycling economy/efficiency does not vary dramatically between individuals, such that power provides a reasonable proxy for actual metabolic demand.

With that in mind, consider a running power estimator, i.e., one that relies on GPS, acceleromters, etc., to calculate power:

1) since power is being determined indirectly, the reported values provide no further insight into someone's actual performance than does knowledge of their pace and changes in elevation (although if accurate, estimated power does have the benefit of pinning a single number on things), and

2) running economy varies much more between individuals than does cycling economy/efficiency, such that the calculated power may not provide a valid/reliable indication of actual metabolic demand.

Combining the above with the fact that runners don't generally have to be able to rapidly change pace the way cyclists do, that they aren't as influenced by environmental conditions (e.g., wind), that they tend to utilize flat surfaces (tracks) for structured training, etc., and I can't see a running power meter having a significant impact on how people actually train and perform.
Co-developer, WKO4 software, and originator of the following concepts: Coggan classic training levels, normalized power (NP), intensity factor (IF), training stress score (TSS), power profiling, quadrant analysis, the Performance Manager Chart (PMC), the WKO4 power-duration model, auto-phenotyping, elevation-corrected power, new individualized training levels ("iLevels"), gross power released (GPR), gross power absorbed (GPA), kurtotic index (KI), and maximum effective pedal force (MEPF) metrics to describe pedaling, and running effectiveness for runners. Some lead, others follow.
Last edited by: Andrew Coggan: Feb 2, 15 10:24

You changed your mind.

Yes, I have on some things - in particular, I underestimated the potential utility of accurate, real-time data for pacing purposes, which is where a number of people have found benefit. But, changing one's mind is allowed, isn't it?

In cycling I thought you were an advocate of pacing by RPE, only using a power meter to ensure you don't start too fast at the beginning of a time trial?

Why would you decide that a power meter is useful for pacing when running but not for cycling?

It isn't as though people have not been able to pace perfectly well without power meters - although I'm aware some have been stupid enough to pace using heart rate.

Are you really claiming that a running power meter enables better pacing than paying attention to breathing, leg sensations, temperature sensations, etc etc? If it feels hard it is hard, as you often quote, given the unreliability of power meters, the fact you can't hold the same power in hot weather as cold weather, when fresh, when fatigued, good days bad days, I would contend that using power to pace when running is risky, misleading, even dangerous.

To quote Charles Howe: "Perceived exertion modulates power, power calibrates perceived exertion."

IOW, I was talking about the usefulness of power vs. other objective measures, I.e., heart rate, estimated pace (e.g., from GPS or footpod), or even accurate spot checks of pace, e.g., mile splits.

I've recently starting using power measurements for running and for back of the pack runners like me, I think it going to turn out to be very useful.
Let me explain. When I am doing low or moderate intensity runs then its pretty easy to just going according to perceived effort, but when I try to do higher intensity runs my perceived effort is sky high and therefore not that useful in gauging how hard I should go. If I use HR the problem is my HR starts out low and when I try to get in in the zone I often overshoot my mark. However when I use power its very easy for me to keep my power at a certain number and keep it there for the duration of the effort. Furthermore, its real easy for me to fine tune my efforts and adjust my efforts my small amounts (e.g. do one interval at 240 watts and another at 250 watts). It also helps me when the terrain is varied. I'm sure if you're an experienced runner you don't need such an aid but if you're a back of the pack runner like me and want to keep things as simple as possible when doing higher intensity workouts then a power meter is the way to go. The other thing I would aid is the technology is relatively cheap if you don't mine running with your phone because then you don't need a compatible watch and the cost of the Styrd is only 200 dollars. Also when determining my running FTP I kept things simple and ran as hard as I could for 30 minutes and use the average power of the last 20 minutes to give me an FTP of 250 (I know I'm weak).
Anyway all this of course is my personal opinion but so far running with power has made running fun for me.
Steve Upshaw

They just released the 2.0 version which is now a pod attached to a single shoe.

Out of curiosity has anyone used this yet in terms of evaluating shoe choice vs power vs speed etc.

IE say multiples of 15 minutes at pure marathon pace (steady state) on a treadmill at 1%?....say as some sort of proxy for running economy?

If the accuracy is close enough I could see this as valuable in the same way that we look at CRR for tires, IE the power to overcome the shoe choice, so to speak.

Thoughts?

Maurice

Do you know what the protocol was when comparing the Stryd to the force plate? Specifically, was it compared on a track, hilly roads, mountainous trails, etc? Did it correlate better on less "complex" terrain and if so, how far off was it on the more complicated terrain?

A force plate is a rigid platform containing multiple strain gages. You can put one in the floor or under the surface of a treadmill have somebody step on it, but you can't, e.g., bury one under sand and measure the ground contact forces under those conditions.

My mistake. I thought it went in the shoe, like an orthotic.