Strava elevation issues (and some stryd comments)

Anyone else having issues with Strava putting you under sea level or showing elevation profiles that are too flat?

recent example: https://www.strava.com/segments/13608365 the elevation difference is off by 1,000ft and it has me under sea level.

here’s the run the segment was created from; here, the elevation difference is correct, but the elevation change/mile is way off seeing how the avg change per mile is actually 130ft/mile: https://www.strava.com/activities/768135029

connect not having the issue, so it points to Strava, not my barometer: https://connect.garmin.com/.../activity/1437493347

and for kicks, here’s Stryd (select elevation; this elevation profile is better than both Strava and Garmin’s elevation): https://www.stryd.com/...run/4545533397958656

At first, I thought it was the stryd sensor messing things up, but I’ve since seen issue with mtb’ing friends on Strava with their elevation profiles.

and random stryd comment: I think it would benefit from right/left pods or a higher sampling rate. The variability present in my post-hoc chart is much more variable than what I saw on my watch. And, having a heads-up glasses display would be much better for run power, as looking down at a watch when trying to peg a particular wattage (was trying to stay at 250w for kicks on the slow run above) is not optimal. Also, stryd distance and pace is off on their website compared to GPS and Garmin footpod, and Stryd pace (live and post-hoc) is very off on a treadmill (compared to the footpod which correlates strongly with PE and GPS outdoors - not comparing to treadmill belt speed).

random stryd comment: I think it would benefit from right/left pods or a higher sampling rate. The variability present in my post-hoc chart is much more variable than what I saw on my watch. And, having a heads-up glasses display would be much better for run power, as looking down at a watch when trying to peg a particular wattage (was trying to stay at 250w for kicks on the slow run above) is not optimal. Also, stryd distance and pace is off on their website compared to GPS and Garmin footpod, and Stryd pace (live and post-hoc) is very off on a treadmill (compared to the footpod which correlates strongly with PE and GPS outdoors - not comparing to treadmill belt speed).

  1. A higher sampling rate would increase, not decrease, the moment-to-moment variability.

  2. Asymmetry while running could definitely be interesting from an injury perspective. OTOH, you can’t be too asymmetrical while running (in terms of common metrics) and still travel in a straight line, so it would require a “deeper dive” into the data to pick anything up. (Anecdotally, I have found that running a few days per week seems to diminish my on-bike asymmetry, which I assume is a result of forced training of my motor control system.)

  3. I have found the Stryd’s footpod pace (internally reorded and viewed post-hoc) to be within 1% of my carefully-calibrated treadmill. Indeed, considering the inherent errors (and bias) of GPS and the fact that other footpods don’t account for variations in stride length from the assumed value, I would tend to trust it over anything else. OTOH, I have noticed that the live pace shown on my Garmin 920XT in Stryd’s IQ app is A) noisy, and B) inaccurate. I therefore assume that there’s something going on on the receiving end of the data.

I don’t put too much stock in Strava’s information. It doesn’t always copy correctly from Garmin Connect, and it has been really bad lately. It has been doing random mapping of my runs (sometimes showing that I ran in a completely different neighborhood), and chopping hours off of my bike rides. It doesn’t really bother me because I don’t pay for the premium service.

Thanks:

I’m seeing a lag in the live data (garmin, wahoo app and Stryd app), so a higher sampling rate could be better.

The pace on the treadmill is off on all of the 3 above devices/apps, so not garmin, and it’s consistently off by 3min/mile… and the Stryd app link above has the total time wrong - garmin isn’t off on time, too; so that puts the avg pace is off on Stryd, too.

I’m liking Stryd a lot and these are little errors that can be fixed

Thanks:

I’m seeing a lag in the live data (garmin, wahoo app and Stryd app), so a higher sampling rate could be better.

The pace on the treadmill is off on all of the 3 above devices/apps, so not garmin, and it’s consistently off by 3min/mile… and the Stryd app link above has the total time wrong - garmin isn’t off on time, too; so that puts the avg pace is off on Stryd, too.

I’m liking Stryd a lot and these are little errors that can be fixed

Lag is one thing, variability is another.

Have you tried downloading/uploading your Stryd via the phone app, and looking at the speed/pace in the PowerCenter? That way you can cut Garmin and Strava out of the loop, to see where the issue arises?

Yep, see the Stryd link above which came from uploading to the phone app.

With lag in live time, my variability in power increases on very undulating terrain

one additional comment and then I have to sign off - but I’ll read your response(s):

To the comments that Stryd is really only good for monitoring form, this is something I disagree with. What I would respond with is that running is like cycling with the exception that it’s like riding a bike where the wheel can suddenly change into an octagon or square - obviously, that’s going to increase power while decreasing speed. So, if using that analogy or using a more realistic analogy of it being like a brake pad rubbing, Stryd measures power for an activity that is much more dynamic that cycling, and from my experience, it is doing it pretty well.

Now, I will say that Stryd (as is the same with HR and GAP), the power drops too much on downhills if a watt is representative of a Calorie. I think that’s largely due to the ‘torque’ factor being isolated to the movement and change in motion of the foot (or body with the original chest strap version). But, the energy required for knee and hip flexion are other sources of ‘torque’ not captured by Stryd that do increase Calorie expenditure and also increase mm lactate/l blood, so if FTP is equal to 4mm/l, Stryd is not providing that measurement. I equate it to holding a high cadence on a descent while biking; I can still maintain very high power but if the torque were being measured as a function of change in forces between the tire and the ground, power would be reduced on downhills more than it should be, but because torque is measured as force applied to the pedal, power is still accurate while cycling downhill. Stryd running power is not accurate while running downhill according to what I’d describe as my fine-tuned PE and perceived lactate level in my legs, chest and remaining body depending on the effort. Stryd needs to factor in change of movement at the knees and hips; I’m fairly certain, from data I’ve personally collected and analyzed in my own hobby-studies while conducting research at Ithaca College and Harvard, Stryd’s foot sensor can capture change in knee and hip angles from a footpod, though two footpods would be better, as minor imperfect bilateral movements can have huge consequences in running form.

I’d go so far as to say that with two footpods, a stick figure could be created (side view) allowing the user to see their form in live time, and a second stick figure could be created representing ideal form. Put these in a heads up display, overlapped, and allow the user to match the stick figures movements to reduce “form power”. That would be my ideal use of Stryd: Power and the ability to see when my ‘wheel’ isn’t round and to fix it with visual feedback.

Now, I do fully expect you to deconstruct that pseudoscience above to the nth degree, but I hope in the process you can take the feedback as something worth considering and addressing, as this is not just my perception of Stryd, but also the perception of my peers who are using the new Stryd. Thank you

So Andrew, you recommend to use the IQ Power Data Field and the Pace Data Field that Garmin has included to get a better accurate Pace?

My Styrd Power still not arrived (Europe) but I can’t wait to run with a better pace (rather than GPS and Garmin Fodpod) on my hilly forest trails. :slight_smile:

Sadly that most Garmin watches only allow to use 2 IQ-Datafiels simultaneously. Would be also nice record other values from the Stryd Power directly into the watch.

Is there a way to merge down the data easily and synchronize it without any offset from the Stryd recorded file into the Garmin FIT file?

one additional comment and then I have to sign off - but I’ll read your response(s):

To the comments that Stryd is really only good for monitoring form, this is something I disagree with.

Who said that?

What I would respond with is that running is like cycling with the exception that it’s like riding a bike where the wheel can suddenly change into an octagon or square - obviously, that’s going to increase power while decreasing speed. So, if using that analogy or using a more realistic analogy of it being like a brake pad rubbing, Stryd measures power for an activity that is much more dynamic that cycling, and from my experience, it is doing it pretty well.

Now, I will say that Stryd (as is the same with HR and GAP), the power drops too much on downhills if a watt is representative of a Calorie. I think that’s largely due to the ‘torque’ factor being isolated to the movement and change in motion of the foot (or body with the original chest strap version). But, the energy required for knee and hip flexion are other sources of ‘torque’ not captured by Stryd that do increase Calorie expenditure and also increase mm lactate/l blood, so if FTP is equal to 4mm/l, Stryd is not providing that measurement. I equate it to holding a high cadence on a descent while biking; I can still maintain very high power but if the torque were being measured as a function of change in forces between the tire and the ground, power would be reduced on downhills more than it should be, but because torque is measured as force applied to the pedal, power is still accurate while cycling downhill. Stryd running power is not accurate while running downhill according to what I’d describe as my fine-tuned PE and perceived lactate level in my legs, chest and remaining body depending on the effort. Stryd needs to factor in change of movement at the knees and hips; I’m fairly certain, from data I’ve personally collected and analyzed in my own hobby-studies while conducting research at Ithaca College and Harvard, Stryd’s foot sensor can capture change in knee and hip angles from a footpod, though two footpods would be better, as minor imperfect bilateral movements can have huge consequences in running form.

So if I understand you correctly, you’re judging the accuracy of Stryd’s power measurements on downhills based on your perceived exertion?

I’d go so far as to say that with two footpods, a stick figure could be created (side view) allowing the user to see their form in live time, and a second stick figure could be created representing ideal form. Put these in a heads up display, overlapped, and allow the user to match the stick figures movements to reduce “form power”. That would be my ideal use of Stryd: Power and the ability to see when my ‘wheel’ isn’t round and to fix it with visual feedback.

Now, I do fully expect you to deconstruct that pseudoscience above to the nth degree, but I hope in the process you can take the feedback as something worth considering and addressing, as this is not just my perception of Stryd, but also the perception of my peers who are using the new Stryd. Thank you

Many people love the idea of using technology to help them achieve their “perfect” running form. The problem is that the former can’t easily be defined, at least for a given individual. For example, Bill Rogers had a famously asymmetrical arm swing…would he have been able to run even faster if he had corrected the “problem”? Peter Cavagnaugh was smart enough to know that the answer to that question was “hard to say, but I’m not the one who is going to mess with success.”

So Andrew, you recommend to use the IQ Power Data Field and the Pace Data Field that Garmin has included to get a better accurate Pace?

My Styrd Power still not arrived (Europe) but I can’t wait to run with a better pace (rather than GPS and Garmin Fodpod) on my hilly forest trails. :slight_smile:

Sadly that most Garmin watches only allow to use 2 IQ-Datafiels simultaneously. Would be also nice record other values from the Stryd Power directly into the watch.

Is there a way to merge down the data easily and synchronize it without any offset from the Stryd recorded file into the Garmin FIT file?

FWIW, I am using a Garmin 920XT with the Stryd IQ app (not just the data field). Data from the footpod goes directly to Stryd’s Powercenter. Data from the Garmin goes to Garmin Connect (if it is an outdoor run that I bother to download), which then ports the GPS data over to Powercenter for merging. In either case, I then download the .fit file containing all of the data for further analysis in WKO4.

The above is the only workflow with which I have experience, so can’t speak to what happens if you include, e.g., TrainingPeaks in the loop.

I already use the HRM Run Strap from Garmin, these metrics are fine for me. It would be nice, to get the other metrics like power, form power, etc. from the pod to a single fit file. There is an app called fitfilerepairtool (http://fitfilerepairtool.info) wich you might know, how can handle all kind of fit file optimization, editing, etc. p.p.
I understand, that at least you need a software or platform, like WKO4 or GoldenCheetah, that can handle and show up all the metrics.

I don’t believe these numbers are generated through anything more than magic.

Remember Coggan appears on the company promotional website…and his professional reputation is linked to it…

So if you get criticised by him for saying the numbers don’t work for you it might well not be you at fault and instead that the device doesn’t work.

I don’t believe these numbers are generated through anything more than magic.

Remember Coggan appears on the company promotional website…and his professional reputation is linked to it…

So if you get criticised by him for saying the numbers don’t work for you it might well not be you at fault and instead that the device doesn’t work.

Am I too blind to see where you got criticized by Andrew Coggan? You might be a little tiny bit off topic? :slight_smile:

I do not question who is related to whatever and tho I do not question the data this device gives me.
Also I am old enough and have a brain, to think about something before I buy it… But thanks for your little advice!

From my point of view it does work pretty well (Power and almost perfect and on point Pace), even better than anything else out on the market today.
The other metrics are very similar to the Garmin HRM Run data.

If the numbers do not work for you, thats bitter for you. But I doubt you have a Stryd Power at all?

There are plenty of people who say homeopathy “works pretty well” for them…in-fact I suspect there is a correlation between amount spent on first appointment and belief that homeopathy works…how much did you spend on your running power meter?

Glad you didn’t see the criticism anyway.

Why would I have a running power meter? I think it is bogus and that proponents of running powered meters should declare their conflicts of interest

I don’t believe these numbers are generated through anything more than magic.

Remember Coggan appears on the company promotional website…and his professional reputation is linked to it…

So if you get criticised by him for saying the numbers don’t work for you it might well not be you at fault and instead that the device doesn’t work.

  1. Beliefs are not data.

  2. Yes, I am a consultant for Stryd.

  3. There is already quite a bit of independent data out there showing that, e.g., the power data reported by Stryd is highly correlated with metabolic rate.

  1. There is already quite a bit of independent data out there showing that, e.g., the power data reported by Stryd is highly correlated with metabolic rate.

http://hetgeheimvanhardlopen.nl/wp-content/uploads/2016/11/The-breakthrough-of-running-power-meters.pdf

So why when you comment on running power meters do you not declare your conflict of interest and that you work for the company?

So why when you comment on running power meters do you not declare your conflict of interest and that you work for the company?

I thought it was common knowledge?

Anyway, if you want to know for whom I have been a consultant, you can always read my CV (middle of page 15):

https://iupui.academia.edu/AndrewRCoggan/CurriculumVitae

Yeah right. I believe you 🙄. Just add a line to your signature saying you are employed by Stryd as a consultant.

Yeah right. I believe you 🙄. Just add a line to your signature saying you are employed by Stryd as a consultant.

Dear Eerke, wait and time will tell us. :slight_smile:

I still have hard non-believers in my club talking that a cycling power meter is rubbish. Surprisingly the same people are really bad in good pacing and really good in over pace themselves pretty bad. :slight_smile: No Pro Cyclist and no Triathlete doubts in the usefulness of such a Tool.

Simliar discussion came up when I started to understand the benefit of a turbo trainer a couple of years ago, while in winter training. Most of my team mates laughed about the fact, that I did all my winter bike workouts indoor. Surprisingly I catched less colds and always had a much better form over the whole next season, compared to them. Now as Zwift is coming up, all did a complete U-turn and started to ride indoors. Funny from my point of view.

If you think about it carefully, you should see that there is a big big chance, that the same effect take place in running sports and running powermeters.
As always it is in this beautiful life, of course there will be some people with benefits, and some doubter.

Have a nice day!