Stryd is not all it's cut out to be...maybe

This morning I ran twice with one watch hooked up to a Stryd and set to Always use for Pace and Distance and another just on GPS. In both cases the Stryd came up with a longer distance. First up was a trail run that we've been running for years so we know it is close to 9km. Stryd returned 9.31 where my GPS Trail Run gave 8.99 and my wife's 945 8.97km. Next up was parkrun. For those who don't know, parkrun is always 5km. Our parkrun has been accurately measured using a calibrated wheel so we know it's 5.04km from the wheel. And every GPS run we've done has been 5km +/-20m or so. This morning Stryd gave 5.24km while the GPS watches gave 5.02 and 5.01 respectively.

This wasn't the first time out with the Stryd. Several runs on a 400m track and a couple of other trails runs should have seen it calibrated by now. I haven't looked over earlier runs as I did want to make sure I had a number of different runs with the Stryd.

Let's see how things progress. I have done a couple of runs on a 400m track but not sure if I had the Stryd set up properly so can't comment on that except to say the watches gave the same distance. However, I now have the Stryd where I want it so can. and will, do some more comparison runs to see how this plays out.

  • I had always understood that Stryd was better than GPS for pace, but not for distance.

    Because each GPS point has a Circular Probable Error of about 5m, one point may be somewhat behind your true position, while the next point is well in front - this makes calculating instantaneous pace difficult. However, by stringing a whole lot of points together, the errors tend to net out, and the algorithm can work out the most likely path you took, and the distance of that path.

    However, if Stryd's estimation of step length is just 1mm out, this is negligible in the context of pace (which is only reported in increments of 5s per mile/km anyway). But with no means of correcting cumulative errors in step length, overall distance can be off. Pace is not cumulative, distance is.

    Easily fixed. Footpod Pace = Always. Footpod Distance = Indoors.

  • Easily fixed. Footpod Pace = Always. Footpod Distance = Indoors.

    I'm guessing this has probably been suggested before but as I've not use a pod before I probably missed it. I'll give it a try sometime. Thanks.

  • Interesting. That is hard to argue with and my experience is that Stryd cumulative distance is slightly messed up by things like changing shoes and different kinds of workouts. 

  • That’s definitely not true. I have had 2 Stryd models. The Stryd Live — which I eventually upgraded in firmware to a full Stryd was about 1.3% short for distance. The Stryd Wind I about 2.2% short for distance. 

    Without calibration it gives noticeable slow paces and short distances. But highly reproducible on a given course and shoes. 

  • It may depend on location which is better for distance. If you are in an urban canyon situation like NYC or Chicago that are notorious for GPS reflections then the dead reckoning from Stryd is probably better even with cumulative errors. In open sky situations in suburban and rural areas it could easily be the reverse that GNSS is better. 

  • Just in case you haven’t changed the setting, always make sure you turn off the Stryd auto-calibration on your Garmin watch - I find this helps.

    I’ve used Stryd for over 5 years and on flats and the treadmill, I’ve found it gives the most accurate distance.  You need to be mindful that a lot of treadmills actually come in at a lower distance than they actually measure (based on the consensus of studies that I’ve read).

    However I live in a very hilly city and that is the Stryd’s Achilles Heel when it comes to distance measurement - Stryd is great when you keep a consistent running form (stride length, cadence etc) but when this varies more e.g on hilly or uneven terrain, then that’s when the distance measurement becomes a little more off (although still acceptable). In such circumstances I revert back to GPS.

  • Without calibration it gives noticeable slow paces and short distances.

    Agreed - You have to be completely oblivious not to notice this, its really apparent, especially if you move from one model of Stryd to another. And that is before the shoe change issue which no one seems to discuss, but again is blatantly apparent.

    A guy on Youtube has compared 4 Stryds at once believe it or not, and he sees some needing calibration of 98% and others 102% 

  • I do t have a Stryd, but what you say makes sense to me. A foot pod is really as close to the source of power/motion as you can get, so it should be quite good at per-step metrics and power calculations. But as for total distance, it's just a form of dead reckoning, and errors are going to be cumulative the longer you run.

    With GPS, you always have a magic of error (within 10s of meters from pure satellite reception alone) but it is constant at any given point. Plus, the faster you go the less the errors matter because they stay static and become a smaller variable in the accuracy. So I would imagine gps total distance to be better overall.

  • It may depend on location which is better for distance.

    I'd have to say this is the most accurate of comments made here.  I live in the Pacific Northwest and the landscape is nothing but Fir trees for as far as one can see.  Even residential areas are surrounded.  For obvious reason, this poses a challenge for GPS.  That being said, I find distance from my Stryd much better than GPS.  I suppose I might have a different opinion if I ran in open areas, but I don't so I can't say which is better.  I switch between two sets of running shoes and I don't see a difference as others have stated.  As far as changing pace, there is a slight difference in distance measurements with my Stryd, but as I said earlier, for me it's still much better than GPS.  There is no correct answer to if the Stryd is better or not.  The answer is the same as many other threads here about features/accessories.  It works for some and for others it doesn't, but no one can say it does or doesn't work for all.

  • You want to share a .FIT file which i can analyze? Send it as a private message if you want. Slight smile