Stryd vs Runn 15% difference on treadmill

For anyone who needs further convincing, just think about how the Runn and the Stryd work.

The Runn uses a light sensor to measure reflective strips on your treadmill belt as they go by. Dead simple, and accuracy isn't going to be affected by your pace.

The Stryd, on the other hand, uses a tiny accelerometer to try and measure the movement of your foot, which is moving in a roughly elliptical motion, and from that deduce your forward progress. On a treadmill, as far as the accelerometer is concerned you're not making any forward progress, so it has to try to figure out how fast your feet would be propelling you if you were outside. Then add to that a bit of movement between the Stryd and your shoe, and between your shoe and your foot, which will increase at faster paces. Then add a bit of slippage between your shoe and the running surface, which again will get exponentially worse at faster paces. Lastly, your stride is going to change at different paces, and as others have mentioned your shoes will affect your stride as well.

npurdy1112 wrote:

Very similar Stryd experience here. I had the thing track calibrated to every shoe I own. My pace/distance started to show fast/long. Re checked some shoes and the calibration factor was now different. Same shoes, track, protocol, etc. Stryd told me there was nothing they could do to help and would not even exchange me out for a new unit.

Nothing is ever their or the products fault. The blame was always given to something else. My error, treadmill, shoe placement, track, didnt follow protocol, maybe a future update will fix your issue, etc. etc.

Yup - this is exactly my experience also with the Stryd mindset. It's like a cult. Or the bad running coach who blames the athlete's failure to respond on the athlete, rather than the training.

What I've always found perplexing is the insistence that a) Stryd should be perfect out of the box, no calibration needed, but b) if you do need to calibrate it, then you need to do so separately for every shoe you wear. Those two points seem inconsistent to me.

Bolded sentence is quote is a good summary of this topic. Runn is best. Darkwave pointed out how she had some inaccuracies due to setup and how they are easily fixed. So as long as it's set up right, I'd trust that over the Stryd.

I do think the Stryd is close enough for most of us (as long as it's not one with an issue). I know it must be reasonably close since my HR to pace ratio agrees very closely to outdoors as it's currently calibrated. Good enough for me but I absolutely concede that if I had the Runn sensor it would be better.

npurdy and darkwave, I also don't get the claims that Stryd has that it is perfect out of the box. There is no way that is true. Had I not calibrated mine, it would be way off.

Good discussion. I may need to get myself one of those Runn sensors. : )

The Stryd is a great piece of gear and definitely more accurate than your treadmills speedo or even GPS watch in any situation where there are obstructions to the sky. The rest of the time it’s just as good. No calibration necessary for different shoes LOL. You guys are weird.

Darkwave I agree with your friend, the C2950 definitely speeds up over time even with speed set fixed.

I do 1 hour+ runs regularly and if I want to hold constant pace I have to dial the treadmill speed down. If I want to go for a 'PB' I ket it creep and negative split.

As the OP suggested, Runn is very simple kit and works really well. But when you calibrate it you can't use the treadmill speed for all the reasons mentioned above.

One thing though the flex bed throws off the incline reading no matter how many times I try to calibrate it.

I'm the 2950 owner mentioned earlier in this thread by Darkwave and it's good to see others with a similar problem on their Nordictrack's. The speed creep isn't drastic but it is definitely noticeable. I have a whiteout mark on the right side of my treadmill (opposite the NPE Runn) and at random times I will just lap my Garmin and then count 100 revolutions. I have the formula set in my Excel log so I just enter the data in after the run. The manual splits always corroborate what the NPE is stating for speed (via Zwift). Early in my runs I may get 11.1 / 11.2 but by the end I would have treadmill at 8 and Runn at 8.2/8.3.

LTHR hurts- I did mess around with my incline of the NPE quite a bit and used the app to calibrate it (I did not calibrate the NPE speed to the TM). After some tinkering I no longer get the massive elevation runs and will get a couple hundred feet over 12+ miles, which is about right since I run at .5 or 1%. I assume you have the NPE towards the back after those grooved slats end? Is the sensor as close as it can be to the belt vs the recommended pencil width?

Thanks for the info cnmm688 ... did I get that right? Yes, I have the NPE right at the back left but a good 1cm from the track. As you know Zwift doesn't care about incline so it hasn't bothered me but I will try closer and see what it says. It says to calibrate incline when you're not on it which sounds daft when there's a flex bed. Which did you do?

LTHR hurts--yes small typo I am actually cmm268 when I use my username on these boards:)

Yes, I dont care about elevation either my concern was that it was throwing off my Strava account when I would do a 10 mile run with 3k feet of elevation!

Overall though, very impressed with the NPE device and coupled with Zwift has made treadmill workouts somewhat enjoyable.

All you need to do is measure the length of the belt by placing a few temporary markers(labels work?) as you measure 1 rotation. Then you put one mark(with white out pen for example) and count revolutions at different speeds and you can easily calc the MPH to see if the treadmill is reporting the right speed. With a large motor load/not loaded makes little difference.

darkwave wrote:

In contrast, even after calibrating the Stryd on a track it can be off (both on the treadmill and the road, as well as when I return to the track). More importantly, it is not consistently off. It will be fine for 8 or 9 runs out of 10 and then then it starts reading very off - up to 20 seconds fast or slow per mile. And even within the run it will not be consistent. I could be running a consistent pace on the treadmill (as assessed by the speed on the treadmill, my perceived effort, and my heart rate) and my mile splits as reported by Stryd will vary notably both up and down.

Stryd would be good, however, if the fixation on the shoe is not good the pod shakes around on the shoe and that leads to varying in speed up and down.

The clip is a holy sh!t and mounting the pod is crucial. I mentioned that to Stryd already.

Also laces are soft, which is a problem by itself.

Glad to see I’m not the only one using both Runn and Stryd!

And I have a question for you all:

In my Garmin Fenix 6X Sapphire I have set Runn at «Always» for speed and distance, and Stryd is only paired through the Stryd Workout App and not in Garmin’s native system.
Nevertheless, in all indoor workouts, Runn is in charge of speed and distance the first 10–15 minutes, then Stryd takes over and messes everything up. My Runn device is calibrated with my mill and is more or less 100 % accurate, while Stryd measures speed about 10–15 sec/km too slow. So obviously I want to stick to Runn.

This happens in the Stryd Workout App as well as in the native treadmill app with Stryd Zones data field.

Any advice?

Am I missing something simple in setup because my stryd will not work at all on the treadmill. Thanks in advance

You can remove Stryd from the paired sensors in your watch.

By the way, if you care about accuracy, you should not calibrate the Runn to your treadmill, because then you would be just replicating what the treadmill display says, which can be off.

Stryd is not paired as sensors in my watch, and Runn is calibrated to the stickers – not the treadmill itself.

I tried so many forums about this issue, but it seems like I’m the only one experiencing this. Not even NPE knows how to fix it…

This is quiet strange.

Are you connected via ANT+ with your Runn?

Do you have the newest Stryd pod? This seems to work much better (accurate) on treadmills now.

This thread is quite something.

Basic premise: Always assume that your treadmill is wrong

There are several different types of wrong:
- True speed is lower
- True speed is higher (very rare)
- True speed is affected by load on belt
- True speed is affected by incline (and decline)
- True speed oscillates
- True speed decreases slightly throughout session
- True speed has a constant offset (i.e. ±0.5mph)
- True speed has a variable offset (i.e. 0.2mph slower at 5mph, and 1.0mph slower at 12mph)

You can test your treadmill either manually or with professional tools (for high-end mills). That's the only way to know. Doing it manually can be tedious, especially if you do it alone while running at high speeds. And for cheaper mills there's often no way to calibrate after testing, so you'll have to create a [true speed / mill speed] table and look up the setting you need for the target speed.

As another user wrote, the Stryd is based on an algorithm with lots of approximations which is apt to be wrong. I'm not sure why anyone would think that a foot pod is more accurate than a belt sensor.

The NPE Runn has two optical sensors which measures the time it takes for one (or more) reflective strips taped to the treadmill band to pass between them. This is extremely simple. The distance between the optical sensors is fixed and a known size, so the speed measurement is the simplest possible: distance/time. Increasing the number of strips on the band means increasing the data frequency, you get updates more often.

The simplicity doesn't mean users are barred from intentionally using it wrong. Some like to "calibrate" it to show exactly what their treadmill shows. I guess it feels good to get that +0.5mph ego boost, especially on Zwift.