I have the same problem. It's now clear that fenix 6 uses internal accelerometer to adjust the pace when gps signal is not 100% strong. I also believe it can be something Garmin potentially could fix via software update. In my case, I see biased (slower) pace even if the trace on the map is quite good (maybe the signal is still good for tracking ma not 100% affordable for pacing, and it uses accelerometer). Another strange behaviour is that for real pace between 5.20 and 6.00 min/km the F6 pace is terrible....if i run faster (<5 min/km) the F6 pace tends to be quite good.
Your drop-outs seem odd, but if it's the peaks and valleys that concern you - Stryd is not the solution.
Pace graphs from Stryd as actually very choppy - don't expect smooth lines. Garmin smooths GPS pace, Stryd is unsmoothed, so if you look at a pace graph from Stryd you may have a shock!
This isn't my graph, but it's posted to Stryd's facebook page - first two runs are with GPS, 2nd two are witth Stryd.
Great thanks, it is the first graph of its kind that I see without a foot pod, as silentvoyager says you will live in a very open area and well covered by GPS. Great
I too have had similar results to @johanwiren - my run was a steady 5k run until I speeded up at the end. This is from GPS. It's not typical I have to say! I pace myself with HR, not pace generally.
How well does a Stryd work with a varied running pace and also varied cadence?
Way better than any GPS based device. Due to the inherent limitations of GPS (sample rate and >3m accuracy per sampled point in best case scenario), it will never be able to compete with a foot pod for speed/ distance/ cadence/ change of speed & direction. Add to that poor signal due to built up areas and tree cover.
This is a good review and specifically looks at accuracy and precision…
“Under good conditions most of the watches are remarkably good, but when things get a little tough the differences become more apparent. However, none of the watches have GPS accuracy that is good enough to be used for displaying your current pace.”
The Stryd works great for me running trails in the Cascades with many changes in pace including switching between running and walking. Why would you think making a dramatic change in pace would change on a Stryd regardless if running in the open or in mountainous terrain? It's not affected by the limitations of GPS pace.
However, none of the watches have GPS accuracy that is good enough to be used for displaying your current pace
It has been shown here and in other threads that Garmin's current pace doesn't come from GPS alone, so your comment is moot. Garmin has invented an algorithm, it seems, where it tries to be smart and mix in accelerometer data into pace calculation. And that apparently doesn't work the way it should.
If the pace was coming from GPS alone, even though it could be inaccurate, over longer distance it would average out and be consistent with the distance travelled over time. Garmin's pace is not. Garmin's pace, on average, is always slower than what you'd get from the distance and time. Here is an example from my today's trail run in a local watershed park:
The data comes from parsing the activity TCX file, so that is exactly what the watch has recorded without any corrections. Avg Instant pace is calculating by averaging the pace from all samples (1 second recording was used) and the split pace is what the watch displays for every split. These should match, but with exception of the last split there is a consistent bias. If you look at the entire run, the instant pace, on average, is off by 0:35/mile.
