Hi,
with running power now available I thought I'd try to set up an in-depth protocol to see how it compares to Stryd with the hope of somehow reconciling the quite different numbers both systems seem to be producing. I'll be collecting more data in the days/weeks to come and would be happy about some further ideas on how to compare the two.
Methodology:
- 6 km relatively flat (16 m elevation) run with FR 935, HRM-Tri & Stryd
- Varying speed (Fartlek style ;))
- Data import into Matlab for analysis
Caveat:
After typing this text, I realized using Stryd as a speed and distance source can skew the results towards a more favourable outcome for Garmin. Speed is one of the important variables entering Garmin's power calculations (see here). By providing the high-quality speed measurements from the Stryd, Garmin's algorithm doesn't have to rely on the lower accuracy GPS measurements). I will repeat this tomorrow :).
First let's have a look at running power from both devices over the whole run. The two curves look similar, however Garmin's power estimate is always higher, which is also evidenced by the 329 W Garmin gives as average power, compared to 247 W from Stryd.
Shifting both curves to the same baseline (i.e. subtracting the mean power) shows that apart from the offset, both curves are quite similar, although Garmin appears to be more "noisy":
Let's now have a look at how well the two power measures actually correlate. I have removed some outliers where either of the two devices shows a power below 100W - this is where I had to stop to cross a street and I'm only interested in the power during the run and not the ramp up/down behavior. I'm getting a correlation coefficient of 0.88 - which is pretty good. Furthermore, I calculated that the Garmin power is roughly 32% higher than the Stryd reported power.
So at first glance, both devices report quite closely correlated power values, with Garmin showing an approximately 32% higher value on average. Since there are a lot of different power measures in running, maybe both systems are reporting a different power measure. Garmin says the measure "propulsive power applied at the road" (see here), whereas Stryd measures the "positive machanical energy" (see pdf here) (don't be confused by the term energy, power is just energy per time). Looking at those definitions, they seem to be the same, however this is not my area of expertise. Both companies also talk about muscle metabolic efficiency and how it relates to power. While Stryd cites a maximum achievable efficiency around 25% (see previous pdf), Garmin quotes research showing between 39% and 60% ( see here). This very strong discrepancy leads me to believe that they are indeed measuring different power values that are however closely related. Maybe someone with more in-depth knowledge of the topic can chime in here.
Conclusion:
- Stryd and Garmin power is closely correlated
- Garmin power is higher, possibly due to different definitions of power used by both companies
- More data is needed, especially without the Stryd as the speed/distance source
