Changes

Stryd

1,361 bytes added, 09:59, 15 June 2017
no edit summary
I've found Stryd to be particularly useful in [[High Intensity Interval Training]], especially when using the [[Polar V800]] that has a native power support. With the V800, I can set the watch to display the maximum power for the current lap, which gives me a far better idea of how hard I'm pushing it on each interval than is possible any other way. You can see in the graph below that in short HIIT intervals the heart rate is a lagging behind power estimate, and my heart rate is still rising after I've slowed up.
[[File:StrydHIIT.jpg|center|thumb|400px|A HIIT interval from my [[Katana Run]] workout.]]
Below is an example from another runner during a marathon plus, mountainous training run. I've applied a lot of smoothing to the heart rate and power data, then scaled and aligned the two tracks. Over the first mile or so you can see the resting heart rate, rise up as of a warm-up, then the heart rate and power match reasonably well up and over the first hill, again suggesting that the Stryd power estimate is probably reasonable. For the second half of the run, you see heart rate rising above the power track, which I believe is an indication of [[Heart Rate Drift]], probably due to heat and dehydration. This shows one of the benefits of even a broad approximation of power compared with using heart rate for longer training runs.
[[File:Stryd HR Drift.jpg|center|thumb|400px|]]
However, it's also in indication of some of the weaknesses of mathematically estimating our rather than measuring it. If we look at the steep descent on the second hill, it looks like there is still some underestimation of power. The other possible issue is that Stryd is estimating for a typical runner, and each person's efficiency under different conditions is likely vary quite a bit. It's important to remember the limitations of this type of estimation, and avoid confusing it with what you'd get from a power meter.
[[File:Stryd LpV Descent.jpg|center|thumb|400px|]]
==Statistical Analysis of Power Estimate==
I've not performed any statistical analysis of Stryd's power estimate, mostly because I can't work out a reasonable way of doing this. Ideally, I'd be able to compare Stryd's power estimate with VO2 measurements from respiratory gas exchange, but I don't have the equipment, nor is it likely to be practical. The approach I've used above is to compare Stryd's power estimate with heart rate, but while heart rate is a well-established method of measuring exercise intensity, it's flaws (as noted in detail below) me know that he can only be used for relatively short periods of steady state exercise. (I'm hoping that someone will come up with a portable VO2 meter that will measure oxygen and carbon dioxide in your breath.)