Changes

* '''Ground Contact Time (GCT)''' will require more analysis, but my initial results suggest it's measuring a bit shorter than reality (based on a comparison with high speed video.)

* '''Smoothness. '''This is the most interesting metric, and I've enjoyed getting useful and actionable real time feedback on my running form. However, I've hit a couple of issues. The first is that the smoothness value seems to depend directly on the tightness of the chest strap. If I have the strap as tight as I'd normally have a HRM strap the smoothness values are so bad they don't register right. If I tighten the strap so it's as short as it goes, my smoothness values start to get reported. If I tighten the strap by knotting it so that it's constraining my breathing then my smoothness values reach the "average" range. This problem is exacerbated by the way that my breathing changes the smoothness value; breath in and it reports I'm smoother than when I breathe out. The second issue is that I really want the metrics shown on a watch not my phone, as it's tough to run so you can keep an eye on even a small smartphone. The Wahoo app works well on a treadmill where you can prop your phone up or even use a tablet instead.

=Data Export=

The Wahoo fitness app will allow you to export your data as a CSV, which allows for more detailed analysis. Unfortunately the format is not documented, so I've had to reverse engineer it a little. The smoothness values are exported as ma_smoothness (overall average of the other three), ma_smoothness_x (left-right), ma_smoothness_y (up-down), ma_smoothness_z (back-forward). Wahoo claim these smoothness values are normalized so that an average runner has a smoothness of 100 in each axis. The values are also available as ma_jerkx, ma_jerky, ma_jerkz which look like the raw figures. Not surprisingly the up-down jerk are more than twice the magnitude of the other two.