We're actively looking for ideas on how to improve Weekly Challenges and would love to hear what you think!Submit Feedback
Strava is a popular app that serves a social network for all athletes. In particular, the app is popular with runner and cyclists. However, sometimes we want to see the data differently from what is served to us. Below is a GPX file containing a mountain bike ride I completed a few weeks ago. The challenge is to parse it and create report snippets like these:
Feel free to try anything else on top of these! I will not be providing a starting workflow since the results are pictured above, and connecting to the .gpx file is part of the challenge!
The distance between lat/long observations divided by the time elapsed results in some pretty phenomenal speeds!
Or perhaps Joe can really climb Belcher Hill at 60mph? What's your secret Joe? Newfangled e-bike? Secret high speed chair lift?
My first draft is attached (named minimal), and upon looking at the data, I learned that in the Charting tool, the Chart Type of Line does not allow us to set an X-Axis, so what it uses is the record sequence. This means if we want to use Time as the X-Axis, each record should represent the same interval if we want an accurate representation of the data. In the data provided one record only represents one second for about half the records, the visualization in the thread question does not take these factors into account. Without taking them into account, the resulting chart is distorted.
Here is an example of the difference between Raw and Densified (note that with the Densified version we can see the moments when there was no movement, speed stays at 0 for a bit of time, and elevation is flat):
Attached is my complex route with a way to densify the data with respect to each second of time, and then interpolate the coordinates and elevation. An additional chart with distance as the X-Axis was also created.
Side note, instead of using a Multi-Row Formula tool twice, one option is we can use a Join Multiple on record position keeping only records that join from all inputs, with one data stream skipping the first record. I am not sure if it is more efficient than multiple Multi-Row Formula tools.