2015 ∙ 4 ∙ 25  /  Work

Comparing the Elevation Profiles of Two Ridiculous Runs

A colleague of mine is contemplating running Sinister 7  this summer. For the uninitiated, as I was, Sinister 7 is a punishing 161km run through mountainous terrain that starts in Crowsnest Pass, Alberta and ends up in a different time zone (not really). For most people, the question is why on earth would I do something like this? His question to me was, 'do you know how to extract elevation data from Google Earth?'

What he really wanted to know is, having already run the  Canadian Death Race, what could he really expect from Sinister 7? How similar are the distance and changes in elevation between the two courses?

In order to provide him with a comparison I needed to extract all of the longitude, latitude, and altitude data provided by each course in the form of a Google Earth file. I processed the files in various ways.

  • I reshaped the coordinates so that they were in tabular form.
  • I extracted the legs of each course separately so that I could group the coordinates by leg and assign them an order ID. Extracting separate files meant having to merge them back together.
  • I used the Haversine Formula (great circle distance) to calculate the distance between each set of coordinates and the next in order to calculate the overall distance of each course and each leg. The reason for this was to normalize each course onto the same scale - one data point per kilometre - for better comparison / presentation, rather than simply calculating the distance.

I'm not any closer to knowing if he should go, but I do have a better sense of how these two ridiculous runs compare.

> Check out the dashboard on Tableau Public.

Caveats: Using the great circle distance between data points to calculate the overall distance gives me fewer kilometres for both courses than stated on the respective course sites. The data files I used have a significantly different number of rows: 10,767 for the Canadian Death Race and 2,204 for Sinister 7. This difference will likely affect the elevation changes calculation, skewing the results in favour of The Canadian Death Race.

@datadaydream