#LondonTUG and Visualizing Fitness Data

Posted by

Yesterday I had the pleasure of attending and presenting at the London Tableau User Group, which was hosted at Tableau’s office in Southwark.

While it feels like I am constantly talking about #MakeoverMonday and Triathlon as part of my daily conversations with friends, family and colleagues, this was the first time I presented on those topics to a reasonably large crowd of 100+ people. And it was my first live demo in Tableau to a big audience.

But thankfully everything worked as expected from a technical side and being passionate about those two topics helped me overcome the fact that my notes didn’t show on the screen, so I was flying blind. Somewhat.

You can find my presentation on the LondonTUG Workplace page if you’re interested. There are also some more or less amusing photos from my talk and the evening.

Live demo

Aside from my presentation, which focused mostly on MakeoverMonday to introduce the project to anyone who didn’t know it, I also did a live demo in Tableau, visualizing Fitness data. Given that Triathlon is ‘my thing’, I visualised a swim, a bike ride and a run, using different maps and a couple of ‘styles’ for the vizzes and the cycling dashboard.

Click on the images below to check out the Tableau Public dashboard



One fine day in Provence...

London running

The data for these dashboards comes from Strava and after connecting to it via Tableau’s web data connector (currently in beta), I ended up loading it into EXASOL for demo purposes, because that gives me nice and fast response times.

Below are a few screenshots of the fields available through the web data connector. This list includes some calculated fields I have added which you can identify through the sign before the field name:

The process of the demo

I had about 10 minutes for my demo so in preparation I created 3 sets for which I picked selected activities. Why? Because the 1,095 sporting activities in my data lead to a very long list to pick from when filtering, so I chose my activities ahead of time, created a set for the swim, the bike ride and the running dashboard, to ensure that I could easily filter to those activities in the demo.

I then used latitude and longitude for each viz, but chose different Mapbox background maps, depending on the viz.

How the vizzes were created and why I did what I did

I aim to make all my choices for creating a viz very deliberate, so the following paragraphs talk about the how and the why of the views you saw above…

For the swim I picked a satellite background to showcase the beautiful beaches of the starting and finishing location. For the purpose of the Tableau Public version (which comes without my narration), I also added some information above the map, including the date, location, distance and duration of this race.

To create the line of the swim, I created a line chart and placed Path ID on the Path shelf, before changing the line colour and removing the halo.


One fine day in Provence...For the bike ride I picked a simple Tableau map (light) during my demo but changed it to a Mapbox Emerald map for the uploaded dashboard, because I like the topographic look of it, which highlights the mountain region where I rode.
I added the map solely for the purpose of providing a reference point for where this ride happened. While most cyclists will know about Mt Ventoux, as it is the holy grail of road cycling, I wanted to ensure everyone else could also place it…

For the cycling dashboard I added a few additional worksheets. 3 key performance metrics, including the total elevation gain, the duration of the ride and the total distance covered. The placement of these sheets was deliberate so that the total elevation aligned with the peak of the mountain (note: total elevation covered the entire ride; Mt Ventoux summit by itself was only part of that, but I still like the visual reference it provides). The duration was placed in the middle to ‘stretch across’ the viz below, while the distance covered was aligned with the small map of the ride to show how big the loop was.

Lastly I included an elevation profile of the ride. I did this by placing Path ID on columns and elevation on rows, sizing it to ‘entire view’ and making it a bar chart. The Path ID is a field where my location was tracked every few seconds or minutes to create a data point along the way. This creates basically thousands of way points over time and by putting them on columns, you get a profile of the ride from start to finish, left to right.
By making this a bar chart (rather than a line chart), I was able to have a nice viz that looked like a mountain and a horizontal ‘slice’ through my ride.

I made the dashboard very wide but not very tall to ensure the profile of the ride (‘the mountain’) resembled in some way the actual gradient. I didn’t get this spot on, but better than the standard dashboard dimensions which made it look like an almost vertical climb…

London runningFor the run I selected a Mapbox Streets map which looks a bit like the Strava maps.

I created a line chart with Path ID on ‘Path’ and using the Activity Name on colour to highlight the different runs I had done in London over time. I also reduced the opacity of the map a bit to make the colours of the lines come out more and removed the line halo.


Some closing comments…

I had a really fun time presenting at the TUG and enjoyed meeting so many new people and seeing interested people who asked great questions and paid attention. That’s always nice when you’re the presenter :-).

If you have any questions about the above, about the data or the vizzes, please leave a comment below or connect with me on Twitter

At this point, thank you to Sarah, Paul C, Paul B, David and Nick for hosting me, for inviting me to speak and for bringing together a great event with an even greater crowd. I look forward to returning at some point, if you’ll have me back :-).

Thanks also to my Makeover Monday partner Andy, whose feedback in preparation has been invaluable and whose encouragement is always helpful!



  1. Hi Eva,
    Which data connector are you using for the strava data? I can only find the Information Labs one that doesn’t give all the dimensions and measures. I’d love to be able to look at my strava data in a similar way as you have done with the elevation profile


    1. I have used a beta version of a connector Tableau is developing, They haven’t published it yet for wider use, I’m afraid.
      Alternatively you could go through the Strava API


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s