Iron Viz – Better to be a Drogba than an Owen?

Category : Tableau

My winning entry for the Wiki Data Iron Viz Contest, looking at how quickly football fans write off young players. Taking data from Wikipedia, the entry examines the scoring records of 50 of the most accomplished strikers and goalscoring midfielders of the past twenty years,

Background to the Viz

Any Wikipedia table data is difficult to scrape as each user defines the table structure. So if you want to scrape lots of table data from multiple pages then it becomes a nightmare. We also had to calculate an age for each season, which is harder than it seems, and is not perhaps 100% correct.

And as my Python skills are a bit weak, for this task, the data for each player and for each season was scraped using Outwit Hub, a player at a time, after the tools KinimoLab and ImportIo had failed to make sense of the different table structures on each player page. We only had time to scrape 50 players.

As with all Wikipedia data, I have no idea this data is correct!

I took inspiration, including using similar fonts and layout, from the interactive stories of the New York Times and Washington Post, and the way they combine simple, but effective, charts with detailed analysis.

I wanted a viz that had something to say about the data, that wasn’t afraid to use lots of text, as well as including a viz that let the user find their own story.

There are some limits to how you can do this in Tableau (well for me anyway). I couldn’t use Story Points as it created two scroll bars which would be confusing to use.

The dashboards sheets also have a maximum height of 4000 pixels meaning I had to split the content over a few pages, and I found changing the layouts on the dashboards using containers frustrating, as moving one thing changed everything else.

Thanks for the signifiicant contribution of Paul Tomkins.

Leave a Reply

Your email address will not be published. Required fields are marked *