With another amazing year of Wimbledon wrapping up this weekend, I wanted to share some interesting stats about the tournament, and what better way to show these than through Power BI?
Click on the image below to access an interactive Power BI dashboard and views stats from the matches played so far in the tournament:
Interesting Facts:
- 128 Total Players from 38 different countries participated in the tournament
- All players in the Quarterfinals are seeded (only one major upset – Djokovic)
- Andy Murray seems the most rested based on Total Sets Played, and Total Games Played
- France had the most players compete in the tournament – 16
- If Federer and Murray both win, it will mark the 26th time they play each other, with Federer winning their last matchup at Wimbledon in 2015
Building the dashboard took about 5 hours to get the color scheme, layout, and do some analysis of the data in order to highlight some interesting findings. The data is sourced from espn.com, and I used a neat web scrapping tool called import.io to get the data out and into a CSV file format that I could consume. After some data cleansing using SSIS and a basic SQL Database, I created an ETL that grabbed the import.io file and exported into a cleansed CSV file that became the data source in my Power BI dashboard.
As with most BI projects, the challenging part is making sure that 1) The data is available, 2) The data is in a format that can be consumed, and 3) The data is accurate. I spent quite a bit of time working with the import.io tool, and modeling the data in a way that allowed me to do the kind of analysis that I was interested in. Not all of the answers I was looking for were readily available in my dataset, and I had to spend some time deriving them through data transformations and calculations. While maybe not the most fun part of the project, nor what clients/end users see at the end, your ETL and data modeling are very important components of any BI implementation, and should not be overlooked. In the end, what use is a “cool” dashboard if it tells you nothing? 🙂