We’re going to move towards thinking about sports for a bit with a few objectives in mind:
- Visualization
- Ranking
- Prediction
Last year’s men’s Big South basketball season
Let’s start with groovy look at last year’s men’s Big South basket season:
This kind of chart is called a chord diagram. The beginning and ending width of a ribbon connecting team A and team B form a measure of how well team A performed against team B, as described below. The color of a ribbon is determined by the stronger team. The total width of an arc for a team is a measure of how well the school did in the conference.
You can hover over one of those ribbons or a team logo to get a bit more information about the games played between the teams. You could find, for example, that UNC Asheville leads the conference at 16-2. All the arcs emanating from UNC Asheville are blue, since they’ve scored at least as many points as all of their opponents.
It looks like UNCA did pretty well! They won the Big South tournament, too:
From regular season games to tournament prediction
There’s a company called Kaggle that runs data based competitions. I’ve participated in their March Madness competitions for years. You can see my competition list on my profile page.
I’ve also built a couple visualizations on top of Kaggle’s data. As a simple example, here’s where I stood relative to the competition in 2022:
Here are a couple more complicated examples:
Over the next couple of weeks, we’re going to take a look at sports data. In particular, we’ll learn how to
- Rank teams that play each other and a common set of opponents.
- Make predictions for future games and tournaments.
Of course, we’ll look at some data visualizations to see if they can help illuminate the situation. Here’s a simple, not so flashy example that I wrote a few years ago: