The Future of Statistics: Player Tracking Data in the NBA
26 Mar, 2019
The Future of Statistics: Player Tracking Data in the NBA
One of the most prominent experts in basketball analytics shares the power of player tracking data to extract insights for measuring athletes, formulating game strategy, and evaluating team performance at an elite level.
Tracking data is a powerful analytical tool in the world of basketball, and Luke Bornn is one of the foremost figures in this innovative area of sports data.
As Vice President of Strategy and Analytics for the NBA’s Sacramento Kings, Bornn’s role is to essentially help the Kings make more objective decisions through advanced tracking data and statistical modelling.
Until very recently, Basketball coaches were driven by box score data, which is simply a structured summary of the results of the game. In comparison, tracking data can measure things that are happening off the ball, while also measuring spacing, movement, and team formations.
“Tracking data really catches things within the sport that coaches talk about and the way coaches think,” said Bornn.
“Coaches don’t always talk about the outcomes that are measured at that level of data, but they do talk about movement and spacing and formations. So tracking data really allows us to measure features of the game which are much more closely aligned to the way coaches think.”
“Tracking data really catches things within the sport that coaches talk about and the way coaches think”
In the past teams would only measure data for in-game performance, whereas with tracking technologies NBA franchises can obtain far more data points.
Bornn explains how measuring training performance data leads to much more effective insights and a higher accuracy and understanding of shooting skills.
“If you think about the data that you’re getting within a basketball match, you’re getting maybe around 200 shots in total, and if you’re thinking of looking at measuring skill in a particular area of the court, you need a lot of more shots than that to know if a player is good or bad,” said Bornn.
“A typical player shooting a corner three point shot, you might only see one or two shots from there in a game, so you would need to see several games to know if a player is good or bad at shooting corner threes. However, if you look within practice, many of these guys are taking thousands and thousands of shots every day, so using tracking data from training allows you to understand a players skill level much more rapidly than if you just look at a match.”
“There are a lot of clubs who are spending in this area, whether its purchasing data or hiring analytical talent, that don’t have the culture or the process in place to make decisions from this data,” said Bornn.
“You could have a significant investment that is not impacting your club which is not ideal.”
Secondly, is having analytical talent in your staff to be able to handle complex data.
“Five or ten years ago clubs would hire people with very basic technical skills, the ability to manipulate excel spreadsheets for example,” said Bornn.
“But in the past few years, especially with the advent of tracking data, this new data has such a volume and complexity that it requires a whole different technical skill set. It requires the ability to handle massive data sets and also spatial temporal modelling.
“The talent that is needed don’t just have simple analysis backgrounds, but full machine learning and statistics backgrounds so that they are able to model these big complex systems.”
Click here to see how Hudl's basketball solutions can work for your club.
