How Big Data and Analytics are Transforming Elite Basketball

Published Jun-01-19

From devising game plans to improving player and team performance big data is changing the way basketball is being played.

Second Spectrum, United States

The Story:

How Big Data and Analytics are Transforming Elite Basketball During the last decade, big data has made big, impactful strides into basketball helping to improve individual and team performances as well as the way new players are recruited. The overwhelming majority of NBA (National Basketball Association) teams in the US now have data analysists on their staff and the league also runs an annual big data hackathon.

A Slam Dunk for Big Data

Among the NBA teams that have let data into the locker room is the Houston Rockets. Their ascent from mid-table mediocrity to being one of the sport's top teams isn’t just down to its squad of brilliant players. Big data has played an enormous part.

It all started at the back end of the noughties when the Rockets were one of four NBA teams that installed a video tracking system that mined raw data from games. When analyzed the data revealed which shots had the highest rates of success. They were two-point dunks and three-point shots from outside the three-point line.

Since 2008/2009 the number of three-point shots as a percentage of all shots in the NBA has increased considerably. During the 2017-2018 season, the Houston Rockets made 1,184 three-point shots, more than any other NBA team in the sport's history. This was a major contributing factor to their breaking their record for most wins in a season, with 65 victories.

Machines with a Coach’s understanding of the Game

Since the 2017-2018 season, Second Spectrum has been the official video tracking technology provider for the NBA. Cameras installed in NBA arenas collect 3D spatial data including ball/player/referee locations, movements and much more. Spatiotemporal pattern recognition technique identifies and augments data extracted from the video to generate insights for coaches and teams.

“There are things that basketball coaches want to know, and the problem is they can’t know them because they’d have to watch every second of every game and remember it. And a person can't do that. But a machine can," said Second Spectrum CEO Rajiv Maheswaran in a 2015 TED Talk.

"The problem is a machine can't see the game with the eye of a coach at least it couldn't until now. So, we've taught the machine things like passes, shots and rebound things that most casual fans would know.

“And then we moved onto things slightly more complicated, events like post-ups and pick and rolls....Now we've gotten to a point where today, the machine understands complex events we have taught a machine to see with the eyes of a coach."

Match Input

Another example of how data analytics works within teams can be seen at the Philadelphia 76ers Sixers where player development coach Ivana Seric who has a PhD in computational fluid dynamics develops daily strategies. She also writes programs that interpret data that the team and league supplies and sits in on the daily coaches' meetings. Her input plays a significant role in how the team plays offense or defense.

Big Data For Sporting Success

Basketball, like many sports, is becoming more data driven and just like a star player this information is crucial for success.

Share on      
Next Story »

What Can we Solve for You?