Basketball and Business Data Analytics Solutions for the NBA
A tool to predict the entertainment value of basketball games and other innovations for the NBA.
National Basketball Association, United States
Big data and analytics are making huge inroads into sports. From an athlete’s rate of respiration to the timing of a touchdown, the speed of a lap and the percentage of first serves won, sporting activities produce a wealth of statistics. Understanding and learning from them can make all the difference between winning and losing.
These days many major sporting teams have analytics experts on their payroll who process the data and help their side turn it into a competitive advantage, whether in real-time during an event or in training, pre-match preparation and recruitment.
Sport probably hasn't even scratched the surface when it comes to realizing the enormous potential of data. And one entity that's very excited about its possibilities is the National Basketball Association (NBA) in the US.
Basketball and Business
In September 2017 the NBA held its second hackathon for students and recent graduates from 55 colleges and universities across the US and Canada. The open innovation contest presented participants with some of the analytics-based problems the league faces on both the basketball and the business side.
They were posed by the NBA League Office and on the basketball side, prompts included modeling shot distributions and proposing rule changes, while on the businesses side, participants were tasked with creating business intelligence tools to predict game entertainment value and customer value.
"Analytics continue to play an increasingly important role at the NBA," said Evan Wasch, NBA senior vice president of basketball strategy and analytics. "I am excited to see the innovative ideas that come from a collection of the brightest young statisticians, engineers and developers."
Hundreds of number crunchers worked for 24 hours and their final submissions were reviewed by members of the League Office Analytics group. Three finalists for each track were selected to give presentations to the audience and expert judging panel. Following their deliberations, first, second and third places for each track were decided.
The first place of the basketball track was 'Predicting the Future Shot Distribution of the NBA' by Stanford Sports Analytics Club. The team analyzed the distribution of shots taken across the league to predict what the average teams’ short chart would look like ten years into the future. They also recommended a rule change related to the shot value of mid-range jumpers.
The business track first place was awarded to DataBucket for their 'Predicting Entertainment Value of NBA Games in a Real-Time Interactive Tool'. They defined four components of "entertainment value" and predicted each with independent regression models. The models were strengthened with external data from Google Trends and Instagram’s open API as input, along with stats for games.
There was a mixture of prizes for the winning finalists, including prize money and expenses-paid trips to NBA matches.
Next Story »