What is it about?
This article introduces a novel method to identify and define archetypes of players in the NBA. A weighted network clustering approach is developed to leverage the many types and categories of statistics, to include the relatively modern tracking data. The algorithm is used on historical data, through the introduction of tracking data, to identify and explore the evolution of player types in the NBA.
Featured Image
Photo by Barna Bartis on Unsplash
Why is it important?
Our analysis identifies eight historical player archetypes according to the tracking data era. It also identifies outliers in each season, from outstanding contributors such as Giannis Antetokounmpo (2019-20) and Stephen Curry (2015-16) to lackluster performers such as Nikola Pekovic (2014-15) and Manu Ginobili (2016-17). We show an improvement in Win Shares explanation in our archetypes, compared to traditional positions and previous research.
Perspectives
I hope this article appeals to both sports fans and technical aficionados, and everyone in between! It was a pleasure to research, and I hope the ideas are taken and improved upon for anyone inspired to do so! I'm always up for a hearty discussion and feedback, from both a sports and technical perspective. Cheers.
Megan Muniz
Colorado School of Mines
Read the Original
This page is a summary of: A weighted network clustering approach in the NBA, Journal of Sports Analytics, December 2022, IOS Press,
DOI: 10.3233/jsa-220584.
You can read the full text:
Contributors
The following have contributed to this page
