How can a background in basketball analytics be utilized to explore complex sport science questions?
After working together for the past few seasons Senior Manager of Basketball Analytics for the LA Lakers Phil Chang and Fusion Sport’s Dr. Marcus Colby share their experience and key learnings in this presentation for the 2021 Sports Biometrics conference.
In addition to theory, the presentation provides an insightful guide to assist in building similar data analytics frameworks to inform your performance optimization strategies.
Watch now below and continue scrolling for the presentation summary from Dr. Marcus Colby including links to relevant resources and further reading.
For the vast majority of my consultancy I’m used to working alongside performance and medical staff, so when I started working with Phil Chang a few seasons ago it suddenly created a whole new dynamic for looking at performance optimization given his analytics background.
Phil and I truly enjoyed putting this together and hope you enjoy the discussion as much as we did, and that the below summary provides further guidance to assist you in your own data analytics projects.
Learn more about Dr. Marcus Colby
Utilizing Data Analytics for NBA Teams
Diving straight in, Phil and I discuss some of the key foundations from which we built our optimization strategies specific to basketball, the NBA, and the LA Lakers’ player and team dynamics. Some topics covered:
- Formulating optimal work-to-rest time
- Utilizing league-wide workload data
- Exploring the age curve and player projections
Analyze your schedule
This topic we dive into quite thoroughly and something I want to stress is that your schedule dictates a lot of your high performance program. Analyze it in depth and be sure to leverage your takeaways proactively when managing athletes.
Usable Data: Planning your Analytics Projects
Phil makes some brilliant points in the presentation around choosing the right data to look at and keeping your eye on the ultimate goal and application of the data.
Similarly, don’t be afraid to utilize the assistance of front office staff who may not have a sports-oriented background. Zooming out and looking at your data through a new lens can lead to new ideas and strategies, and help you remove the sport science bias.
Resources and Further Reading:
- Learn about the multi armed bandit problem
https://lilianweng.github.io/lil-log/2018/01/23/the-multi-armed-bandit-problem-and-its-solutions.html
- Clustering NBA Players
https://sports.sites.yale.edu/clustering-nba-players
- Archetypal Analysis
https://fusionsport.com/blog/archetypal-thinking/
- It’s a Hard-Knock Life: Game Load, Fatigue, and Injury Risk in the National Basketball Association
https://www.researchgate.net/publication/325207204_It’s_a_Hard-Knock_Life_Game_Load_Fatigue_and_Injury_Risk_in_the_National_Basketball_Association
- Long-Distance Traveling in Basketball: Practical Applications Based on Scientific Evidence.
https://www.researchgate.net/publication/344544877_Long-Distance_Traveling_in_Basketball_Practical_Applications_Based_on_Scientific_Evidence - Air ball R package – Jose Fernandez
https://github.com/josedv82/airball