Embedded Files
NCAA: Using Matchup Data to Gain Competitive Advantage
NCAA: Using Matchup Data to Gain Competitive Advantage
Overview
Research project studying the viability of the use matchup point differential data during a basketball game. Specifically this project looks at lineup matchups in the first half of a collegiate basketball game and their potential insight into second half matchup performance. For this project, matchups are defined to be every unique combination of 5v5 lineups on the floor at the same time.
^Click on the image to be redirected to the Google Collab Python Notebook
NBA: Unassisted Scorers are Team Stars
NBA: Unassisted Scorers are Team Stars
Overview
Tableau dashboard providing insight into the relationship between unassisted scoring and offensive prowess in the NBA. Players with the highest share of baskets that are unassisted, also happen to be the leading scorers at their respective teams. The dashboard displays the "green light" concept since star players have 'a pass' to produce offensively in solitary, whereas other players in the NBA do not.
^Click on the image to be redirected to the Tableau Dashboard
AUTO-Scout
AUTO-Scout
Overview
NCAA: AUTO-SCOUT IS A PRIVATE PROJECT.
AUTO-Scout is a Python package that generates player tracking stat scouting reports for all team rosters in NCAA MBB as well as for historical reports for all players in NCAA MBB since 2014.
The scouting reports are outputted in .pdf format and include data on player shot efficiency, AST%, TOV%, and other stats, per possession type. The software also includes the option for blank sports on every page for the addition of commentary or shot-charts, to be added manually. The program runs on average for 3.5 minutes to scrape the tracking data, roster images, play-by-play data and generate a scouting report for a full roster for a given season (can be current season).
^Click on the image to be redirected to a couple of output examples for NCAA AUTO-Scout
Page updated
Report abuse