College - Author 1
College of Science and Mathematics
Department - Author 1
Statistics Department
Degree Name - Author 1
BS in Statistics
Date
6-2013
Primary Advisor
Matthew Carlton
Abstract/Summary
This paper analyzes and calculates an advanced NBA statistic that is becoming more and more widely used in the NBA. The Adjusted plus-minus (APM) statistic measures a player’s contribution, independent of all other players on the court. The most appealing aspect to the APM is that it only attempts to capture how a team’s scoring margin changes with a particular player on and off the court. Scoring margin in basketball effects winning percentage greatly, so it only makes sense that players with high APM’s will increase their team’s scoring margin and, therefore, help win games. The APM statistic is not an exhaustive approach to finding NBA talent. Many quantitative and qualitative factors, in addition to the APM, should be investigated before acquiring a player. However, the APM needs to be addressed and carefully examined along with these other factors. In any sport, basketball included, it is natural (and human) for experts to just believe what they see. However, much of what NBA scouts and upper-management see they subconsciously over or under-value, and that can lead to errant judgments. Paying an average or below average player way too much can be a huge mistake to an NBA franchise. Therefore, a comprehensive in-depth analysis must be conducted both statistically (with traditional and advanced measures) as well as visually. The APM statistic is growing quickly in popularity as a quantitative tool, and will continue to do so, because it measures better than anything else presently something that is critically important; how much does a player actually contribute to team success? Armed with this “arrow” in his quiver, the smart GM can make more informed personnel decisions for his team by more effienctly allocating his limited resources. Some possible very under-valued players are examined in this paper that could potentially help a GM make better decisions. This in turn will maximize his chances of building a winning team over time.
URL: https://digitalcommons.calpoly.edu/statsp/35