This is a review of the NBA research conducted by James Tarlow, applying multiple least square regression, including a lagged dependent variable.
Within sports, it is always assumed that experience improves a team and leads to championships. This is especially true in the NBA. Teams made up of younger players are viewed negatively for their lack of experience. When teams comprised of younger players losses a playoff series it is attributed to a lack of experience. Young coaches with less experience are criticized in the same manner.
Analysts and fans always seem to look at the experience of individual players and never the aggregate experience of the team. Teammates who have played together for a long period of time gain an innate sense of how their teammates will react during different situations. This improves team efficiency but to what extent?
An econometric study is undertaken to examine the relationship between experience and winning in the post season. It first looks at which, if any, types of experience have a positive effect on team performance and what extent that relationship has on winning games.
The study uses data from the 1979-2009 seasons including 4,020 players. Three models are developed using multiple least square regression, including a lagged dependent variable. The first model looks at player experience. Player experience is divided into two categories, NBA experience, and postseason experience. Variances of player NBA experience and player playoff experience are included in order to control for the diversity of player experience within each team. Coaching experience is examined in the second model and is also divided into NBA experience and postseason experience, but also includes the coach’s winning percentage for postseason games and coach tenure. Lastly, the third model analyzes chemistry, which is defined as the number of years the five players playing the most minutes during the regular season have been teammates playing for their current team.
Results from the first model regarding player experience indicate that NBA experience does not contribute at a statistically significant level. However, postseason experience is a different story. Results indicate that player postseason experience helps a team make it into the playoffs but does not increase their ability to win postseason games.
Results from the second model regarding coaching experience are similar to those regarding player experience. Regular season coaching experience does not appear to have any effect on postseason success. However, having postseason coaching experience and the amount of that experience does appear to contribute to winning in the postseason.
The third model dealing with chemistry determines that chemistry is statistically significant in regards to postseason wins. Each year of shared experience between teammates increases the expected number of postseason wins.
All of this information helps coaches and analysts when making decisions regarding trades. Knowing the effects that the experience of player or a coach may or may not have regarding postseason success will help make wiser decisions. The chemistry model provides more potentially useful information as it indicates that teammates who play together for an extended period of time will get better and improve a team’s efficiency. This would seem to be an argument against making trades simply because a team is not performing well at the moment. Leaving the current players together, rather than making trades, may prove to be a wiser decision in the long run.
Chemistry is not a variable that is typically considered by coaches when making decisions regarding trades, but perhaps it should receive more attention.
Analytics Used: Econometric Study, Multiple Least Squares Regression including a Lagged Dependent Variable
Find out how Sports Analytics Expert Victor Holman can give your team the competitive advantage.
How mature is your team’s analytics program? Take the Sports Analytics Maturity Assessment.
Discover the Groundbreaking Sports Analytics Application and Framework coaches and sports analysts are talking about!
Learn all about sports analytics in Victor Holman’s Sports Analytics Blog.