The Superstar Effect: NBA Players and Team Influence

I. Introduction
Individual influence in a group setting has been studied among numerous academic fields for decades. In particular, productivity is often seen as a positive externality, in which the efficiency of a single individual in certain environments may others to also be more efficient, leading to better performance of the overall group [1]. This increase in productivity can be observed in a variety of different environments, including but not limited to the work office, education, and even in a domestic setting. Specifically in sports teams, this phenomenon is coined “the superstar effect”.
In the past, there has been much debate regarding the actual impact of the superstar effect. These studies bring up many important issues that we will address before proceeding with our work. Among these questions are: Who qualifies as a superstar? How do we define a player to be “productive”? Under what metrics will we look at to determine the supposed effect of these players? All of these are valid points that we must carefully define to progress with our experimentation.

II. Background and Data Our first and foremost question is what we base our data and work on: how do we define a superstar? Past work shows numerous ways to address this topic. One common way to determine a superstar is to refer to the number of times he or she has made an All-Pro team, or a player who has played in the NBA All-Star Game for at least 50% of his or her years in the league [5]. Other studies look at fan preferences; for instance, superstars would be those with high All-Star votes [3][4]. In our study, we take a purely numerical approach; we will define a superstar to be a player with high numbers of points scored as well as various other player statistics, in comparison to the rest of the team. This last