What Dr. Paul Beckman and Jennifer Chi propose is this article is perhaps not revolutionary, but still rather intriguing to me: they have discovered a positive correlation between the number of connections a professional baseball player has in the network of Major League players and their subsequent offensive performance. In order to demonstrate this, Beckman and Chi extracted, “opening day rosters for every MLB team from 1979 to 2004.” They subsequently added each player as a node in their network and connected nodes (i.e. players) who were on the same opening day roster. They acknowledge the imperfections in this design due to mid-season trading, yet this approach can still yield meaningful data. Lastly, they used the Pearson product-moment correlation (1.0 means highly positively correlated and -1.0 means highly negatively correlated) to determine if there existed a linear relationship between the network connections and offensive performance. Their results demonstrate a significant positive relationship between ‘connectedness’ and batting average, home runs, runs batted in, and slugging percentage. Each had a coefficient of 0.38, 0.48, 0.47, and 0.37 respectively. This conclusion also makes intuitive sense. If a baseball player, like Derek Jeter, has been exposed to many different players, he has probably picked up on many of their tips and tricks and incorporated them into his repertoire, thereby improving his offensive performance.
Then, you might ask, why is this article interesting at all? I found the article particularly interesting because of the way it then allowed me to take the principle of ‘The Strength of Weak Ties,’ that we learned in class, and think about how it might apply to this scenario. The researchers already found a correlation between number of connections and offensive performance. However, given what we have learned in class, I would also conjecture that the number of weak ties might have an even higher correlation with performance. If the causal mechanism that underlies this improved offensive performance is increased access to different tips and tricks from players, weak ties would offer a player greater access to a wider range of information. Weak ties would connect a player to new clusters of distant knowledge he would otherwise not have access to. In the same way that Mark Granovetter discovered people often get job offers from weak ties, it might follow that Major League Baseball players would receive their most hopeful hitting tips or advice from weak ties. Although the researchers did not measure the strength of each tie, I believe that including this in a further study could produce intriguing results.
Furthermore, by using a network to represent all of Major League Baseball, it would then be possible to track how certain trends (e.g. a particular batting stance, wearing batting gloves, or changes hitting approach) have risen and fallen to popularity by way of the network effect. How might social influence and selection play into the trends we have observed in Major League Baseball throughout the years? How might one represent these trends using a network structure like the one created by Beckman and Chi?
In this manner, I view this article as a highly interesting jumping-off point for further inquiries into the subject of network effects and professional sports. This is not to say that Beckman and Chi’s findings are not interesting, because I do think they are; however, I think the potential extensions of their research are even more intriguing.