A challenge in evaluating fan bases in professional and college sports is how to adjust for capacity constraints. Unlike most consumer categories, teams have a limited number of seats to sell. One way to get around this issue is to look at team revenues. But this approach also has some strong implicit assumptions in that we must assume that teams are trying to price in a manner that maximizes revenue.
The world of social media provides an opportunity to look at fan base support without worrying about capacity or pricing issues. To look at NBA teams “social media equity” we collected follows and likes from Twitter and Facebook. We then created a statistical model that predicts these measures of social media engagement as a function of market size, tweeting activity and team performance for this past season and for the season before that. We then compared each team’s actual follows and likes against the model predictions. This method attempts to control for short term fluctuations in winning percentage and market differences.
The top team in terms of social media equity is the LA Lakers. The Lakers crush the competition both in terms of raw numbers and in our model. In second place, we have the Miami Heat. This one is interesting, and we suspect that the Heat results may be a bit misleading. While the Heat does very well currently it is not possible to separate out how much of the social media equity is driven by the team versus by LeBron. This is something to watch as we collect more social media data over the next few years. In third place, we have another non-surprising result in the Celtics.
It is the next three teams that are surprising as Golden State ranks 5th, New Orleans ranks 6th, and Charlotte ranks 4th. The case of Charlotte illustrates the value of our model based approach. In absolute terms, Charlotte performs relatively poorly in terms of social media metrics. However, when we adjust for team performance and market size, the team does fairly well. This indicates that the Charlotte market has fairly resilient fans, and likely speaks to the potential of the market if a consistent winning team is developed.
At the bottom of the list, the most surprising result is the New York Knicks’ 27th place finish. This is doubly interesting because when we ranked fan bases in terms of “economic” support, the Knicks were number one. What these two results imply is that the Knicks’ fan base is economically valuable but not engaged (at least in terms of social media). The Knicks play in the largest market but have only about 20% of the social media activity of the Lakers.
There were a couple of other interesting findings from this study. First, the number of Twitter followers was uncorrelated with the number of times a team tweeted. This suggests that fans follow based purely on their feelings for the teams, rather than the entertainment of following an interesting Tweeter. We also found a very high correlation between the two social media platforms as the social media equity estimates across the two platforms exceeded 0.91. However, when we looked at the correlation between the social media equity and the economics based fan equity the correlation was just 0.3. We will leave this disconnect between social media and revenues for a future post.
Mike Lewis & Manish Tripathi, Emory University, 2013.