Entertainment - Written Thursday, August 27, 2009 by Denis Hancock - 11 Comments

Starting the comparison of NBA teams on Twitter

A couple of weeks ago I explained why how the NBA – the league, the teams, the players – uses Twitter would be a fascinating and fun research topic. With the help of my colleague Yuan Ding, we’ve been slowly building the data set for this research, with an early focus placed on comparisons between the “official” Twitter accounts for each of the 30 NBA franchises.What makes this data interesting is that while comparing one company to another is inherently difficult (since business models, brand positioning, etc. vary so widely), the underlying objectives of each of the 30 NBA teams should be more or less the same. In turn, if (say) the popularity of different team accounts vary widely, we have a bit more of a “controlled experiment” in which to explore the reasons why.

While we haven’t had time to get into much of the analysis yet, I think we’ll have a lot to work with – the popularity of different team accounts do vary widely.  Using simple follower counts as a starting point (we’re working on a “fan score” to index these I’ll talk about another time, but the basic idea is to account for the fact small market teams should naturally have a smaller audience), here’s the raw statistics on a 29 team sample (Golden State was excluded do to lack of use) as of August 12th:

• Average number of followers: 49,877
• Median number of followers: 7,871
• Lowest number of followers: 2,921
• Highest number of followers: 782,019 (has since grown to 875,510 and counting)

Continue…