Ozzie Sport

Yesterday, I was having a conversation with some one at UCNISS about relationships between people and how these relationships, in sports, be almost predictive in terms of helping to determine sport performance.  Who coached a player?  Are their nodes of really good players that can be traced back to certain practices and coaches and cultures that could help explain why a team, much further down the chain, does well?  I don’t really have the data on that from a performance perspective.  What I do have is data regarding athlete and team follower patterns.  My most recent data, and I’m busy updating that now for the future, comes from late December 2010/early January 2011. This was a list of about 300 twitter accounts with the full list of accounts looked at being found on this post.  To analyse relationships, I took the list of twitter accounts, searched through every file that had an account on that list’s complete list of followed, removed all names that followed the account that were not on the list of Twitter followers.  Each occurrence was then put on a separate line. For example @harry_o, AFL, Collingwood -> @stkildafc, AFL, St. Kilda Saints could be a line.  I then created a chart using nodeXL. This was a complete mess because there were over 19,000 follows: Australian sport accounts follow other Australian sport accounts and they follow them a lot.  To make a better, more potentially meaningful chart, I eliminated the individual accounts as the means of looking at relationships and instead used the leagues.  This resulted in the chart below.

The red lines that you see are all extending from the AFL node: AFL accounts are following or are being followed by accounts from many Australian sporting competitions.  This chart doesn’t really give an idea of the weight so the table below shows the number of lines and the size of each node by looking at the league (consisting of the combined total of all accounts associated with it) and the followed (combined) that follow an account related to that league.  Why does this matter?  Example: More AFL related accounts follow a Cricket Australia related account than Cricket Australia people are following AFL related accounts.  I’d guess this is the Shane Warne effect.  I’m only including the top 105 connections, where there are 20 or more lines for a particular intersection.  The whole thing has 697 pairings, with 1 share pairings for things like Netball Victoria -> Horse racing.

I’m not entirely certain what this tells you. For me, it shows that like accounts follow like accounts. An account about the AFL is more likely to follow other AFL accounts than other leagues. It also shows that Australian sport on Twitter is very interconnected. For the major sports, there do not appear to be huge pockets of isolation away from other sports with the major expectation possibly being the AFL.

There might be something going on with the AFL. The NRL has issues in that AFL does not appear anywhere on its follow list. I don’t know where or why this is. There are NRL accounts that follow the VFL. The ANZ Championship doesn’t appear to be following the AFL either. Sport people following the AFL appear to be baseball players, tennis players and skateboarders. Confused.

Related Posts:

• One week Twitter change for Australia/New Zealand sport related accounts
• Most popular Australian (and New Zealand) sport related Twitter accounts (January 2)
• Most popular Australian (and New Zealand) sport related Twitter accounts (December 11)
• Most popular Australian (and New Zealand) sport related Twitter accounts (January 1)