I start off 2012 with a look at an article from 2011 — this piece in the Royal Statistical Society's Getstats campaign by Leeds University professor Bill Gerrard on his experiences as a statistical analyst with professional baseball, soccer, and rugby teams. Gerrard had famously partnered with Billy Beane on some analytics work with the Oakland A's ownership group but has since transitioned to rugby and cricket. It's worth reading in order to appreciate better his role in the team and the challenges involved in integrating statistical performance analysis in a professional sports team.
Some of the lessons learned are quite sobering for a startup company such as mine — access to in-match data to develop soccer analytics tools is difficult to achieve without having already a developed tool that has been demonstrated to be effective, which of course requires in-match data. Another is that a solution cannot be imposed entirely from outside, but rather integrated with existing assessments made by the technical staff. I disagree slightly with Gerrard when he says "Only the team coaches are in a position to produce effectiveness data." Perhaps third-party data suppliers aren't able to do so, but I think that analytics companies such as mine can provide more sophisticated measures of player and team performance that is closely integrated with the team's game plans. In 2012 I aim to make more progress on this kind of analysis.
Happy New Year, everyone! And in an effort to spur more discussion (borrowed from David Cummings), what are your thoughts on the role of statistical performance analysts in professional sports teams?
(CORRECTION: I corrected some factual details about the nature of the business relationship between Gerrard and Beane. Thanks to a high-ranking official for pointing that out.)