The History of Analytics

I was reading this article about the recent Sloan Conference, where they the writer talks about an interview of Bill James, and his discussion about the importance of luck in analytics (ok - I'm paraphrasing that).

But only a bit.

The example used was, at a specific point in time, analytics said that two baseball players were identical.  One was selected by the Red Sox and he has gone on to win two titles and is now a legend in that town.  The other didn't do much past that specific point in time.

This got me thinking that, no matter what, Analytics is always historical.  Any analysis is based on what has already happened.  Any projection is based on historical data.  The only thing you know for sure is that the information you have is absolutely true at that moment.  At least, you are pretty sure about that.

The point is that Analytics, which is becoming more and more fashionable but has been around in elite sport for at the very least half a century (why is it when an activity people have been doing for decades is given a new name its considered a new thing?), is only ever about the past.  That's all it can be.  It can't tell you what will happen in the future.

Having said that, its REALLY useful to know what happened in the past!  Just make sure that it informs your decisions about the future, rather than determines them.

(Photo Credit)