Saturday, December 12, 2015

Introducing FUN

Later today at T-Rank I'll be adding a team stat I call FUN, which stands for

Fortune Unexplained by Numbers


Failure Unexplained by Numbers

This is an equivalent stat to what Ken Pomeroy calls "luck." Here's how it is calculated: T-Rank goes back in time and calculates an expected winning percentage for each team schedule so far, based on their T-Rank profile. Then it compares it to their actual winning percentage. The difference (expected minus actual) is their FUN.

I'm avoiding the word "luck" here for a reason. If I call it luck, that implies that the T-Rank is a rather absolute reflection of a team's quality, and that any deviation from T-Rank's expected winning percentage is simply random variance, or luck.

Undoubtedly, that is a big part of the story. But it's probably not the whole story. 

Let's look at a team like Maryland, which last year was very "lucky" in that it's actual record was much better than it's Kenpom / T-Rank expected record. The reason for this difference is pretty easy to suss out: Maryland went 12-1 in games decided by 6 or fewer points. That's kind of ridiculous. Most of the time, teams will go about .500 in close games -- even very good teams.

But maybe there's something about Maryland that makes them particularly good in close games. Whatever this something might be, there's no chance Kenpom or T-Rank will reflect it. It could be random variance, just plain luck. But it could be something else.

And with Maryland, there's actually a pretty obvious hypothesis about what that something might be: an all-American point guard (Melo Trimble) who can get to the line at will, and hits 90% of his free throws. This is something that is particularly useful in close games. And it turns out that there's good evidence that Trimble does indeed get a disproportionate share of his free throws in crunch time of close games:
So this is the kind of thing that might be something other than random variance that causes a team to outperform T-Rank expectations. 

The other reason I'm calling it FUN is that variation from reasonable expectations is one of the things that makes college basketball so interesting and, well, fun. After all, college basketball is where the unexpected becomes ordinary. Or so I've been told.

[T-Rank says: FUN is luck, nothing more.]