Monday, January 10, 2011

The team that can't shoot straight

While there is undoubtedly much moaning and gnashing of teeth in Hoya-land these days, it's important every once in a while to step back and wonder whether your current mood is based upon small sample-size results.

To wit, one of the biggest drivers for Georgetown's current swoon is the complete loss of an outside shooting touch.  If you take a look at the Team Stats breakout page, you'd find that the team shot 42.6% on 3FGs in the non-conference season, but has only hit them at 25.7% over the first four conference games.

To put this difference into some context, I've gathered the final non-conf. and conf. 3FG shooting percentages for all Georgetown seasons since the 3FG game into the game (1986-7).  This may have been a herculean effort in the past, but thanks to the stats pages over at the Georgetown History Project, it's now a trivial exercise.

As you can see, how well a team shoots from behind the arc in the non-conf. portion of the schedule is generally a good predictor of how well it will shoot in Big East play.  Until this season (that big black dot in the lower right).

How well are the two shooting percentages correlated?  Glad you asked:

There's a bit more going with this graph, so allow me to explain all those lines:
  • The blue line is the 1:1 line.  Points above the line are when the team shot better in conference, points below the lne are when the team shot better out-of-conference.  Since competition in the non-conf. portion of the schedule is weaker, you'd expect most of the points to be below this line, which they are.
  • The thick red line is the linear fit of the data, for all seasons except this year.  That R2 number refers to the goodness of fit of this line.
  • The dashed lines are the 95% prediction bands, based on the fit.  That is to say, there is only a 1 in 20 expectation that a data point would fall outside of the dashed lines.  And this is just what we see:  there are 24 seasons that make up the dataset, and only one would be considered an outlier (1989-90).

In the four Big East games played so far, Georgetown's lousy outside shooting is an extreme outlier.  As we said at the top, the team is shooting 25.7% on 3FGs, while we'd expect them to be making 37.4%, based upon the OOC performance.

Using the stats from previous seasons, there's about 0.12% chance that Georgetown will end the conference season shooting this poorly.  In terms of odds, that's 850:1 against.

So chin up Hoya fan!  It's going to get better.

It has to.

I hope.


    1. Excellent analysis! A large part of the struggle appears to be self-inflicted by players taking unnecessarily long, low-percentage 3's. That seems easy to be an fix....or so it would appear.

    2. Agreed, very good analysis, and looking even moreso at this point in the season.

      Also just wanted to say that I hope you don't get discouraged by the lack of comments on here. I've never commented before but I read this regularly.