Saturday, December 8, 2012

Midgame Woes

That game was a roller coaster ride.  It started great and ended great, but the midsection was ugly.  I can't complain too much.  The Cats are 7-0 with two wins on the road against major conference opponents, even if they are not teams at the top of those conferences.   It was especially nice to see Mark Lyons respond the way he did.  The guy has grit.  What I like about this team is that you never who is going to have a big night.  I thought Brandon Ashley made some critical plays when we needed them, too.

Speaking of the game's middle section, this a pattern that is not unique to this game.  In fact coming into the game, the Cats have shown strength at the starts and ends of games, but weakness in the middle, particular in the latter part of the 1st half.  The graph below shows point differentials for ten minute segments of the first seven games.

For the last ten minutes of games, Arizona has outscored its opponents by a blistering 57 points, or more than eight points per game.  For the first ten minutes of the game, they have also been dominant, with a 48 point differential.  What is astounding is minutes 10:00 through 20:00 where the point differential drops to eight; the last ten minutes of the first half has been a trouble area.   Is this due to poor offense or poor defense?  Well, it turns out it's both.

The graph above shows field goal percentage for the Cats and their opponents for the same segments of games.  Notice that for every quarter of our first seven games but one, the Cats are on top.  Again, for the second half of the first half, the Cats not only shoot the ball poorly, but they struggle on defense as well.  In fact, our opponents have shot the ball better than we have during that time segment, and by far this has been our weakest part of the game defensively. Again, we are dominant at the start and end of games. Remarkably, for the last ten minutes of the game, the Cats are shooting it at 54%

What's causing this pattern?  I have no idea. It might just be a statistical fluke, or perhaps it can be attributed to substitution patterns.  It will be interesting to see if it holds up with a larger sample size.

No comments:

Post a Comment