Smoke and Mirrors?

Happy Valley HoopsEvery time I look at the RPI/Pomeroy Ranking Chart, I get encouraged. How could you not with that lovely upward trend rising from the disastrous Orlando ashes?

But then I start to wonder. What exactly has caused this turnaround? And is it sustainable.

Let’s focus on the Ken Pomeroy’s number simply because there’s more information involved there. Pomeroy generates a rating number based on a team’s offensive and defensive performance against expectations. He then compares that number against all other D1 schools to find the schools relative ranking. One would expect that the rating number would track close to the ranking number. The chart below demonstrates that that is exactly what you find.

Happy Valley Hoops

For the rest of the analysis, I’ll eliminate the ranking number and work strictly with the rating number. What’s of particular interest is that the rating number has been pretty much consistently rising since Orlando. What that means is that the team has been exceeding expectations in each game since then (with a slight hiccup against St Joe’s). It’s not like it’s been a roller coaster up and down with a trend upward – it’s pretty much been a constant improvement. That’s a good sign.

The obvious question is “what has been driving this improvement?”. The first step in trying to analyze it is to break down the rating number into it’s two components. Here’s a plot of PSU’s offensive rating and defensive rating up against the overall rating. The overall rating is in red, defense in blue, and offense in green. In order to effectively compare the ratings I’ve normalized each value as a percentage of the range between the best and the worst team in each category.

Happy Valley Hoops

It’s readily apparent from this chart what’s going on. Our offense has been fairly consistent since the beginning of the season (the first couple of games can be expected to have a large variance since there’s only a small amount of data available). Its plot, in green, is pretty much flat.

The defense is a different story. It’s clearly what is driving the Pomeroy rating. It tracks closely with the overall Pomeroy rating and has improved in each of the last four games.

The natural reaction is to say that the last three teams that we have played haven’t been particularly good so we should expect to have good defensive performances against them. But that overlooks one thing – the fact that our opponents weren’t very good was already factored into the expected performance of the defense. The defensive rating only improves if the team’s defense exceeds expectations.

You can get a feel for what the defense is expected by looking at the opponents’ predicted scores. In the last three games, those predicted numbers were 54 (Princeton), 54 (Denver), 58 (Colgate). The actual points given up in those games and the delta from the prediction were 38 (Princeton – 16 under), 39 (Denver – 15 under), 48 (Colgate – 10 under). Not only were the Nittany Lions holding opponents to less than what was predicted, they were doing it by a large margin.

Now here’s where the subject question needs to be asked  – are these defense performances real or are they just smoke and mirrors? Anyone who watched the Nits games against Princeton and Denver would perhaps say that those opponents were their own undoing. They both flat out stunk the joint up in the first halves when they scored 11 and 10 points respectively. No disrespect meant to PSU defense, but it really didn’t seem to have much to do with the poor performances of those two teams.

Colgate, on the other hand, was a different story. In that game, the Lions did play solid D, particularly Geary Claxton’s effort on Colgate’s leading scorer Kyle Roemer.  Claxton held Roemer to just two points, 16 under his average, on 1-9 shooting. Penn State clearly had the better athletes and they just overwhelmed Colgate.

So I’d say the jury is still out. We don’t yet know if PSU’s defense really has gotten better or not. Even in the case of Colgate, it might just be a case of PSU having the superior athletes, something that they won’t have once they get into Big Ten play. But I do feel strongly that any success that this squad has this year is going to have to come from improve defensive play over last year.  And the numbers do give us something to hope for.