Monday, January 17, 2005

Rebounding Thoughts and Numbers (Part Two)

After calculating all of Kansas's opponents offensive rebounding rates (OffReb%) and comparing that with their results against Kansas, I've found that the Colorado game is the only one in which Kansas has surrendered measurably more offensive rebounds than expected (4.3 more to be exact).

(First, I subtracted the totals from the Kansas game for each school then divided the team's offensive rebounds for the season by the sum of their offensive rebounds and their remaining opponents' defensive rebounds to determine their OffReb%. I then multiplied the OffReb% by the sum of their offensive rebounds and Kansas's defensive rebounds to determine their expected offensive rebounds against Kansas. For example: Texas A&M, in all games (minus their game at Kansas) have 177 offensive rebounds. Their opponents have 251 defensive rebounds. Thus, Texas A&M gathers 41.4% of the rebounds at their offensive end. In the game against Kansas, there were 28 rebounds at Texas A&Ms basket. 41.4% of 28 is 11.6. That would be A&M's expected number of offensive rebounds. In real life, the Aggies grabbed 12 offensive rebounds.)

Now, this is nowhere near a perfect measure primarily due to the fact that I'm not adjusting at all for quality of opposition. With the possible exception of Kentucky and Georgia Tech, Kansas is indisputably the toughest opponent for each of the teams they've played so one would expect Kansas to perform better than average statistically. Six teams were within two offensive rebounds of their expectation against Kansas and three teams (St. Joseph's, Nevada, and UW-Milwaukee) fell short of their expected offensive rebounds by at least the same number that Colorado exceeded expectations (5, 4.3, and 4.9 offensive rebounds respectively).

Barring someone pointing flaws in my methodology or pointing me to a resource that tracks second chance points allowed (maybe Kansas is surrendering about the number of offensive rebounds expected, but giving up more second chance points than the average team) I'll have to conclude that my eyes, guided by my passion, deceive me and that Kansas, though they force a ton of missed shots, don't give up an inordinate number of offensive rebounds.

1 comment:

Anonymous said...

Brett-

I found my way over here after seeing your post at collegeball. This is a nice site!

Defensive rebounding has been my biggest complaint about this season (just ahead of unforced turnovers). I am really surprised, then, to see your conclusion about defensive rebounding. I don't think you've done the numbers wrong, but I am surprised to see the results. The only thing I can think of off the top of my head is that the 'Hawks' opponents' rebounding percentages are artificially inflated by playing teams of a lower caliber than the 'Hawks (like you mention in Part II). It is possible, then, that the 'Hawks are a worse defensive rebounding team than they should be, for a team of their quality. But I have no idea how to go about showing that (unless there is a strong correlation between team quality and Off Reb% allowed and KU is an obvious outlier from that trend).

So, I don't have any idea. Bottom line: I am surprised by your numbers, but don't think they're wrong. Maybe I should worry about turnovers on the other end of the court.

Brian Wilhite
Champaign, IL (I just work here--my loyalties are not divided.)