Extra MustardSI On CampusFantasyPhoto GalleriesSwimsuitVideoFanNationSI KidsTNT

The last class

Are some highly ranked teams already doomed?

Posted: Friday December 23, 2005 7:56PM; Updated: Friday December 23, 2005 11:42PM
Free E-mail AlertsE-mail ThisPrint ThisSave ThisMost PopularRSS Aggregators
Taj Gray and OU were lit up by West Virginia on Thursday night in Oklahoma City.
Taj Gray and OU were lit up by West Virginia on Thursday night in Oklahoma City.
AP
MAILBAG
Have questions or feedback? E-mail Luke Winn.
Your name:
Your e-mail address:
Your home town:
Enter your question:

Winter break, for the college kids, is upon us. This I was reminded of by my little brother, who, as I was working on this column, sent the IM, "I'M DONE, B----!" (done with final exams in his penultimate semester at the U of Wisconsin) and then followed it up with thinly veiled references to an afternoon party ... all while I was still in the SI office, facing a 47-block walk home during the New York transit strike. Well, I'm not done yet (b----). Consider this the final class session for every hoophead's first semester: A statistics lecture.

The subject may sound dreadful. But what we're trying to do, just one-third of the way through the season, is not mundane: pinpointing the ranked teams that fit a "champion's" profile -- as well as those that may be setting themselves up for disappointing early exits in March. Knowledge that, if applied properly, could score you big points in your NCAA tournament pool.

The key stat we'll use to examine teams is offensive and defensive "efficiency" -- how many points a team scores or yields per 100 possessions -- because it is a far more telling figure than points scored or allowed per game. It allows us to compare slow teams with speedy teams, and is adjusted to factor in the level of competition (and served up to a ever-growing following at kenpom.com).

Let's start with a couple of questions: In March, why were Wake Forest and Gonzaga -- who were both booted in the second round of the dance -- such popular Final Four picks on tourney pool sheets, despite not having any semblance of defenses (the Deacs were ranked 76th nationally in defensive efficiency, yielding 94 points per 100 possessions, and the Zags were even worse at 119th, giving up 97 points per 100 possessions) to match their top-10 offenses? And why this season are hoopheads -- myself included -- so compelled to watch Bulldogs star Adam Morrison when he admittedly does not care much at all about defense?

The answer -- to both -- is that we're infatuated with offense. Infatuated to the point that we're all too willing to overlook bad defense, and, in the process, confuse a team that's efficient only on offense with a team that has a strong shot of making a run to the Final Four.

But among the teams ranked in the top 15 of the latest AP poll, quality offense is a given. Thirteen of those 15 are in the top 30 in offensive efficiency (Louisville is 45th and Oklahoma is 46th). The rare attribute is stellar defense: Only five of the AP's top 15 are in the top 30 defensively (while the worst, Michigan State, is 247th, giving up 104.7 pts/100 poss.).

Since this efficiency business is relatively new math on a national scale, we only have a small sample size (2003-04, '04-05 and this season), but the data shows that when picking tourney teams, you need to care about D: Of the teams that reached the Elite Eight in 2003-04 and '04-05, only one each season was ranked outside the top 35 in defensive efficiency (Alabama, in '04, was 49th; West Virginia, in '05, was 83rd). Of the teams that reached the past two Final Fours, every one was in the top 25 defensively. And of the past two national champs, each was in the top 10 (UConn was third in '04; UNC was seventh in '05).

Take a look at the efficiency "resumes" of the last eight teams standing in 2004:

2004 Elite Eight Teams
Efficiency Stats are based on points scored and yielded per 100 possessions, and adjusted for competition
Team Round Adjusted
Off. Eff.
(Nat'l Rk.)
Adjusted
Def. Eff.
(Nat'l Rk.)
Connecticut NC 115.7 (8) 83.0 (3)
Georgia Tech FR 111.1 (27) 83.7 (5)
Duke FF 118.7 (2) 83.5 (4)
Oklahoma St. FF 116.3 (5) 85.8 (9)
St. Joseph's EE 115.1 (9) 85.8 (10)
Kansas EE 111.2 (23) 86.5 (14)
Xavier EE 112.3 (17) 87.9 (19)
Alabama EE 112.0 (18) 92.2 (49)

And now, the last eight in 2005:

2005 Elite Eight Teams
Efficiency Stats are based on points scored and yielded per 100 possessions, and adjusted for competition
Team Round Adjusted
Off. Eff.
(Nat'l Rk.)
Adjusted
Def. Eff.
(Nat'l Rk.)
North Carolina NC 122.1 (1) 85.1 (7)
Illinois FR 120.0 (3) 85.5 (11)
Louisville FF 118.0 (6) 87.8 (16)
Michigan St. FF 118.0 (7) 89.3 (25)
Kentucky EE 111.4 (25) 85.0 (6)
Wisconsin EE 110.6 (28) 86.2 (12)
Arizona EE 116.3 (9) 90.3 (34)
West Virginia EE 113.6 (12) 94.6 (83)

Establishing a multileveled "Law of Defense" -- which claims, for example, that you must have a top-10 D to win a national title, or a top-25 D to make the Final Four, and so on -- would probably be going a little overboard. Ken Pomeroy (of kenpom.com) says that although he does not have full 2002-03 efficiency data on his site, Syracuse was not a top-20 defensive team, and Marquette was likely even lower in the D rankings than the Orange. But, based on the past two seasons, we can at least establish an efficiency "profile" of the type of team that is most likely to make a deep tournament run. And with roughly 10 games -- a third of a season -- under each team's belt, we have enough data to start sorting out the field before conference play begins.

Continue

Search