Daily Dribble: Stats Give Away Final Four?

February 23rd, 2007

I love sports stats. Its as simple as that.. baseball has the best stats obviously, but college basketball has some underrated numbers. And I'm not talking about the RPI, I'm talking about Ken Pomeroy's offensive and defensive efficiency numbers. Its not my place to explain these numbers, but you can check out http://kenpom.com/blog/index.php/weblog/stats_explained/ for what the stats mean. But basically a good offensive efficiency rating means a team is good at offense and a good defensive efficiency rating means your good at D. Ken combines these two efficiency ratings into the "Pythag Winning Percentage" which basically predicts how a team will do against an average schedule. For example, last year's national champion Florida, ended up #1 in the Pythag category. They were projected to have a 97.4 winning percentage against an average schedule. The worst, not suprisingly was 1-26 Savannah State with .8% Pythag winning percentage. This means (theoretically), Savannah State would only have won 1 of every 100 games against an average NCAA schedule last year.

Anyways, I don't mean to repeat the information you can find at KenPom.com, but I'm here to point out how Ken's Pythag stat has done such a good job predicting the Final Four of the past recent years.

2004 - W: UCONN (2) - F: Georgia Tech (7) - SF: Duke (1), Oklahoma State (3)

In 2004 (when Ken's Pythag stat starts online), the four Final Four teams were 1st, 2nd, 3rd, and 7th in Pythag Winning %. 4th, 5th, and 6th that year were Pittsburgh, Wisconsin, and St. Joseph's. St. Joseph's went to the Elite Eight, Pitt the Sweet Sixteen, and Wisconsin the 2nd round. The fact that Wisconsin and Pitt didn't go far might make it seem like Pythag % doesn't mean much.. But those 3 teams along with Oklahoma State were all in the same regional bracket. Wisconsin lost to Pitt in the 2nd round, Pitt lost to Oklahoma State, and then Oklahoma State beat St Joseph's. Basically, Wisconsin was the most underrated seed according (a 6th seed despite being 5th overall in Pytag), and perhaps they would have done better if they didn't run into the #4 Pythag team in the 2nd round.

2005 - W: UNC (1) - F: Illinois (2) - SF: Louisville (4), Michigan State (7)

UNC and Illinois were #1 seeds and their finals appearance surprised no one (including the Pythag which had them 1-2). Louisville and MSU however were 4 and 5 seeds, so their appearance in Saint Louis was more surprising. However, according to the Pythag, they were both underseeded and were both Top 8 teams. The other high pythag teams that year were Duke (3), Villanova (5), and Florida (6). Duke and Villanova both made it to the Sweet 16, while Florida lost in the 2nd round. Again though, Florid'a early exit doesn't disprove the Pytah.. because they lost to Villanova. Villanova, despite being the 5th seed, gave UNC a great run in the Sweet Sixteen, which seemed surprising at the time but maybe it wasn't since they were 5th in the Pythag. Duke's upset to Michigan State in the Sweet Sixteen (a 1 to a 5 seed) doesn't seem like much of one considering Duke was 3rd in Pytah to MSU's 7th.

2006 - W: Florida (1) - F: UCLA (3) - SF: LSU (10), George Mason (23)

Last year, not one #1 seed made the Final Four.. and of course George Mason made its magical run. LSU and George Mason were the lowest teams in recent years to make the Final Four (4th and 11th seeds), but they ended up with solid Pythag's of 10th and 23rd.

If you remember, 2nd seed Tennessee got knocked out in the 2nd round by Wichita State.. and 2nd seed Ohio State lost to Georgetown. According to the Pythag, those teams were highly overrated to begin with. OSU was 18th in Pytag and Tennesee was 22nd. Basically OSU should have been a 5th seed.. and Tennessee a 6th seed.

So what does 2007 have in store for us? The point of all the above was that the Pythag is a great predictor of NCAA Tournament success. Theoretically teams that have a Pythag right now will go further. Who does that benefit?

As of now, the Pythag Ranking is:
1) UNC
2) Kansas
4) Florida
5) OSU
7) Georgetown
8) Wisconsin
9) Memphis
10) Duke

These are all Top 20 teams.. so there's nothing too surprising there. But there are some potential top seeds that the Pythag doesn't respect much.

9th ranked Washington State is just 26th in Pythag. These means they should only be a 7th seed as of now.. and a potential 1st round upset victim.

11th ranked Nevada is just 64th in the Pythag. This means the Wolfpack may be a definite 1st round upset victim.

Some unranked teams that might go far?

Maryland and Michigan State are unranked, but 11th and 12th in Pythag. If these teams get seeded in the 7-10 range, a potential #2 seed like Pittsburgh (13th in Pythag) should be very worried..