Email me

March 31, 2006

Conference Performance In The NCAA Tournament

With the mass exodus of talent to the NBA last year, most people expected that the ACC would be down this year and that's pretty much how things turned out. Carolina was better than expected, but that was offset by Duke being not as good as expected. Most people, me included, thought the Blue Devils would be far and away the best team in the country, but they struggled all season with developing their bench and getting consistent production from anyone other than J.J. Redick and Shelden Williams. Boston College was also supposed to be a top-ten team, but rarely played like it. They coasted at times and were more like a 15-20-type team.

The NCAA Tournament isn't necessarily the best way to rate a conference's yearly performance - a true measure would take tournament performance combined with a few computer power ratings and head-to-head comparisons against other top leagues - but it's probably the most significant. Five years from now, few people will remember much about this season other than that Duke failed to reach a Final Four and the ACC was shut out of the Elite Eight. Likewise, many will forget that the SEC had a relatively down year, because that league has stepped up so nicely in the postseason.

I put together a table that shows how each of the six power conferences (sorry, Conference USA, you no longer count) have performed so far in the postseason. In each column, I've bolded the top figure so that you see for example that the Big East led in bids while the SEC has the best winning percentage.

 

NCAA NIT Total Postseason

Conference

Bids W-L Win % Bids W-L Win % Bids W-L Win %

ACC

4 6-4 .600 6 4-6 .400 10 10-10 .500

Big East

8 11-8 .579 4 7-4 .636 12 18-12 .600

Big Ten

6 3-6 .333 3 5-3 .625 9 8-9 .471

Big Twelve

4 4-4 .500 3 0-3 .000 7 4-7 .364

Pac 10

4 7-3 .700 1 1-1 .500 5 8-4 .666

SEC

6 11-4 .733 2 5-1 .833 8 16-5 .761

As you see, the ACC falls in the middle of the pack in most measures. It didn't flop nearly as badly as the Big Ten or Big Twelve, but didn't perform as well as the Big East, Pac Ten or SEC. What's particularly interesting is that Big Ten was believed by many to be the top league this year while the Pac Ten and SEC were almost universally thought to be having bad years.

For my next table, I've added another column for what I call Impact. Impact is simply the number of bids multiplied by the conference winning percentage. This takes into account both how well a league did (win %) and how many teams they sent to the dance. This way, teams that get bids but lose in the first round (half of the 64-team field) can still have a positive impact on a league's rating. In effect, this gives a league credit for it's pre-tournament success and also keeps leagues that send only one or two teams from appearing to be the top conference if one of their teams makes a deep run (like the CAA this year).

Curiously, the Impact rating says that the Big East has had more impact on the tournament than the SEC (to this point). This may seem counter-intuitive, since the SEC has two Final Four teams while the Big East has none, but remember that the Big East had four Sweet Sixteen teams while the SEC was already down to two at that point.

 

NCAA

Conference

Bids W-L Win % Impact

ACC

4 6-4 .600 2.40

Big East

8 11-8 .579 4.89

Big Ten

6 3-6 .333 2.00

Big Twelve

4 4-4 .500 2.00

Pac 10

4 7-3 .700 2.80

SEC

6 11-4 .733 4.40

Comments
 
(1) by Chad Orzel (unregistered) on 03/31/2006 05:37 pm
I don't think it means much to talk about the raw number of wins for a conference, without considering the seeding. For example, a lot of people tried to make something out of the Big East going 0-3 on the first day of the tournament, when two of those three were in the 7-10 game, and the third was Syracuse, who needed a miracle run at the Big East tournament to guarantee themselves a spot.

You shouldn't get the same credit for a #1 seed winning three games as for a #11 seed winning three.

The best technique I've seen is to determine the "expected" number of wins for each seed (0 for teams seeded 9 or lower, 1 for teams seeded 5-8, two for teams seeded 3-4, etc.), and look at how teams under- or over-performed. Of course, that's really tedious to work through...

 
(2) by Dave on 03/31/2006 06:02 pm
Why penalize a team for being good? If you want to measure how a team or conference did versus expectations, then taking seeds into account makes sense. But if a conference has several high seeds, then they earned them. They still need to win games. There aren't many chippies in the tournament any more.

 
(3) by Chad Orzel (unregistered) on 04/01/2006 09:35 pm
Obviously, this comes down to personal taste in the end, but I don't think it's as impressive for a #3 seed to win two games as it is for a #14 to win one. You can only play the teams that they put in front of you, sure, but I think it's more of a credit to a team and a conference for a lower seed to win, than for a high seed to simply take care of business.

I also think that comparing to expected performance helps account for some of the differences in the competition that different conferences face, which is important if you want to make a meaningful comparison. Otherwise, it's too much like college football for my tastes...

 
(4) by Chad Orzel (unregistered) on 04/01/2006 09:35 pm
Obviously, this comes down to personal taste in the end, but I don't think it's as impressive for a #3 seed to win two games as it is for a #14 to win one. You can only play the teams that they put in front of you, sure, but I think it's more of a credit to a team and a conference for a lower seed to win, than for a high seed to simply take care of business.

I also think that comparing to expected performance helps account for some of the differences in the competition that different conferences face, which is important if you want to make a meaningful comparison. Otherwise, it's too much like college football for my tastes...

Comment:
Username or your name/handle: 

   
Posted by Dave at March 31, 2006 03:43 PM | TrackBack

BrightCar Car Maintenance Software  - Software To Manage Your Car Maintenance Schedule
Basketball Scores
Links
Columnists
Caulton Tudor
Ned Barnett
Lenox Rawlings
Tony Kornheiser
Michael Wilbon
Bill Simmons
Blogs
ACC BasketBlog
The Sports Frog
Duke Basketball Report
FanBlogs (college football)
Yoco: College Basketball
StateFans Nation
Carolina Hoops Experience
Carolina Basketball Update
Ken Pomeroy
The SportsProf
Chris's Sports Blog
ACC Info
Charlie Board's ACC Stats Page
Official ACC Site
Ratings
Mike Greenfield's Ratings
Pomeroy Ratings
Sagarin Ratings