Next's Year's Brackets....

[RPS]

Did you predict an NCAA Final of Kansas v. Memphis? Most people's NCAA brackets are a mess. But there is hope, at least for next year. Three optimization and statistics professors in the Stewart School of Industrial and Systems Engineering at Georgia Tech have created a computer ranking system that consistently predicts NCAA basketball rankings more accurately than the various polls, formulas and other computer models available. It's even more accurate than the tournament seeds themselves.

"After correctly picking all four of this year’s finalists, the LRMC method has now identified 30 of the last 36 Final Four participants (83 percent accuracy over the past nine years of NCAA tournaments) as one of the top two teams in their region. Over the same nine-year stretch, the seedings and polls have correctly identified only 23, and the RPI indentified 21."

Link (http://www.gatech.edu/newsroom/release.html?id=1802)

[Franzez]

2009 Final Four Prediction:
1-UCLA vs 1-Duke
2-Michigan State vs 1-Tennessee

[Bluedog]

"After correctly picking all four of this year’s finalists, the LRMC method has now identified 30 of the last 36 Final Four participants (83 percent accuracy over the past nine years of NCAA tournaments) as one of the top two teams in their region. Over the same nine-year stretch, the seedings and polls have correctly identified only 23, and the RPI indentified 21."

Link (http://www.gatech.edu/newsroom/release.html?id=1802)


Uh, looking at the LMRC rankings, http://www2.isye.gatech.edu/people/faculty/Joel_Sokol/lrmc/lrmc.sort0.html, it has Duke ranked fourth and Carolina fifth. So didn't they only get 3/4? Am I missing something?

1. Kansas
2. Memphis
3. UCLA
4. Duke
5. North Carolina
6. TN
7. Wisconsin
8. Clemson
9. Stanford
10. Texas

[shadowfax336]

no because they ranked UCLA ahead of Duke

[Bluedog]

no because they ranked UCLA ahead of Duke

Ah, true. duh. my stupidity.....

[brevity]

Interesting timing with this post. I do a year-in-advance Final Four prediction at the end of the tournament, which produces surprisingly okay results (Kansas and Memphis this year, UCLA and Ohio State last year, LSU in 2006, Illinois and Louisville in 2005). Making picks that early -- when you don't even know about transfers, injuries, or NBA defections -- can be tricky.

One unintended advantage of this approach is that it gives me teams to follow during the regular season besides Duke. Not root for, really, just watch closely.

All that said, I'm having a hard time picking a 2009 Final Four because I'm not feeling strongly about the chances of any team. Right now I'm leaning towards Notre Dame, Michigan State, Texas, and UCLA. But I'd appreciate some thoughts about any other teams I should consider. Not Duke, though; there are other threads on this board that evaluate their chances next season.

[CDu]

Uh, looking at the LMRC rankings, http://www2.isye.gatech.edu/people/faculty/Joel_Sokol/lrmc/lrmc.sort0.html, it has Duke ranked fourth and Carolina fifth. So didn't they only get 3/4? Am I missing something?

1. Kansas
2. Memphis
3. UCLA
4. Duke
5. North Carolina
6. TN
7. Wisconsin
8. Clemson
9. Stanford
10. Texas

Based on this, it looks like the system successfully predicted the Final Four, but missed badly on the Elite Eight. Duke, Tennessee, Wisconsin, Stanford, and Clemson look like pretty bad calls now. Of course, given that every game has a large margin for error in the final score, it's not surprising that the system can miss on certain players.