Ranking the NCAA Tournament Teams 1-68

Last year was not a good year for me in terms of predicting the bracket, despite correctly picking Louisville as the winner.  As a result, I made sure that this year my bracket would be better.  Over the past few months, I have been fine-tuning a logarithm that models college basketball with predictive accuracy.  Turns out this was the right year to create a formula, as Warren Buffet is giving $1 Billion to anyone who can make a perfect bracket through the Quicken Loans Bracket Challenge.

The formula is similar to that of Nate Silver; It incorporates various well-known college basketball rankings system into an average.  My model differs, however, by taking a Weighted Average, based on performance instead of just weighting each system equally, as Silver does with the computer model portion of his rankings.

I won't go into the exact weights each system gets (Any good formula needs a bit of secrecy), but the rankings used are as follows.

Ken Pomeroy's KenPom
Joel Sokol's LRMC Bayesian Model
Joel Sokol's LRMC Classic Model
Raymond Cheong's AASM+HCA
Sonny Moore's "PR"
Associated Press's Preseason Rank

I should note that these rankings have been proved to be the most accurate, more accurate than some of the ones that Silver uses in his model like seeding.  In addition, my model rewards the most accurate models by giving them more weight, which should make the model the most accurate one available.

Now, let's get to the good stuff.  How can I help you fill out your bracket?

While I am not yet finished calculating each teams odds at reaching the championship (Because the model is in its first year, these things take longer), the teams ranking does correlate very well with their odds of advancing far in the tournament, even if some teams have a harder path than others.

Be aware, however, that this does NOT mean that these are the teams most likely to win the tournament.  It should provide a good idea, but some teams (Wichita State comes to mind) just have a much harder path than others.

Rank. Team (Seed)

1. Louisville (4)
2. Arizona (1)
3. Florida (1)
4. Duke (3)
5. Kansas (2)
6. Michigan State (4)
7. Virginia (1)
8. Creighton (3)
9. Villanova (2)
10. Wichita State (1)
11. Wisconsin (2)
12. Michigan (2)
13. UCLA (4)
14. Oklahoma State (9)
15. Ohio State (6)
16. Kentucky (8)
17. Iowa (11)
18. Gonzaga (8)
19. Syracuse (3)
20. Tennessee (11)
21. VCU (5)
22. Pittsburgh (9)
23. Connecticut (7)
24. Iowa State (3)
25. North Carolina (6)
26. Oregon (7)
27. Cincinnati (5)
28. New Mexico (7)
29. San Diego State (4)
30. Oklahoma (5)
31. Baylor (6)
32. Memphis (8)
33. Stanford (10)
34. Providence (11)
35. BYU (10)
36. Saint Louis (5)
37. Xavier (12)
38. Texas (7)
39. Arizona State (10)
40. Harvard (12)
41. Kansas State (9)
42. Massachusetts (6)
43. Colorado (8)
44. George Washington (9)
45. Nebraska (11)
46. Dayton (11)
47. Saint Josephs (10)
48. NC State (12)
49. New Mexico St (13)
50. North Dakota St (12)
51. Tulsa (13)
52. Manhattan (13)
53. SF Austin (12)
54. Delaware (13)
55. NC Central (14)
56. Mercer (14)
57. American (15)
58. Louisiana Lafayette (14)
59. Western Michigan (14)
60. Eastern Kentucky (15)
61. Milwaukee (15)
62. Albany (16)
63. Cal Poly (16)
64. Weber State (16)
65. Mount St. Mary's (16)
66. Wofford (15)
67. Coastal Carolina (16)
68. Texas Southern (16)

EDIT:  I now realize that I posted this after Albany had already beaten Mount St. Mary's.  I didn't cheat; the data was from before the game was even played.  On the bright side, 1 for 1!