Thursday, March 6, 2014

March 6, 2014

Duke's loss to Wake Forest last night dropped the Blue Devils down one seed line, as did Saint Louis' loss to Dayton.

1's: Arizona, Wichita State, Florida, Villanova
2's: Wisconsin, Kansas, Syracuse, San Diego State
3's: Virginia, Creighton, Cincinnati, New Mexico
4's: Saint Louis, Louisville, Duke, Michigan
5's: Massachusetts, Iowa State, Connecticut, North Carolina
6's: VCU, UCLA, Michigan State, Oklahoma
7's: Gonzaga, Kentucky, George Washington, Ohio State
8's: Texas, Memphis, Oregon, SMU
9's: Saint Joseph's, Colorado, Arizona State, Pittsburgh
10's: Dayton, Iowa, BYU, Missouri
11's: Kansas State, Arkansas, Oklahoma State, Southern Miss
12's: Xavier, Baylor, Providence, Saint Mary's, Green Bay, North Dakota State
13's: Toledo, Iona, Stephen F. Austin, Harvard
14's: Belmont, Georgia State, North Carolina Central, Delaware
15's: Boston University, Mercer, New Mexico State, Vermont
16's: UC Santa Barbara, Robert Morris, Davidson, Weber State, Radford, Alabama State

FFO: Stanford, Tennessee, Middle Tennessee, Nebraska
NFO: Clemson, Saint John's, Florida State, Minnesota


  1. Out right big ten champ, top 10 in amost every rating site,10 top 50 rpi wins and a 4 seed for Michigan?

  2. Please explain your reasoning for Saint Louis being higher on the S-curve than Michigan, and for that matter, Creighton, Cincy, and Duke.

    Look at their resumes ... Michigan's is better.

    1. Here's an RPI/record breakdown courtesy of

  3. Michigan is closer to a 1 seed then they are a 4 seed. But dont get me wrong, they're not a 1 seed. They are on the 2/3 line

  4. i came here to ask about Michigan but it looks like it was already taken care of

  5. To unknown - a conference championship is not an independent metric on it's own. It is a function of wins and losses that are already factored into the seeding. They shouldn't be doubly counted. My predictions are not based on any other rating site, or poll, or what anyone else thinks about a team. They're based on a multiple regression analysis from the last three years' tournament selections.

    There are 31 teams with a better record than Michigan. By virtue of their strong SOS and quality wins, they move up past about half of them, but I don't believe they're a top 3 seed. Based on conference numbers only, they would certainly be a 2- or 3-seed, but you can't ignore their mediocre Non Conf numbers (#77 RPI, #92 SOS).

    1. This comment has been removed by the author.

  6. Regarding comparison to the teams I have ahead of Michigan:
    Saint Louis - their recent slump isn't factored into my formula, although it maybe should be. Outside of their one anomolous loss to Duquesne, the Billikens don't have a loss outside the top 50, and only two total losses outside the top 14. You can certainly put Michigan ahead of Saint Louis, but that doesn't move them off the 4-line on my bracket...

    Duke and Michigan are also very close, and I wouldn't fault you for putting Michigan ahead. Duke's overall record (especially Adj WP) is a bit better, they have one more top-25 win, and their performance has been more consistent both inside and outside their conference.

    Creighton and Cincy are much easier arguments. Creighton has a better RPI, better Non Conf numbers, same top-100 record, no bad losses, and swept a #1-seed. Cincy has two fewer losses and no losses outside the top-45.

  7. Final thought on this (for now), Here is a list of Michigan's top-50 wins. #5, #22, #22, #27 (impressive so far, huh?), #40, #46, #48, #48, #49, #49. So if you said top-45 instead, they'd have just 5. "50" is a more or less arbitrary number used to identify "quality" wins, but it's a far from scientific measuring tool.