Random Walker Rankings for NCAA Football




Return to RWR Blog

Beginning & Disclaimer

Can monkeys rank teams?

2003 rankings

What was wrong with the old BCS formula?

2004 rankings

2005 rankings

2006 rankings

2007 rankings

2008 rankings

Our manuscripts

Press coverage

2006 Random Walker Rankings

With the BCS as its one-piece sequin jump suit, college football
is essentially figure skating combined with Microsoft Excel, a
subjective taste test on what emotional voters and soulless
computers perceive as the two teams most worthy to play in the
national championship game. Sunday night we will find out what
matchup the humans and the computers give us.
                                    -- John Buccigross, ESPN SportsCenter

Yes, ladies and gentlemen, it's that time of year again. The leaves have fallen, the smell of fireplace-burning wood hangs in the crisp winter air, and the arguments about who should play in the Division I-A football title game ring in our ears... Yeah, I liked Buccigross's quote better too...

So, what happens next? We'll find out when the bowl matchups are announced. The random walkers definitely prefer Florida to Michigan. At our preferred p=0.75 value, this simplistic measure even says to go ahead and lineup Florida against USC, because the system doesn't know anything about margins of victory (neither do any of the official BCS computer rankings, by decree), nor about dates of games. Some of the official systems take date of game into account, some do not; the ones that do not consider dates treat Michigan's loss to Ohio State and USC's loss to UCLA as if they had happened at any time, not in the last games of their respective regular seasons. At p just about 0.75, Ohio State moves up into the top two. Please don't complain about how an undefeated team isn't always ranked in the top two, else we have to talk about Boise State... But of course in the revised BCS system, the voters have 2/3 of the power and will possibly make the decision on their own without the official computer rankings (which we again emphasize we are not part of) affecting the outcome. Who will play Ohio State in the Championship game? Will Notre Dame make a BCS bowl? Who will get tasked with playing Boise State?

We remain grateful to both Kenneth Massey for maintaining his comparisons page and to Peter R. Wolfe for making game results available online.

Below we list the Top 25 according to the Random Walker (RW) system of first-place votes and the Random Walker First-Last (RWFL) system at our selected p=0.75. We also make available rankings of all 119 Div-IA teams at p=0.75. The breakdown of the top teams across different p values for both RW and RWFL can be seen in the figures below. Additionally, we plot the fraction of ranking violations---that is, the fraction of game outcomes that have been contrary to the rankings---for each system across the p values. Ranking violations like these are also included on Kenneth Massey's site, and can be used as one measure of the efficacy of a rank ordering; indeed, one could instead choose p to minimize such errors, but we are more comfortable at the selected fixed values for ease of explanation.

2006 Random Walker Top 25 (p=0.75)
Games Through Saturday December 2nd
RW RWFL
RankTeam (Wins-Losses)% of Total Votes Team (Wins-Losses)%(First)-%(Last)
1Florida (12-1)2.6043Florida (12-1)2.3717
2Southern Cal (10-2)2.4950Southern Cal (10-2)2.2838
3Ohio State (12-0)2.4656Ohio State (12-0)2.1911
4Michigan (11-1)2.1151Michigan (11-1)1.8539
5Auburn (10-2)2.0481Auburn (10-2)1.7368
6Louisville (11-1)1.9846Louisville (11-1)1.7162
7LSU (10-2)1.9717LSU (10-2)1.7061
8Boise St (12-0)1.8395Boise St (12-0)1.5138
9California (9-3)1.8001California (9-3)1.5067
10Arkansas (10-3)1.6829Notre Dame (10-2)1.3882
11Notre Dame (10-2)1.6753Arkansas (10-3)1.3470
12Tennessee (9-3)1.6420Tennessee (9-3)1.3370
13Oregon St (9-4)1.6040Oregon St (9-4)1.2614
14West Virginia (10-2)1.5519West Virginia (10-2)1.2311
15Rutgers (10-2)1.5515Rutgers (10-2)1.2075
16Oklahoma (11-2)1.5161Oklahoma (11-2)1.1986
17Wisconsin (11-1)1.4490UCLA (7-5)1.0586
18Virginia Tech (10-2)1.4474Wisconsin (11-1)1.0548
19Wake Forest (11-2)1.4353Wake Forest (11-2)1.0519
20UCLA (7-5)1.4300Virginia Tech (10-2)1.0475
21Boston College (9-3)1.3040Texas (9-3)0.8871
22Texas (9-3)1.2868Oregon (7-5)0.8227
23Georgia (8-4)1.2473Georgia (8-4)0.7886
24Oregon (7-5)1.2349Boston College (9-3)0.7667
25Arizona (6-6)1.1584Arizona (6-6)0.7071




2006-2007 Post-Bowl Random Walker Rankings

While our primary interest here at Random Walker Rankings is in what the rankings say prior to the bowl games, we were particularly interested this year to see how the teams were rated after all the bowl games were played. There isn't any drama about who is ranked #1 and #2 by these methods, since both Florida and USC won their bowl games against the RW/RWFL #3 and #4 teams, Ohio State and Michigan, respectively. But our love of the purported underdog made us particularly want to see whether and how far Boise State would rise from their #8 pre-bowl ranking. Alas, as tabulated below and in the rankings of all 119 Div-IA teams, the Broncos only picked up one spot by passing Michigan, which isn't surprising since previously-unbeaten Ohio State was the only other team in the top 8 besides Michigan who lost their bowl game.

2006-2007 Random Walker Top 12 (p=0.75)
Games Through Monday January 8th
RW RWFL
RankTeam (Wins-Losses)% of Total Votes Team (Wins-Losses)%(First)-%(Last)
1Florida (13-1)2.9482Florida (13-1)2.7324
2Southern Cal (11-2)2.4905Southern Cal (11-2)2.2605
3Louisville (12-1)2.2616Louisville (12-1)2.0142
4Auburn (11-2)2.2099Auburn (11-2)1.9153
5LSU (11-2)2.1275LSU (11-2)1.8764
6Ohio State (12-1)2.0720Ohio State (12-1)1.7870
7Boise St (13-0)2.0517Boise St (13-0)1.7405
8Michigan (11-2)1.8307Michigan (11-2)1.5563
9California (10-3)1.7369West Virginia (11-2)1.4348
10West Virginia (11-2)1.7295California (10-3)1.4272
11Rutgers (11-2)1.6701Rutgers (11-2)1.3482
12Wisconsin (12-1)1.6135Wisconsin (12-1)1.2423

GT UNC Copyright © 2006-2007 Peter J. Mucha (mucha@unc.edu), Thomas Callaghan, Mason A. Porter

THIS PAGE IS NEITHER A PUBLICATION OF THE UNIVERSITY OF NORTH CAROLINA (UNC) NOR THE GEORGIA INSTITUTE OF TECHNOLOGY (GT), WHERE THIS WORK BEGAN. NEITHER UNC NOR GT ARE RESPONSIBLE FOR EDITING OR EXAMINING ITS CONTENT. THE AUTHOR OF THIS PAGE IS SOLELY RESPONSIBLE FOR THE CONTENT. THE RIGHTS TO ANY AND ALL MATERIALS CREATED BY THE AUTHOR OF THIS PAGE ARE RETAINED BY THAT AUTHOR.