When Stanford upset Southern California 24-23 last Saturday (10-6-07) many people were surprised. I was not.
No. 2-ranked USC was a perfect 4-0 coming into the game and had enjoyed 5 straight seasons of national top-4 finishes, BCS bowl appearances and Pac 10 Conference titles.
Stanford was a 6-touchdown underdog, had lost 41-3 at home to Arizona State a week earlier, had lost to USC 42-0 a year ago, and finished last year with a lousy 1-11 record.
So what happened? Study these three sets of figures to identify some clues. The first is the prior week's AP Top 25 Poll, the second is Sagarin's mathematical ratings of a team's performance strength against shared opponents, and the third is Sagarin's mathematical ratings of a team's schedule strength.
The AP Poll is made up of 65 media types who follow and report on college football teams. Sagarin is Jeff Sagarin, who produces the gold standard among rating services. Sagarin's ratings represent the average schedule difficulty faced by each team, taking into account the rating of the opponent and the location of the game.
AP Top 25 Poll - Sagarin Rating - Schedule Rank
1) LSU 1) LSU 1) Washington
2) Southern California 2) Southern California 2) Notre Dame
3) California 3) Ohio State 3) Colorado State
4) Ohio State 4) Oklahoma 4) Stanford
5) Wisconsin 5) California 5) Marshall
6) South Florida 6) South Florida 6) Tennessee
7) Boston College 7) West Virginia 7) Mississippi
8) Kentucky 8) Arizona State 8) North Carolina
9) Florida 9) Florida 9) Colorado
10) Oklahoma 10) Auburn 10) AA Sam Houston
11) South Carolina 11) Oregon 11) Auburn
12) Georgia 12) UCLA 12) Akron
13) West Virginia 13) Georgia 13) Duke
14) Oregon 14) Cincinnati 14) East Carolina
15) Virginia Tech 15) Boston College 15) LA-Monroe
16) Hawaii 16) Kentucky 16) Florida International
17) Missouri 17) South Carolina 17) Florida State
18) Arizona State 18) Missouri 18) Syracuse
19) Texas 19) Kansas State 19) UCLA
20) Cincinnati 20) Connecticut 20) Brigham Young
21) Rutgers 21) Florida State 21) Oregon
22) Clemson 22) Boise State 22) San Diego State
23) Purdue 23) Purdue 23) Washington State
24) Kansas State 24) Wisconsin 24) Wake Forest
25) Nebraska 25) Texas 25) Miami (Ohio)
How could Stanford upset USC? Well, for one thing, although USC has far more talent, its talent did not show up for the game. USC's talent was "at" the game but not "in" the game.
USC was 4-0 and ranked No. 2 going into the game and Stanford was 1-3 and ranked No. 83, yet Stanford had played the 4th toughest schedule in the country before the Cardinal arrived at USC as the visiting team. USC had played the 38th toughest schedule.
You decide if this was a factor or not. I say it was a factor. Certainly, USC did not think Stanford was that tough of an opponent; the Trojans paid a big price by not being prepared.
Here is another example. I have said for weeks that Wisconsin was overrated. The Badgers were 5-0 and ranked No. 5 going into Illinois while the Illini were unranked, but upset Wisconsin 31-26.
Wisconsin may have been ranked No. 5 but its Sagarin rating was No. 24 and the Badgers had played the 90th toughest schedule. Sagarin's rating for Illinois was No. 40, but the Illini had played the 39th toughest schedule.
Still not convinced? Kentucky was 5-0 and ranked No. 8 going into South Carolina and lost to the Gamecocks 38-23.
Kentucky had a Sagarin rating of No. 16 and had played the 92nd toughest schedule. South Carolina was 4-1 and ranked No. 11, had a Sagarin rating of No. 17 and had played the 47th toughest schedule.
In a clash of unbeatens, Purdue was 5-0, ranked No. 23 and hosted the Ohio State Buckeyes who beat them 23-7. Sagarin's rating for Purdue was also No. 23 and the Boilermakers had played the 124th toughest schedule. There is only 119 Division 1-A schools.
Ohio State was also 5-0 and ranked No. 4 with a Sagarin rating of No. 3 and had played the 60th toughest schedule.
Some fans and pundits put no credence in Sagarin. I put a lot of credence in the Sagarin's team ratings and schedule rank, and now you know why. I can with some accuracy predict when a team is overrated and due for an unexpected loss.
Copyright © 2007 Ed Bagley