In the giddy post-game atmosphere after the BCS Championship Game, but from the depths of his post-drubbing-of-Ohio State-induced depression, ESPN college football analyst Kirk Herbstreit, a former quarterback at Ohio State, made a woeful assessment of the Big Ten’s image in the football world. He said that the Big Ten would be placed "between Conference USA and the MAC" because of Ohio State’s second straight failure in the BCS Championship Game (and the conference’s overall 3-5 record in bowl games).
I decided to test this assessment with some simple, readily available statistics--the final Coaches Poll rankings and the records of each team. In my view, number of losses is most directly related to poll ranking. Strength of schedule is the second most important factor. Since so much of a team’s schedule is composed of conference games, conference affiliation is often substituted for a more complete analysis of a team’s actual strength of schedule.
Herbstreit’s assessment would be born out if teams with similar numbers of losses in the Big Ten, Conference USA and the MAC found themselves in similar poll positions.
I expanded the analysis to include all ranked teams and all teams with five or fewer losses after the bowl season. There were 39 teams that received votes in the final Coaches’ Poll. Where the vote totals were close, I grouped up to three teams (example: I called Georgia and USC tied for 2nd even though USC had ten more votes (these two teams were reversed in the final AP poll)). There were ten unranked teams with five or fewer losses. I called this group "tied for 45th".
For each conference, I totaled both the rankings and losses for all the included teams. To determine what I call the "conference respect factor" (CRF), I merely divided the total number of ranking points by the total number of losses for each conference.
On a micro scale this system works beautifully. LSU had a rank of 1 and 2 losses. Their team respect factor (TRF) was 1 divided by 2 or 0.5—the lowest of any team. They were awarded the national championship. Hawaii had the highest TRF—17; ranked 17th despite only one loss. Otherwise, Troy and New Mexico had the highest factors—they ranked 45th (estimated) with four losses—a team respect factor of 45 divided by 4 or 10.5.
The conferences’ respect factors landed somewhere in between. Here is the ranked list of CRFs.
Big 12 (5 teams) --- 4.2 --- Top-heavy Big 12 had no 5-loss teams
SEC (8 teams) --- 4.8 --- Includes three 5-loss teams
Pac 10 (5 teams) --- 5.2 --- Includes Cal, only 6-loss team in the final poll
ACC (5 teams) --- 5.3 --- All 3-loss or 4-loss teams
Big Ten (6 teams) --- 5.7 --- Still a ways from C-USA or MAC
Big East (4 teams) --- 7.2 --- Big East powers recently moved to ACC
MAC (1 team) --- 9.0 --- Bowling Green worst 5-loss team in country
Indies (2 teams) --- 9.0 --- Both teams unranked.
C-USA (4 teams) --- 9.1 --- Includes two unranked teams
WAC (3 teams) --- 9.7 --- Hurt by Hawaii’s TRF of 17
Sun Belt (2 teams) --- 10.0 --- Both teams unranked.
Mountain West (4) --- 10.1 --- Includes two unranked teams.
(My apologies for the format of the above. I still haven't figured out how to align tables in this system)
These figures show a clear grouping of major conferences (Big 12 thru Big Ten anyway) and minor conferences (MAC thru Mountain West). A loss hurts a major conference school about half as much as one does a minor conference school. The Big East sits in the middle, still digesting its Conference USA (Louisville, Cincinnati) and Div I-AA (Connecticut, South Florida) pickups, but held somewhat aloft by holdover West Virginia, who gets treated like an ACC school (TRF of 3). With a better record, Notre Dame would skew the figure for Independents as the Irish are treated like a major conference school, while the other independents (Western Kentucky, Navy and Army) are treated like minor conference schools.
Getting back to Herbstreit’s misery moan, the Big Ten has a long way to go before finding itself in MAC-land. To reach a CRF of 9.0, the average rank of the six Big Ten teams with five or fewer losses would have had to be 34.5 rather than 21, dropping the number of Big Ten teams in the top 25 from five to one. Or put another way, six Big Ten teams would average 2.5 losses and end up in the same poll rankings as seen this year. Can you imagine a final poll with three 2-loss teams from the Big Ten and only one in the top ten? As for Ohio State, they had a final TRF of 2.0, fourth only to LSU (0.5), Georgia (1.25) and USC (1.25) and ahead of such New Year’s Day-or-later bowl winners as Kansas (6.0), Texas Tech (3.5), West Virginia (3.0), Missouri (3.0) and Tennessee (3.0).
A shorter, but more painful fall may come in December 2008 if voters decide to put two 2-loss teams into the 2009 BCS Championship Game ahead of a 1-loss Ohio State or Michigan team and create the potentially competitive match that was left unplayed in 2008. If that happens, loyal Big Ten man Herbstreit will really have something to be depressed about and Big Ten fans will join the clamor for a playoff system.
I decided to test this assessment with some simple, readily available statistics--the final Coaches Poll rankings and the records of each team. In my view, number of losses is most directly related to poll ranking. Strength of schedule is the second most important factor. Since so much of a team’s schedule is composed of conference games, conference affiliation is often substituted for a more complete analysis of a team’s actual strength of schedule.
Herbstreit’s assessment would be born out if teams with similar numbers of losses in the Big Ten, Conference USA and the MAC found themselves in similar poll positions.
I expanded the analysis to include all ranked teams and all teams with five or fewer losses after the bowl season. There were 39 teams that received votes in the final Coaches’ Poll. Where the vote totals were close, I grouped up to three teams (example: I called Georgia and USC tied for 2nd even though USC had ten more votes (these two teams were reversed in the final AP poll)). There were ten unranked teams with five or fewer losses. I called this group "tied for 45th".
For each conference, I totaled both the rankings and losses for all the included teams. To determine what I call the "conference respect factor" (CRF), I merely divided the total number of ranking points by the total number of losses for each conference.
On a micro scale this system works beautifully. LSU had a rank of 1 and 2 losses. Their team respect factor (TRF) was 1 divided by 2 or 0.5—the lowest of any team. They were awarded the national championship. Hawaii had the highest TRF—17; ranked 17th despite only one loss. Otherwise, Troy and New Mexico had the highest factors—they ranked 45th (estimated) with four losses—a team respect factor of 45 divided by 4 or 10.5.
The conferences’ respect factors landed somewhere in between. Here is the ranked list of CRFs.
Big 12 (5 teams) --- 4.2 --- Top-heavy Big 12 had no 5-loss teams
SEC (8 teams) --- 4.8 --- Includes three 5-loss teams
Pac 10 (5 teams) --- 5.2 --- Includes Cal, only 6-loss team in the final poll
ACC (5 teams) --- 5.3 --- All 3-loss or 4-loss teams
Big Ten (6 teams) --- 5.7 --- Still a ways from C-USA or MAC
Big East (4 teams) --- 7.2 --- Big East powers recently moved to ACC
MAC (1 team) --- 9.0 --- Bowling Green worst 5-loss team in country
Indies (2 teams) --- 9.0 --- Both teams unranked.
C-USA (4 teams) --- 9.1 --- Includes two unranked teams
WAC (3 teams) --- 9.7 --- Hurt by Hawaii’s TRF of 17
Sun Belt (2 teams) --- 10.0 --- Both teams unranked.
Mountain West (4) --- 10.1 --- Includes two unranked teams.
(My apologies for the format of the above. I still haven't figured out how to align tables in this system)
These figures show a clear grouping of major conferences (Big 12 thru Big Ten anyway) and minor conferences (MAC thru Mountain West). A loss hurts a major conference school about half as much as one does a minor conference school. The Big East sits in the middle, still digesting its Conference USA (Louisville, Cincinnati) and Div I-AA (Connecticut, South Florida) pickups, but held somewhat aloft by holdover West Virginia, who gets treated like an ACC school (TRF of 3). With a better record, Notre Dame would skew the figure for Independents as the Irish are treated like a major conference school, while the other independents (Western Kentucky, Navy and Army) are treated like minor conference schools.
Getting back to Herbstreit’s misery moan, the Big Ten has a long way to go before finding itself in MAC-land. To reach a CRF of 9.0, the average rank of the six Big Ten teams with five or fewer losses would have had to be 34.5 rather than 21, dropping the number of Big Ten teams in the top 25 from five to one. Or put another way, six Big Ten teams would average 2.5 losses and end up in the same poll rankings as seen this year. Can you imagine a final poll with three 2-loss teams from the Big Ten and only one in the top ten? As for Ohio State, they had a final TRF of 2.0, fourth only to LSU (0.5), Georgia (1.25) and USC (1.25) and ahead of such New Year’s Day-or-later bowl winners as Kansas (6.0), Texas Tech (3.5), West Virginia (3.0), Missouri (3.0) and Tennessee (3.0).
A shorter, but more painful fall may come in December 2008 if voters decide to put two 2-loss teams into the 2009 BCS Championship Game ahead of a 1-loss Ohio State or Michigan team and create the potentially competitive match that was left unplayed in 2008. If that happens, loyal Big Ten man Herbstreit will really have something to be depressed about and Big Ten fans will join the clamor for a playoff system.
No comments:
Post a Comment