SWC75
Bored Historian
- Joined
- Aug 26, 2011
- Messages
- 33,988
- Like
- 65,563
There are constant arguments about the best size for an NCAA Division 1A playoff. We’ve had two teams for years and that has been unsatisfactory. We are moving to 4 teams in 2014. Will that be enough? I decided to try to look at it logically.
What is a national championship contender? The generally recognized sources for “national championships” have been the writer’s poll which began in 1936, the coach’s poll, which began in 1950 and the BCSD, which began in 1998. These organizations have chosen 87 national champions in the 77 years from 1936-2012, (more than the number of years due to split titles). 83 of these 87 teams have either won all their games or all their games but one, (meaning they lost one game or tied one game). The other four:
1965- Michigan State, Arkansas and Nebraska were all 10-0-0 and ranked #1-2-3. All lost in New Year’s day bowl games. The highest ranked conqueror was Alabama, who, ironically had been 10-0-0 and voted #1 in 1964, then lost in a bowl game. But there was no poll after the bowls in 1964 so they went into the books as the national champion. But the Associated Press decided to poll their writers after the 1965 bowls for the first time ever, (they’d actually done it eighteen years before but that was “unofficial”- this was “official”). And the writers voted Alabama #1, even though the Tide at 9-1-1 still had an inferior record to the three 10-1-0 teams. The coach’s did vote after the bowls and so Michigan State and Alabama were co-national champions for that year. Alabama lost their openr to Gerogia by a single point and were later tied by Tennessee, so they were 3 points from a perfect record. .
1974- Oklahoma had the best team in the country and cruised to an 11-0-0 record, out-scoring their opponents 473-92. They were easily #1 in the writer’s poll. But the Sooners were on probation and the coaches had agreed to not vote for a team on probation so they voted Southern California, who was 10-1-1 as their #1 team. But nobody really thought they were the best team in the country.
1990- Colorado went 11-1-1 and was voted #1 by the writers. They played the toughest schedule in the country and one of the toughest in memory, playing 7 ranked teams and losing their one game by a point. They were only 3 points away from being 13-0-0. But they were also a 5th down away from being 10-2-1, using one to beat Missouri in a mistake the refs later acknowledged. Unlike Cornell in 1940, they refused to concede the game to their opponent. The coaches apparently didn’t care for that and voted for 11-0-1 Georgia Tech, who had played only three ranked teams.
2007- It was a “bonfire of the vanities” year as all the major contenders lost twice except Ohio State. LSU was considered the best of the rest and they beat the Buckeyes for the title
Note: in 1960, another “bonfire” year, Minnesota was voted #1 despite having been upset by Purdue, then got beat by Washington in the Rose Bowl to finish 8-2. They went into the books as the national champion, (in both polls), because there was no poll after the bowls. Had there been they would surely not have been voted #1: that probably would have gone to their conqueror, Washington, (10-1-0), Mississippi, (10-0-1), or Missouri (10-1-0), with their one loss being forfeited due to Kansas having an ineligible player).
So of the four teams with more than one non-winning game, three were co-champions with teams that either beat all their opponents or all of them but one. 83 of 87 teams is 95%. I think we can deduce from this that the definition of a national championship contender is a team that played a strong enough schedule to be considered a contender and won all their games or all but one.
I then looked at the rankings form one source, the writer’s poll, (because I wanted to avoid confusion from a team being ranked slightly differently from in one pole than another). I looked at how many times the teams ranked at each one of the top 20 positions has a record of having won all their games or all but one. Here’s the count I came up with:
#1- 73 teams have won all their games or all their games but one
#2- 71
#3- 56
#4- 52
#5- 41
#6- 36
#7- 19
#8- 18
#9- 10
#10-10
#11- 8
#12- 15
#13- 6
#14-10
#15-10
#16- 7
#17-7
#18- 5
#19- 6
#20- 5
The obvious cut-off here is the first six teams. Even at the #6 spot about half the teams ranked there have been undefeated or won every game but one. Teams ranked this high are typically teams that have been national championship contenders all year and they frequently have as good a record as some of the teams ranked in the top 4. They will tend to be from the “power” conferences or, occasionally, to be a non-power conference team that has totally dominated its conference and done well against power conference teams over several years and thus gained credibility, (Boise State). No team from a power conference with an undefeated or one loss record, (since we don’t have ties anymore), would be likely to be ranked lower than 6th. The undefeated or one loss teams ranked below sixth will tend to be mid-majors who have not established themselves as major national contenders, (Marshall a decade ago).
Thus it seems to me that the ideal playoff size would be 6 teams. You could have the #3/#4 teams play the #5/#6 teams and then the winners take on the #1/#2 teams. But I hate byes- I think any team in a tournament should have to win the same number of games as anyone else to win it. And there will be years when the #7/#8 teams might be viewed as contenders, as well. Thus, I’d go with an 8 team tournament- if it could just be a straight invitational. But I suspect that the power conferences would want automatic qualifications for their champions in an 8 team tournament. We could go to 16 teams and give automatic invitations to the champions of the 10 Division 1A conferences and at-large invitations to the highest ranked 6 teams that did not win their conference. Hold the first three rounds of a 16 team tournament in December and the championship game in a bowl game on a rotating annual basis. You can keep the other bowls, (I’d like to pair them down to a dozen with an 8 D1A win limit but that’s another subject). National championship controversies would be a thing of the past as they are in every other sport and every other level of this one.
A two team playoff was always in adequate. A four team playoff will be better but some worthy teams will still be left out. An eight team playoff would end serious controversies but a 16 team play6off may be needed to please everybody.
Can it be done? Well, FCs has a 20 team playoff. Division II has 24 teams and Division II has 32 teams. So yes, it can be done.
What would it look like in 2013? It might look like this:
A two team playoff:
Florida State 13-0 (689-139) vs. Auburn 12-1 (522-312)
A four team playoff:
Florida State 13-0 (689-139) vs. Michigan State 12-1 (387-165)
Auburn 12-1 (522-312) vs. Alabama 11-1 (465-136)
An eight team playoff just based on the BCS rankings:
Florida State 13-0 (689-139) vs. Stanford 11-2 (432-242)
Auburn 12-1 (522-312) vs. Baylor 11-1 (639-254)
Alabama 11-1 (465-136) vs. Ohio State 12-1 (602-277)
Michigan State 12-1 (387-165) vs. Missouri 11-2 (507-292)
An eight team playoff with automatic bids to the ACC, SEC, Big 10, Big 12 and Pac 14 champions:
Same as above, (this year).
A 16 team playoff:
Florida State 13-0 (689-139) vs. Louisiana-Lafayette 8-4 (415-323)
Auburn 12-1 (522-312) vs. Utah State 8-5 (424-225)
Alabama 11-1 (465-136) vs. Bowling Green 10-3 (460-192)
Michigan State 12-1 (387-165) vs. Rice 10-3 (408-298)
Missouri 11-2 (507-292) vs. Clemson 10-2 (482-253)
Ohio State 12-1 (602-277) vs. Oklahoma 10-2 (382-256)
Baylor 11-1 (639-254) vs. Oregon 10-2 (561-259)
Stanford 11-2 (432-242) vs. South Carolina 10-2 (409-240)
Comments: I’m not convinced that Auburn is really better than Alabama or even Michigan State so the two team playoff is inadequate, as it almost always is. The four team playoff looks a lot better and includes an Auburn-Alabama rematch with the possibility of the game everybody really wanted to see: Florida State vs. Alabama. The eight team playoff includes two teams that were in the national hunt all year but suffered late losses: Baylor and Ohio State. And Stanford and Missouri probably beat as many good teams as anyone. This year the champions of the “Big Five” conferences are in the BCS top 8 so there’s no different with automatic qualifications. The 16 team playoff looks bloated, as it usually will. But, again it’s perfectly doable as evidenced by the larger playoffs in other divisions. And it has the one match-up I most wanted to see: Baylor and Oregon. Still, if I had my druthers, I’d pick the eight team playoff with no automatic bids. Even if the power conferences demanded them, (which they normally wouldn’t need), I think that’s the ideal size for a D1A playoff.
What is a national championship contender? The generally recognized sources for “national championships” have been the writer’s poll which began in 1936, the coach’s poll, which began in 1950 and the BCSD, which began in 1998. These organizations have chosen 87 national champions in the 77 years from 1936-2012, (more than the number of years due to split titles). 83 of these 87 teams have either won all their games or all their games but one, (meaning they lost one game or tied one game). The other four:
1965- Michigan State, Arkansas and Nebraska were all 10-0-0 and ranked #1-2-3. All lost in New Year’s day bowl games. The highest ranked conqueror was Alabama, who, ironically had been 10-0-0 and voted #1 in 1964, then lost in a bowl game. But there was no poll after the bowls in 1964 so they went into the books as the national champion. But the Associated Press decided to poll their writers after the 1965 bowls for the first time ever, (they’d actually done it eighteen years before but that was “unofficial”- this was “official”). And the writers voted Alabama #1, even though the Tide at 9-1-1 still had an inferior record to the three 10-1-0 teams. The coach’s did vote after the bowls and so Michigan State and Alabama were co-national champions for that year. Alabama lost their openr to Gerogia by a single point and were later tied by Tennessee, so they were 3 points from a perfect record. .
1974- Oklahoma had the best team in the country and cruised to an 11-0-0 record, out-scoring their opponents 473-92. They were easily #1 in the writer’s poll. But the Sooners were on probation and the coaches had agreed to not vote for a team on probation so they voted Southern California, who was 10-1-1 as their #1 team. But nobody really thought they were the best team in the country.
1990- Colorado went 11-1-1 and was voted #1 by the writers. They played the toughest schedule in the country and one of the toughest in memory, playing 7 ranked teams and losing their one game by a point. They were only 3 points away from being 13-0-0. But they were also a 5th down away from being 10-2-1, using one to beat Missouri in a mistake the refs later acknowledged. Unlike Cornell in 1940, they refused to concede the game to their opponent. The coaches apparently didn’t care for that and voted for 11-0-1 Georgia Tech, who had played only three ranked teams.
2007- It was a “bonfire of the vanities” year as all the major contenders lost twice except Ohio State. LSU was considered the best of the rest and they beat the Buckeyes for the title
Note: in 1960, another “bonfire” year, Minnesota was voted #1 despite having been upset by Purdue, then got beat by Washington in the Rose Bowl to finish 8-2. They went into the books as the national champion, (in both polls), because there was no poll after the bowls. Had there been they would surely not have been voted #1: that probably would have gone to their conqueror, Washington, (10-1-0), Mississippi, (10-0-1), or Missouri (10-1-0), with their one loss being forfeited due to Kansas having an ineligible player).
So of the four teams with more than one non-winning game, three were co-champions with teams that either beat all their opponents or all of them but one. 83 of 87 teams is 95%. I think we can deduce from this that the definition of a national championship contender is a team that played a strong enough schedule to be considered a contender and won all their games or all but one.
I then looked at the rankings form one source, the writer’s poll, (because I wanted to avoid confusion from a team being ranked slightly differently from in one pole than another). I looked at how many times the teams ranked at each one of the top 20 positions has a record of having won all their games or all but one. Here’s the count I came up with:
#1- 73 teams have won all their games or all their games but one
#2- 71
#3- 56
#4- 52
#5- 41
#6- 36
#7- 19
#8- 18
#9- 10
#10-10
#11- 8
#12- 15
#13- 6
#14-10
#15-10
#16- 7
#17-7
#18- 5
#19- 6
#20- 5
The obvious cut-off here is the first six teams. Even at the #6 spot about half the teams ranked there have been undefeated or won every game but one. Teams ranked this high are typically teams that have been national championship contenders all year and they frequently have as good a record as some of the teams ranked in the top 4. They will tend to be from the “power” conferences or, occasionally, to be a non-power conference team that has totally dominated its conference and done well against power conference teams over several years and thus gained credibility, (Boise State). No team from a power conference with an undefeated or one loss record, (since we don’t have ties anymore), would be likely to be ranked lower than 6th. The undefeated or one loss teams ranked below sixth will tend to be mid-majors who have not established themselves as major national contenders, (Marshall a decade ago).
Thus it seems to me that the ideal playoff size would be 6 teams. You could have the #3/#4 teams play the #5/#6 teams and then the winners take on the #1/#2 teams. But I hate byes- I think any team in a tournament should have to win the same number of games as anyone else to win it. And there will be years when the #7/#8 teams might be viewed as contenders, as well. Thus, I’d go with an 8 team tournament- if it could just be a straight invitational. But I suspect that the power conferences would want automatic qualifications for their champions in an 8 team tournament. We could go to 16 teams and give automatic invitations to the champions of the 10 Division 1A conferences and at-large invitations to the highest ranked 6 teams that did not win their conference. Hold the first three rounds of a 16 team tournament in December and the championship game in a bowl game on a rotating annual basis. You can keep the other bowls, (I’d like to pair them down to a dozen with an 8 D1A win limit but that’s another subject). National championship controversies would be a thing of the past as they are in every other sport and every other level of this one.
A two team playoff was always in adequate. A four team playoff will be better but some worthy teams will still be left out. An eight team playoff would end serious controversies but a 16 team play6off may be needed to please everybody.
Can it be done? Well, FCs has a 20 team playoff. Division II has 24 teams and Division II has 32 teams. So yes, it can be done.
What would it look like in 2013? It might look like this:
A two team playoff:
Florida State 13-0 (689-139) vs. Auburn 12-1 (522-312)
A four team playoff:
Florida State 13-0 (689-139) vs. Michigan State 12-1 (387-165)
Auburn 12-1 (522-312) vs. Alabama 11-1 (465-136)
An eight team playoff just based on the BCS rankings:
Florida State 13-0 (689-139) vs. Stanford 11-2 (432-242)
Auburn 12-1 (522-312) vs. Baylor 11-1 (639-254)
Alabama 11-1 (465-136) vs. Ohio State 12-1 (602-277)
Michigan State 12-1 (387-165) vs. Missouri 11-2 (507-292)
An eight team playoff with automatic bids to the ACC, SEC, Big 10, Big 12 and Pac 14 champions:
Same as above, (this year).
A 16 team playoff:
Florida State 13-0 (689-139) vs. Louisiana-Lafayette 8-4 (415-323)
Auburn 12-1 (522-312) vs. Utah State 8-5 (424-225)
Alabama 11-1 (465-136) vs. Bowling Green 10-3 (460-192)
Michigan State 12-1 (387-165) vs. Rice 10-3 (408-298)
Missouri 11-2 (507-292) vs. Clemson 10-2 (482-253)
Ohio State 12-1 (602-277) vs. Oklahoma 10-2 (382-256)
Baylor 11-1 (639-254) vs. Oregon 10-2 (561-259)
Stanford 11-2 (432-242) vs. South Carolina 10-2 (409-240)
Comments: I’m not convinced that Auburn is really better than Alabama or even Michigan State so the two team playoff is inadequate, as it almost always is. The four team playoff looks a lot better and includes an Auburn-Alabama rematch with the possibility of the game everybody really wanted to see: Florida State vs. Alabama. The eight team playoff includes two teams that were in the national hunt all year but suffered late losses: Baylor and Ohio State. And Stanford and Missouri probably beat as many good teams as anyone. This year the champions of the “Big Five” conferences are in the BCS top 8 so there’s no different with automatic qualifications. The 16 team playoff looks bloated, as it usually will. But, again it’s perfectly doable as evidenced by the larger playoffs in other divisions. And it has the one match-up I most wanted to see: Baylor and Oregon. Still, if I had my druthers, I’d pick the eight team playoff with no automatic bids. Even if the power conferences demanded them, (which they normally wouldn’t need), I think that’s the ideal size for a D1A playoff.
