Controversy and lingering doubts are what make college football great. I like that some people will never agree on who was the best team in a season. That started to diminish in the BCS era, and it might go away altogether starting in 2014 (with the new four-team playoff). We will still always have to wonder about some great seasons, though...
The main issue is that of undefeated teams not being ranked #1. It's not such a big deal. In recent years few teams have ended up with perfect records, but look back into the past - it used to be common for several teams to go undefeated and untied. People were used to it and it was never a big deal when an obviously lower-tier undefeated team did not get into the biggest bowl games or didn't win a Mythical National Championship. A great example would be Arizona State in 1975: they went undefeated, but at the time they were an unproven program. They only got to play in the Fiesta Bowl, a minor bowl at the time. They won that game and were the only undefeated/untied team in the country. Nonetheless, the polls ranked 11-1 Oklahoma #1, and rightly so. The 12-0 Penn State team of 1973 was ranked 5th by AP. The 12-0 Toledo team of 1971 was ranked 14th. Plenty of examples can be found. Look at the 1913 season - there were six major unbeaten/untied teams, and a 10-0-1 team. Who was the best? In 1922 there were 8 major teams without a loss. How about 1977? - six major teams with 11-1 or 10-1 records. We quite often need the help of a computer algorithm. Teams that play weaker schedules and don't win impressively cannot be considered the best. I believe computer programs can usually sort out the answers in those seasons better than humans can.
There are a handful of teams that I would personally assign a higher ranking to than my program does. Those would be 2008 Utah, 1997 Michigan, 1980 Georgia, 1976 Pittsburgh, 1959 Syracuse, 1950 Oklahoma, 1920 Texas, 1919 Texas A&M, and 1907 Yale. That is a pretty small number of complaints. Looking at what needed to be done to maximize the ratings of these teams (and some others) was a large part of what shaped my program. In the end, anyone who tries this will find no algorithm is perfect, and a few teams will always be left with questionable rankings. I think I have minimized the odd rankings as best is possible.
My two cents on some seasons:
2003 - Choosing between 13-1 LSU and 12-1 USC is difficult. They had two common opponents, and the results are interesting. LSU won at Arizona by 46, while USC won at Arizona by 45. LSU beat Auburn by 24 at home; USC won at Auburn by 23. LSU won by one more point in each case. However, the fact that USC was on the road for the Auburn win does make that win more impressive. Quibbling over points in similar victories never settles anything, though, so keying in on each team's lone loss is probably the way to go. LSU lost to an 8-5 Florida team, while USC lost to an 8-6 Cal team. Florida was ranked significantly higher than Cal in most computer ratings, and so it appears that LSU's loss was less of a blemish than USC's.
1997 - No problem with #1 Nebraska, but most people would probably want to see Michigan at #2 (if not #1). I believe Michigan deserves to be ranked 2nd. Nonetheless, few mathematical methods will place them that high. Here is an attempt to explain why... Compare Michigan and Florida State. FSU had only two close games: a 3-point loss at #4 Florida (hardly a bad game - throw in home field advantage points for FSU and it was like a tie) and a 7-point win over #24 Clemson. Meanwhile, Michigan had four wins by 7 or less, including a struggle with #37 Notre Dame (who beat #94 Hawaii by only 1). Michigan beat Wisconsin by 10, while Syracuse beat Wisconsin by 34. The best team Michigan beat was #9 Washington State, winning by only 5. Arizona State beat Washington State by 13. Florida State beat Ohio State by 17, while Michigan beat OSU by 6. Look at the champions list and you'll see that Entropy has highly regarded Michigan throughout history. I call Michigan the best in 1925, 1926, and 1964, where many do not. Finally, consider that few, if any, computer ratings had Michigan #1 for 1997 (none that I know of did). On Massey's comparison, Michigan's average rank was #3, and one system had them 7th! So, Entropy's #3 ranking is perfeclty reasonable.
1993 - Some thought Auburn was the best team in the land. I don't know why. They did not play in a bowl because they were on probation, but I do not judge them on that. Of the ten D-1A teams they beat, only one is in Entropy's Top 20. That game was Auburn beating #5 Florida by 3 at home. Winning by 3 at home is effectively a draw. They only beat #69 Mississippi State by 4 (at home) and #54 Vanderbilt by 4. They won only two of their games by more than 20 points. Finally, Auburn beat D-1AA Samford 35-7. Samford was defeated by equal or greater margins by AA foes Central Florida and Southwest Missouri. It was a rare sort of down year for the SEC - only two SEC teams made Entropy's top 25. I believe ranking Auburn #11 is accurate. Maybe the team was not very motivated, knowing it was not bowl eligible.
1984 - Only BYU went undefeated in '84. However, in the regular season they played just one top-40 team, #33 Air Force, winning by only 5. Their bowl opponent was a mediocre Michigan team (6-6); BYU only beat them by 7. BYU had close games with #54 Hawaii (18-13), #61 Wyoming (41-38), and #58 Pitt (20-14). With some Top 20 teams on their schedule, BYU would probably not have gone undefeated. No other team was particularly great in 1984, but the best of them was Florida.
1983 - No one went undefeated in 1983. The best team is an easy choice: Nebraska. They dominated their regular season schedule, and lost in the ultimate close game 31-30 at Miami. Give points to Nebraska to compensate for Miami's home field advantage, and they were the best team that night. Miami's loss was a 28-3 crushing by Florida. Miami beat Florida State by only 1 - three other teams beat Florida State by more, including a sketchy Tulane. Also, Nebraska had the best record (12-1) in Division 1A. Miami got the MNC by virtue of timing.
1980 - Georgia won the MNC, but they clearly were not an overpowering team. Half their wins were by 7 points or less. They beat #42 Tennessee by 1, #44 Auburn by 10, and #56 Ole Miss by 7. The best team they beat was #9 Notre Dame (by only 7). Still, how can a two-loss team be ranked above Georgia? Well, examine the losses of the other top teams:
With the exception of Oklahoma's, these losses are epitomes of the "quality loss." Having not played any of these teams, Georgia hits a mathematical brick wall of sorts. Georgia's schedule was the weakest in the top ten, and many of the top ten teams had greater average margins of victory.
The NCAA's archives show that among major polls and computers, five different teams have been awarded #1. It is quite common for predictive rankings to not have Georgia #1 in 1980. However, I fully agree that Georgia deserved the Mythical National Championship.
1973 - Entropy has Ohio State at #1 with a tie on their record. The tie was at #5 Michigan. Michigan was also undefeated outside this game, and tying an undefeated Michigan team in Ann Arbor is as good as winning in my book. Notre Dame topped the AP Poll. Notre Dame's only flaw was a struggle to beat #39 Michigan State 14-10. Ohio State and Michigan both soundly shut-out Michigan State. I have no strong opinion on who should be #1, but judging by Ohio State's incredible scoring defense (four shutouts, two games allowing 3 points, and two games allowing 7), they were a truly elite team.
1970 - Nebraska is a fairly well-accepted #1 for 1970, but close inspection shows Notre Dame deserves consideration. Notre Dame beat Army by 41. Nebraska beat Army by 28. Both teams beat Missouri by 14, and both had close games with LSU (Nebraska by 5, Notre Dame by 3). Nebraska tied a USC team that lost 4 games. In the end, Entropy gives Notre Dame the nod because they were the only team that beat powerful Texas. UPI had Texas #1, and the NFF had Texas and Ohio State #1. There were several teams people might argue for #1...
1967 - A difficult year for rankings. The AP #1 Southern Cal lost to Oregon State. Oregon State lost to BYU and Washington. USC did beat a strong UCLA team that beat Tennessee. Tennessee lost to Oklahoma. Oklahoma lost to Texas, and Texas lost to USC! USC barely won a few of their games. I kind of view this as a year where no one deserved the #1 ranking.
1966 - Alabama was the only major power with a perfect record, and since Alabama had a pretty incredible run through the Sixties, it would seem fitting for this team to be #1. However, just as happened in 1973, there were also two undefeated powers who tied each other in a rather famous game: the 10-10 tie when Notre Dame visited Michigan State.
Looking at Notre Dame's results, it is easy to believe they were the best team. They shut-out six opponents, including both Oklahoma and USC on the road. Notre Dame's schedule was stronger, and their average margin of victory was greater than that of Alabama. The tie was on the road at Michigan State, so one could argue Notre Dame was the better team in that game too.
1965 - Alabama probably claimed AP's #1 by virtue of winning their bowl game, while Michigan State lost theirs. However, as to who was really the best team, Michigan State has a great argument. They had a better record than Alabama: 10-1 versus 9-1-1. Michigan State's loss was by 2 at #2 UCLA. Alabama's loss was by 1 at #16 Georgia. Michigan State's closest win was by 4 at #6 Purdue. Alabama had two close calls, winning by 1 over #10 Ole Miss and by 3 over #23 Mississippi State (which had a losing record).
1964 - Arkansas did not win impressively against Oklahoma State, Wichita State, Rice (each won 4 games), or Baylor (won 5). They did beat two elite teams in Texas and Nebraska, but both were close games (1 and 3 points, respectively). No other team proved that much better than Arkansas, but Michigan probably would have been favored if the two played. Michigan's only flaw was a one-point loss to a Purdue team that finished #18 in the AP Poll, and which owns the #9 Entropy ranking.
1960 - A strange season. The Big 10 takes six of the top ten positions, meaning the conference generally dominated its non-conference opponents. The AP Poll had Minnesota #1 in its final release (before the bowl games), despite a loss to 4-4-1 Purdue. After Minnesota lost the Rose Bowl to finish with two losses, I would guess most people would choose Mississippi for #1 since their only blemish was a tie. However, I think Iowa, with a single loss to Minnesota, deserves #1. Mississippi gets their due in 1959 and '62.
1959 - I'm annoyed by this season. I would like to see Syracuse take the #1 spot. They dominated most of their opponents. The only close call was a 20-18 win over #6 Penn State. Syracuse's ranking isn't determined as much by their performance, as it is by the limited linkage between regions of the country. Teams played fewer non-conference games at this time, and non-conference games are a primary key for accurate rankings of teams relative to teams they did not play.
Southeastern teams played tough schedules, and dominated their small number of non-conference opponents in 1959. Mississippi's schedule strength was ten points greater than that of Syracuse. The result is that almost any network-based ranking system will rank Mississippi #1 and many will rank LSU #2, whether they really deserve it or not. We will never know the truth.
1956 - Some claim Oklahoma does not deserve #1 because of their weak schedule. Well, we also have to look at how they performed against that schedule. Oklahoma beat Texas 45-0, while Baylor beat Texas 10-7. Baylor beat Tennessee in the Sugar Bowl, and Tennessee is a team some mention as a possible #1 when claiming Oklahoma didn't deserve it! I completely back Oklahoma.
1943 - It's fascinating that Notre Dame is a unanimous #1 by all the historians I've seen. The one team Notre Dame lost to (Great Lakes Navy) was defeated by the only major undefeated/untied team - Purdue. Today there would be an outrage over Purdue not being named #1. I am happy to be one of the few backing Purdue for the national title.
1941 - One system out there (Houlgate's) actually ranked Alabama #1 for 1941. Ludicrous. The University of Alabama should not claim a National Championship for 1941 (which they do - check their media guide). Alabama lost to two teams that were good but not great (8-2 Vanderbilt and 8-1-1 Mississippi State). Almost everyone else agrees with Entropy: Minnesota #1.
1936 - Minnesota's results seem impressive at first glance, but it is an illusion. In most seasons their schedule would have been imposing, but odd things happened this year. For example, 1936 was probably Michigan's worst season in history, with only one win. Four of Minnesota's eight opponents had losing records. Minnesota lost to their best opponent (Northwestern). Entropy gives 9-1-1 LSU the #1 ranking.
1926 - Alabama had one very poor result - they beat Sewanee just 2-0. Sewanee was 2-6 and ranks #160. Alabama tied Stanford. Stanford had two other close games. None of the major teams had a perfect record, and so a one-loss Michigan team takes #1.
1925 - Alabama did not win impressively against one average team (Georgia Tech) and one bad team (Mississippi State). They won the Rose Bowl by 1 point, meaning they should be ranked just above Washington (which they are). Washington tied a 4-2-2 Nebraska team. Alabama's schedule was simply too weak for the rest of their games to build a strong case for #1. Against eight strong opponents, Michigan allowed only 3 points. Michigan was probably the best team in the land.
1921 - Cornell's scores were impressive, but they played mostly very inferior teams. None of their opponents make Entropy's top 20. Meanwhile, Iowa played an elite schedule and won impressively. Iowa was the only team to beat Notre Dame. I don't understand why this team has not gotten more credit in the annals of college football history.
I have not studied the game in the pre-1900 era enough to have opinions. If you have comments on those seasons (or any season for that matter), please write to me (link on the main page).