As Clemson and LSU prepare to face off on January 13, let’s take a brief look at the history of college football national championships.
The first college football season in 1869 looked nothing like the sport we know today. Princeton and Rutgers played a pair of games against one another that more closely resembled soccer than the gridiron game with which college football fans are familiar in the 21st century. At the time, the idea of a national championship did not exist — though both teams make retroactive claims to the national title for the 1869 season.
This is a consequence of college football’s long, convoluted history with determining the national championship in its sport. For much of the sport’s history, there have been halfhearted attempts to determine the best team in the country. Different systems have determined different champions, and without a postseason national championship tournament under its aegis the NCAA has happily afforded a range of ranking systems official status toward determining the top team in the land.
As we approach the conclusion of the 2019-2020 college football season and wind down 150th anniversary celebrations, the concept of a definitive national championship is once again at the forefront of our consciousness. Clemson and LSU are set to duel in New Orleans on January 13 for the College Football Playoff national championship, the second straight year that the game will feature a pair of undefeated schools.
That should also make this the second straight season where there is unanimity of opinion on who is the best team in the country. The College Football Playoff and its predecessors — the Bowl Championship Series, the Bowl Alliance, and the Bowl Coalition — were all created with the intention of rendering the split national championship a thing of the past. More often than not, they have failed to achieve that goal.
That all comes back to the NCAA, which lost any chance to superimpose their own college football tournament at the FBS level when they lost NCAA v. Board of Regents of Oklahoma et al. in the U.S. Supreme Court in 1984. After that point, control of the top subdivision was firmly in the grip of the schools and conferences that make up the membership.
Thus the NCAA is all too happy to declare multiple systems as official national championship selectors. If they can’t control the playoff system at the top, they can throw a wrench into the machinery of the system devised by the teams at the top of the subdivision.
In this week’s Sunday Morning Quarterback, let’s look at a contentious history of national championship declarations and the teams that choose to claim those titles.
The first attempts to determine the national champion each season
For the first three and a half decades of college football history, the sport’s largely regional character and a relative dearth of intersectional contests between top teams rendered the concept of a national championship largely moot. Nobody really considered them or spent much time arguing over who was the best team in the country.
Caspar Whitney, the owner and editor of Outing magazine, was the first to name national champions in college football when he used a mathematical system to rate the best teams in the country from 1905 through 1907. Whitney was also a key figure in the creation of All-American selections when he published the first in 1889 during his time with Harper’s Magazine in collaboration with Walter Camp. He tired quickly of picking national champions, though, ending the practice after three years of tabbing Yale as the best team in the land.
After Whitney, the idea of national championships really didn’t take flight until the 1920s, when a series of mathematical systems started to come online and spit out national championship selections. What made this trend unique was that the economists, statisticians, and mathematicians devising the various championship algorithms also opted to apply them retroactively on the first half-century of college football seasons.
Frank G. Dickinson, an economics professor at the University of Illinois, devised the first of these new systems in 1926. Retroactively declaring Notre Dame and Dartmouth the national champions of the two previous seasons, Dickinson then named Stanford the top team of 1926. Dickinson’s selections ran through 1940, as he turned his attention toward other areas of economic study outside intercollegiate athletics.
A year later, sports publicist and statistician Carroll Everard “Deke” Houlgate, Sr. introduced another ranking system that went further in retroactively applying national championships. Houlgate applied his calculations backward into the 19th century, declaring champions for all years between 1885 and 1926 except 1886 and 1906. Working with his wife Dorothy, Houlgate continued publishing the annual rankings in newspapers through 1949.
When Richard C. “Dick” Dunkel, Sr. introduced his rankings in 1929, he did so without any retroactive application. From 1929 onward through the present, several generations of the Dunkel family have continued to calculate and release rankings each season.
Through the 1930s, other ranking systems were introduced that calculated out the top teams in the country. William F. Boand first released the Azzi Ratem system in 1930, the Williamson System came online in 1932, Litkenhous in 1934, and Poling in 1935. All except Litkenhous applied a limited retroactive calculation.
Only once during this period did anyone attempt to calculate out the national championship from the inception of college football. In 1933, Parke H. Davis did a retroactive study that named national champions for every year except 1910, 1917, and 1918. This effort by the former Princeton tackle, coach at three different schools, and longtime football historian effectively boiled down to an eyeball test. Unlike the other rating systems employed in the 1920s and 1930s, Davis was working with no algorithm or weighted factors beyond his own subjective selection.
And then the polls were born, complicating the picture further
These first nine efforts to determine the national championship, both retroactively and contemporaneously, were all the work of individuals tinkering with formulas and sifting through records. In 1936, however, the concept of how national champions were crowned completely changed with the introduction of a national football poll by the Associated Press.
The news organization first released a poll two years earlier, toward the end of the 1934 season. That first effort was aborted after its only release on November 15, leading AP sports editor Alan J. Gould to declare a three-way tie the following season between Princeton, Minnesota, and SMU. Other sports writers petitioned Gould to reinstate the poll on a regular basis to find a consensus on top teams rather than relying on one man’s opinion, and the poll has been operational ever since the 1936 season.
For a decade and a half, the poll functioned alongside the various ratings systems that remained in operation after the college football mathematics boom of the 1920s and 1930s. The Helms Athletic Foundation launched their own poll in 1941 that operated for the next four decades, and in 1950 the American Football Coaches Association launched its weekly coaches poll.
That started a proliferation of polls in the 1950s. The International News Service, Football Writers Association of America, Football News, and National Football Foundation all launched polls between 1952 and 1959 as the format of collecting votes and aggregating a increasingly became a staple of postgame newspaper coverage.
With multiple polls, however, came the possibility of a split national championship. As the number of retroactive rankings increased, more seasons came up conflicted as different mathematical designs and different sets of voters frequently failed to see eye to eye about the best teams.
What does this mean for how we think of national championships?
In a diffuse system with more than 100 football teams spread across the United States, there is always going to be a level of subjectivity when it comes to determining a national champion. How one chooses to calculate schedule strength, how much emphasis one puts on margin of victory relative to point-spread expectations, how one chooses to weigh more recent results against earlier games, and any number of other factors can wildly impact how two computers or two sets of voters perceive teams.
Because the NCAA has declared so many different rating systems and polls as official national championship selectors, that inevitably leads to a lack of consensus on who is the best team in any given year.
We tend to think of this as a problem of the past. Even in more recent seasons, though, the BCS and the College Football Playoff have often failed to satisfy at least one system that opts to declare a champion outside of the two finalists in what is perceived to be the national championship game.
The most famous case of this happened in 2003, when the Associated Press decided to crown USC the national champion despite LSU and Oklahoma playing in the BCS title game. This split led the AP to withdraw the rights to its poll as part of the BCS formula, leading to the creation of the Harris Interactive Poll to replace it as one of the two human polls in the hybrid system.
That is by no means the only instance of a split championship during the past two decades, however. We have also seen Group of Five powerhouse UCF claim a national championship after finishing 2017 at No. 1 in the Colley Matrix ratings as the only undefeated team in the country.
Over the past 150 years since the first college football game in New Jersey, fewer than one-quarter of seasons have ended with one team earning unanimous national championship selection. With 374 national champions declared over 149 seasons prior to the January 13 showdown between Clemson and LSU, there are far more years where multiple teams have a legitimate claim to champion status.
A three-way split for the national championship has been a more common occurrence than years where one team has earned all the accolades. This too is no historical anomaly; as recently as 2011, Alabama, LSU, and Oklahoma State could all make credible claims to the national championship.
The 1993 season saw Florida State win the Orange Bowl over Nebraska to claim the national championship as part of the Bowl Coalition. While the coaches and the Associated Press both agreed on the Seminoles as the top team in the country, various systems also afforded the Cornhuskers, Auburn, and Notre Dame opportunities to claim a title as well.
In 1921, and again in 1981, a half-dozen teams could claim a share of the crown. Five different champions have been named on a dozen occasions, and 14 more times there were four teams named by different services as the best in the nation.
More of these championships are claimed than one might expect
There was a lot of grumbling when UCF opted to thumb its nose at a College Football Playoff system that failed to even give them an opportunity on the field as they claimed the national championship via Colley. The Knights, however, were only following in a grand tradition of claiming titles from a range of selectors.
Out of the 374 national champions that have been named by one system or another dating back to 1869, only 260 are claimed as national championships. In many of these cases, such as Alabama’s 1941 Houlgate championship, there is only one selector going against conventional wisdom. All the same, if a team is named champion by an official NCAA selector, the team has every right to claim that crown.
Rare are the occasions when a team turns down the national championship. Like UCF, Utah ended the 2008 season as the only undefeated team in the country. Shut out of the BCS championship game as the Mountain West champion, the Utes instead took down Alabama in the Sugar Bowl to finish the year 13-0.
Though they were named national champion by three of the six BCS computers, Utah opted not to claim a share of the national title. Similarly, Wisconsin does not claim part of the 1942 title they technically share with Ohio State and Georgia.
Auburn could claim five national championships, but the Tigers only recognize titles in 1957 and 2010. Notre Dame could recognize as many as 22 championships; the Fighting Irish only claim half of those crowns. Alabama only claims 17 of a possible 22 championships, Ohio State only eight of a possible 17, Oklahoma seven out of 17, and USC 11 of a possible 17 titles.
On the other side of the ledger, teams claim titles where nobody even selected them as the best in the land. The Penn Quakers did this in 1907 after going 11-1 with their only loss coming against Carlisle. Columbia did the same thing in 1933, claiming the national championship after going 8-1 with a Rose Bowl win over Stanford. Neither school was actually selected as the top team; rather, they made the claim themselves and dared everyone to call them out.
No system can produce a definitive national championship every year
Part of what makes college football so wonderful is its chaotic nature. Rarely does everything go as we expect it will, and chalk means even less in this sport than it does in other intercollegiate competitions.
Average No. of National Champions per Year by System
Coalition (1992-1994): 3.0/year
Alliance (1995-1997): 1.7/year
BCS (1998-2013): 1.8/year
CFP (2014-present): 1.4/year
The past quarter-century of college football history has been marked by efforts to rein in the chaos. Starting with the Bowl Coalition in 1992, various systems have tried to find a way to crown a decisive, definitive national champion. It hasn’t always worked out as planned.
Out of 27 seasons since the launch of the Bowl Coalition, only 10 of them have ended with a single team named the national champion by every official NCAA selector. For comparison, the national championship has been split between two teams on 12 occasions over the same span, there has been a three-way tie in four different years, and the 1993 season ended with four teams all able to stake some claim to the title.
The College Football Playoff has seen teams split the title twice in five years. Before that, the BCS produced a decisive king of college football on fewer occasions (six) than those years where the system created a controversy around a split (seven). Ten of the 36 years where there is only one champion named have occurred in the past two decades.
The steady march toward the current four-team system has certainly eliminated some, though not all, of the chance of seeing the national championship split between two or more teams. It is far rarer to see a split now than it was in the 1980s. Even so, the mission has left wiggle room for a team to swoop in and lay claim to some part of the title.
So as you watch Clemson and LSU in the final college football game of the 2019-2020 season, realize that — win or lose — both teams will likely leave the field at the Mercedes-Benz Superdome in New Orleans with the opportunity to claim part of this year’s national championship if they so choose.