How to crown a national champion

by Nov 15, 20216 comments

It was 101 years ago Henry L. Farrell fittingly described it as the mythical national championship. One hundred one years later, not much has changed. There have been several attempts to do so, each one reactive, with a heavy dose of recency bias. The political climate seems ripe once again to usher in the newest attempt to correct a failed system. And unless sane minds prevail, we’ll end up with more of the same.

The problem isn’t that Cinci might not get in this year, just as it isn’t that UCF or Boise State have been snubbed in years passed. It’s not SEC bias, real or imaginary. Nor is it we lack variety or a more exciting postseason. Tailoring your solution to address one or more of these is short-sighted and won’t produce lasting, meaningful change. The problem is the current postseason construct does not include all the teams that have earned a shot at the title based on that year’s performance.

The polls

The NCAA, for whatever reason, was never obliged to determine a national champion. In the early 1930s, an Associated Press sports editor by the name of Alan Gould was the one-man operation that produced a weekly top 10. The fact it grew to a panel of 60 voters did nothing to alter his stated original purpose, which was “to develop interest and controversy. … Sports then was living off controversy, opinion, whatever. This was just another exercise in hoopla. … Making it a Top 10 was an arbitrary decision.” For more than six decades, unbelievably, this was the gold standard, and safe to say it accomplished its goal, but it doesn’t accomplish ours.

Let’s pretend for a moment the AP’s goal had been pure – to determine the national champion faithfully and accurately. How many actual games are these voters watching in less than 24 hours? How are they even remotely close to being in position to properly evaluate even two teams, much less 10 or 25? The coaches’ poll, which began in 1950, was positioned even worse. A coach is painstakingly reviewing game film of his next week’s opponent. He can probably tell you more about that one team than anybody in the country, which is great, except the evaluation’s shelf life is a week and neglects every other team, the one exception technically being his own, which certainly doesn’t produce an elephant in the room. How can such a chief stakeholder be expected to not show favoritism without recusing himself?

And then there were two

Our reactive tendencies produced the Bowl Championship Series (1998-2013) and its predecessors, the Bowl Coalition (1992-1994) and the Bowl Alliance (1995-1997). Despite their differences, all three were based on the faulty premise that each and every year produces two and only two teams worthy of a shot at the title.

The BCS even attempted to appear to address the egregious flaw that plagued the sport for the 50-plus years prior by instituting computer rankings. However, Thomas Callaghan, Peter Mucha, and Mason Porter in their mathematical review of the BCS pointed out “the creators of most of these systems guard their intellectual property closely. An exception is Colley’s ranking, which is completely defined on his website. Billingsley, Massey, and Wolfe provide significant information about the ingredients for their rankings, but it is insufficient to reproduce their analysis.” If we cannot reproduce their analysis, we are left with no alternative but to blindly trust the validity of their algorithms. Even if we could do so safely, which is preposterous, the computer ranking was only one factor in the BCS formula, which also included the coaches’ poll, the AP poll, and later the Harris poll. Egregious flaws were not only baked in but were, in fact, the main ingredients.

Playoff? Playoff?!

That leaves us to address the misnomer that is the College Football Playoff. As Steve Deace of the Michigan Podcast accurately described it, it’s not a playoff; it’s an invitational. And who is doing the inviting? In regressive fashion, it’s a committee, only now it’s not sports writers, but seven current athletic directors, a college professor, two former Big Ten offensive linemen, a vice president for athletics and community wellness, a businessman responsible for the operational and financial performance of Entergy’s five utility companies, and a retired head coach. How does an athletic director who is responsible for as many as 23 varsity sports find time to judge the merits of 25-plus football teams in a 12- to 60-hour period every week for six weeks? Forget athletic directors. Name me one person who could do this. Say his/her sole profession is to evaluate college football teams. A cursory review disqualifies 77 percent of this panel. Let’s assume the remaining three members are that dedicated. They’d still be the minority faced with the uphill battle of convincing the majority of the superior quality of their findings.


Essentially, all the CFP has done is expand the field from two to four, which brings us to today. Rumors are raging the “playoff” is in for more expansion, possibly 12 teams. It is very important you get this right. And by “you”, I mean anyone who is of rational, objective, evidence-based thought, not necessarily those who are stakeholders or stand to profit monetarily from the current or future construct. You water this thing down with 12, there’s no going back. Has there ever been a time in history when there were 12 teams in a given year that legitimately deserved a shot at the title? How about nine? Frankly, the answer is a resounding no. And based on the results of the last seven years, we also know four is not enough. Full disclosure: for more than a quarter century I’ve believed eight was the right number. But I hope the facts I’ve submitted to a candid world prove its merits.

It has to be eight for a variety of reasons. We’ve established you can’t just vote for who you think is best. It has to be competitive in nature. It must be a playoff, where the question is answered on the gridiron by the principal characters. And in the spirit of competition, let’s eliminate the idea of giving one or more teams a bye in the first round.

Very few things in this world are perfect, but that shouldn’t stop us from trying. We also shouldn’t use it as an excuse to ignore blatant shortcomings on the grounds that it’s better than what we had. The power-5 conference champions should receive automatic bids. What this measure lacks in perfection it makes up for in its eliminating subjectivity. Will the non-conference strength of schedules suffer as a result? Probably. But I don’t know how much worse it can get than seeing powerhouses play Charleston Southern, Mercer, Stony Brook, New Mexico State, The Citadel, etc. And I’m willing to live with not seeing those marquee non-conference matchups if it means taking the power out of the hands of those behind closed doors to place it in the hands of the athletes.

That leaves three spots. Outside of settling beforehand on a reproducible algorithm that produces the best evidence-based ranking of teams, I don’t see how to avoid some level of subjectivity surviving the selection process. But cutting the subjectivity down from absolute to just 37.5 percent is a substantial improvement. As an incidental biproduct, your Cincis, Central Floridas, etc. have three mathematical possibilities of getting in. Before this year, they had no shot. Again, three shots are a whole lot better than none. If by some miracle Cincinnati ultimately receives legitimate consideration this year, the non-power-five will have proven it has one shot. Nevertheless, three is better than one.

If you want to keep the bowls, knock yourself out; incorporate four into an eight-team playoff. Let the teams left out, vie for the other bowl games as they always have. All the other details are secondary and tertiary. The greatest sport on Earth deserves a proper playoff. It’s incumbent upon the people to make it a reality. How to go about that, there I cannot help you. But it starts with knowing what’s right.



  1. Tremendously well researched and very well written article. A football fan for over 5 decades, i have also thought the 8 team (including the conference champs of the power 5 conferences) format to be the best format to decide a true national champ. No playoff that doesn’t include the regular season champs not only doesn’t have credibility, it advocates for schools to play the easiest schedule possible. I think if the power 5 champs were guaranteed a berth in the playoffs, then many bigger schools might be tempted to play tougher non-conference opposition to toughen up their team for the run for being conference champs; and thus a berth in the playoff. There would still be teams crying foul for the voting for the other 3 slots, there will always be complaining regardless how it is set up. If you want to win the national championship, win your conference 1st. If you are one who advocates for more Cinderella stories, tune into Walt Disney. Disney can make your dreams come true, but let’s stop dreaming about a national champ and decide it on the playing field instead.

    • Thank you for the comment! And you make an interesting point about the impact this would have on the strength of the non-conference schedules. Perhaps it is premature to predict this would result in exclusively cream puff schedules. The incentive to prepare your team for the conference slate and/or strengthen your schedule to better position your team to grab one of those three at-large berths might be too great to pass up.

  2. Well written article and going back with past attempts at getting this right, shows a lot of inherent problems with trying to get to the desired outcomes.
    Strength of schedule is also a very subjective exercise that is only doing the alignment in hindsight.
    Flaws in every attempt are uncovered only a week later when teams expected to win are then beat, again throwing all the polls to then try to decipher the true teams that should be in the final 4.
    This is never done in a vacuum, so there is always biased results and perceived miscalculation as to the end result.

    Overall, great piece showing the problems beginning to end of having people with already not having enough information or film study of games in making a good unbiased decision.

    Will be watching for further opinions from this writer.

    • You’re right. So many variables to consider. Ranking really is an impossible exercise. One thing I think gets overlooked when it comes to strength of schedule is, say a team is highly-ranked when you play each other, but a rash of injuries and a couple tough losses causes that team to almost tank it. These are young, emotional young men. It’s not just who you play, but when you play them. It’s just one more thing potentially worthy of consideration in the arduous task of comparing teams.

  3. “Has there ever been a time in history when there were 12 teams in a given year that legitimately deserved a shot at the title? ”
    Response: Has there ever been a time when there were 8 teams in a given year that deserved a shot at the title? The answer is also a resounding NO.

    “We also shouldn’t use it as an excuse to ignore blatant shortcomings on the grounds that it’s better than what we had.”
    Response: The article then proceeds to excuse a main flaw of its proposed system, the continued use of a committee to select the at-large teams to send invitations to, with “…I don’t see how to avoid some level of subjectivity surviving the selection process. But cutting the subjectivity down from absolute to just 37.5 percent is a substantial improvement.”

    “it’s a committee, only now it’s not sports writers, but seven current athletic directors, a college professor, two former Big Ten offensive linemen…”
    Response: You do a great disservice to one of those linemen. John Urschel played for Penn State, but he also played in the NFL for 3 seasons. And the article also left out that Urschel has a doctorate in mathematics from MIT where he created dozens of models.

    • The eight-team format is designed to reward major conference champions and mitigate the impact voting has on the selection process. There are five major conferences, byes are unfair and undeserved, so eight is the most logical number.

      I acknowledged that subjectivity is still a part of this new proposed system, but 37.5 percent is, again, substantially better than 100 percent.

      I highly doubt John Urschel’s math models are either employed or hold significant weight in these discussions. I think you’re assuming a lot and giving the committee the benefit of the doubt when I’m not sure they deserve it.

Submit a Comment


Keith Eichholz

Lead writer for the Voice of College Football. Thorough, evidence-based, critical thinker. Husband to a beautiful wife, father to a terrific kid, always looking forward to football Saturday.

Recent Articles

McCarthy makes case

McCarthy makes case

You might think it unfair to expect McNamara to be better than the quarterback of the reigning national champions in any area, let alone most. But, my friend, that is the entire point. It’s not enough for McNamara to be as good as Bennett. He has to be be better than Bennett because Bennett has better players — significantly better. And you got to make up the ground somehow. McCarthy is Michigan’s best shot at that somehow.

UM’s QB controversy looms

UM’s QB controversy looms

If you were to take all of the first- and second-string B1G quarterbacks and those coaches had their pick of any of them to run their offense, Jim Harbaugh is the only one who would choose McNamara before McCarthy. The reason Harbaugh is the only one who’d take McNamara first is because he’s the only one compelled to justify making him the guy to this point. Yes, he beat Ohio State, won a B1G championship, and got to the playoff. But you got your doors blown off by the most talented team in the country in Georgia. And if you can win a conference championship starting your second-best quarterback, just how inferior is the Big Ten compared with the SEC? If Michigan wants to be nationally relevant, it has to stop insisting on playing with one arm tied behind its back.

Penn State proves clutch

Penn State proves clutch

The second marquee B1G matchup of the year was a thriller, as Penn State outlasted Purdue for an improbable victory Thursday night, 35-31.