What’s Wrong with the College Football Playoff? Talent Concentration is Boring
The 2020 college football playoff (CFP) will feature four teams, following a selection structure that was put in place starting in 2014. For decades, the NCAA resisted a football tournament, unlike every other major sport in the U.S. College basketball ends its season with a playoff bracket, pro football has its Super Bowl to culminate playoffs in its two conferences, same as the National Basketball Association and Major League Baseball. However, the CFP tournament is extremely narrow — just four teams, which is fewer slots than the sport has conferences.
The effect of the CFP structure has been to create a strange crisis of boredom. The same few teams keep getting selected. The 2020 outcome is “a four-team playoff that is so clubby and familiar, you could have accurately outlined at least three-quarters of it on a hotel napkin a half decade ago,” writes Jason Gay, lead sports columnist for the Wall Street Journal. “1. Alabama 2. Clemson 3. Ohio State. 4. Notre Dame.” The Ohio State Buckeyes won the first playoff at the end of the 2014 season, were selected three other years, but were also the fifth team in the rankings in two years. Alabama and Clemson have appeared in every CFP except one. This is a rare year that the Fighting Irish were selected instead of the Oklahoma Sooners or the Oregon Ducks, but the strange crisis is that the same superteams keep getting in, and nobody knows why.
I have a hypothesis that explains the superteams. Let’s be clear up front that it is not bias on the selection committee that explains why the Crimson Tide dominates the SEC year after year, the Tigers of Clemson dominate the ACC conference, or the Buckeyes frustrate the Michigan Wolverines in The Game. Those three teams are genuinely top-tier programs, now more than ever.
My superteam hypothesis is that the CFP structure has reinforced the recruiting strength of leading football programs.
Decades ago, talented high school football players tended to be recruited locally and play for one of the college teams nearby. Texans played for the Longhorns or Aggies. Michigan kids played for the Spartans or Wolverines or maybe Notre Dame. But now young players know that if they play for an Alabama or Clemson, they are much more likely to be in the CFP and therefore get drafted higher and earn a much bigger contract in the NFL. The CFP is causing a concentration of talent, reinforcing the superteam phenomenon.
There have been three distinct eras — and structures — in college football. The CFP era that started in 2014 was preceded by a structure known as the Bowl Championship Series or BCS that featured a championship game sans playoff as well as a series of traditional bowl games from 1998 to 2013. The traditional structure in the literal past century was for conference champs to face off in the big bowl games: SEC in the Sugar, Big Ten versus Pac Ten in the Rose, and so on. The conferences have evolved under the new structures, but that’s another story. For now, let’s turn to the data on talent and see whether it has become more concentrated since 2014.
For my analysis, I used data on the top-ranked 100 individual players per year over the 20 year period 2000–2019. Each player has a composite ranking from 247 Sports, defined by the company as “a proprietary algorithm that compiles prospect ‘rankings’ and ‘ratings’ listed in the public domain by the major media recruiting services. It converts average industry ranks and ratings into a linear composite index capping at 1.0000, which indicates a consensus №1 prospect across all services.” Only five players earned a 1.0000 ranking, including Jadevon Clowney and Vince Young. Among this elite group, the lowest ranked player had a 0.9337 score.
The ranking services identify the most elite players with a “5-star” rating, about 40 players per year. Comparing the average number of 5-star recruits between the two eras, CFP and pre-CFP, I found there was indeed a concentration of talent in the top programs, notably the top seven programs. Alabama, unsurprisingly to anyone who follows American football, is the top recruiting team in the CFP era. However, the surprise is that Clemson was #6, and other colleges with exceptional recruiting have not seen equivalent team performance on the field, most notably Florida State.
In the BCS era, dominant teams such as Texas, Miami, USC, and Florida were great on the field and off (recruiting). Clemson was also good but not great back then.
This first show hints at, but doesn’t exactly prove, the concentration hypothesis. Another look at the data over a wider range of teams, sorted from most to fewest average 5-star recruits, shows the increasing skew more clearly. (see below). Moreover, the standard deviation — a measurement of concentration — rises from 0.67 to 0.74. That’s a big shift over 20 years of data.
In this second chart, one sees that over fifty colleges were able to recruit at least one 5-star player before the CFP era, but since then, only thirty-two colleges have done so, and only twenty programs have been truly competitive. Expect even more concentration if the CFP structure continues as is.
If the powers that be expand the playoff bracket to 8 or more teams, the concentration of talent is likely to ease and the competitiveness of games will improve. Other adjustments to the structure could also dilute concentration. I favor a rule limiting one team per conference in the 4-team bracket, though to be fair, that might just further concentrate talent within each conference (which would hurt Georgia and Michigan, but might help Oklahoma, Oregon and Notre Dame).
If college football players were allowed to be paid, talent concentration would be shattered. Suddenly, a 5-star running back at Virginia Tech, BYU, or even Harvard would make sense. The advertising royalties a superstar could make as a big fish in small pond would be much more attractive than the risk of benchwarming for a superteam. Not to mention, this would be a valuable hedge for talented athletes against the very common risk of injury that ruins their pro potential. With such a structural shift, Nebraska might even be good again.
__
Dr. Tim Kane has researched and written about talent management for over a decade. His book, Total Volunteer Force, (free PDF version) applied sports metrics to military HR. Write to him at tjkane@stanford.edu.
