Good Wins In College Football Don’t Always Stay Good

To start the 2017 college football season, Alabama and Florida State played a neutral-site opener billed as a clash of the titans. Alabama had lost in the national championship game after the 2016 season, and Florida State had won the Orange Bowl. The teams were ranked first and third, respectively, in the preseason AP poll. They sold out the first college football game in Atlanta’s new Mercedes-Benz Stadium. ESPN ran promos for weeks. “College GameDay” made a visit. The winner, it seemed, would earn a prized nonconference victory and have the inside track to the College Football Playoff.

That’s not how it played out, though. Florida State quarterback Deondre Francois suffered a season-ending knee injury in the fourth quarter of Alabama’s 24-7 win, and the Seminoles struggled from there. They reached bowl eligibility only after rescheduling a postponed game against Louisiana-Monroe, and so Alabama was credited for beating a 6-6 underachiever rather than an ACC contender.

Alabama made the playoff anyway and went on to win the national championship. But the Crimson Tide’s case, and many others like it, illustrates the baffling catch-22 of the chase for the College Football Playoff: To build a playoff-caliber resume, teams have to compile impressive wins. But along the way, they might derail their opponents’ seasons, thus making their wins over those teams less impressive by the time December arrives.

The numbers behind this reality are revealing: Over the past 10 full seasons, teams ranked in the top 20 at kickoff were 1,823-583 (.758). But teams that finished the season ranked in the top 20 had gone 2,021-375 (.843), meaning that wins against those teams were harder to come by than against teams ranked only before each game. Teams ranked in the top 10 at kickoff were 970-237 (.804), while teams that finished the season ranked in the top 10 were 1,071-129 (.893).

When it comes to the most coveted wins — over ranked teams on the road — that discrepancy holds up. From 2009 to 2018, teams that entered games in the top 10 were 575-81 (.877) at home, but teams that finished in the top 10 were an even more dominant 629-45 (.933) at home. Want to boost your resume by taking down a true, lasting top-10 team on its home field? There simply aren’t many of those wins to be had — only about four or five per year, across the country.

Six years into the playoff system, the playoff committee still hasn’t articulated the basis for a “strong” win: whether it’s against an opponent that was highly ranked at kickoff or highly ranked at the end of the season. The procedures on the playoff website list four selection criteria: conference championships, strength of schedule, head-to-head results and comparative outcomes against common opponents — but not whether they measure “strength of schedule” by opponents’ record at the time of the game or at the end of the season. The closest explanation is that “committee members will be required to discredit polls” that come out before the season.

But that approach makes the job of sorting through resumes complicated every year. Minnesota, ranked No. 17 last week, pulled off one of the biggest wins in school history Saturday when it upset then-No. 4 Penn State. But in doing so, the Golden Gophers knocked their opponent down to ninth. If the Nittany Lions lose at Ohio State on Nov. 23, they might well fall out of the top 10, meaning that Minnesota wouldn’t have beaten a top-10 team, despite outplaying a group that the committee placed among the top four at the time of their game.

In the Big 12, Baylor has a flicker of hope for the playoff, but that would likely require beating Oklahoma twice — once at home on Saturday and once in the Big 12 championship game — which would in turn drop the Sooners to 10-3 at best, possibly pushing them out of the top 15. In the SEC, LSU escaped a hostile environment on Sept. 7 with a win against then-No. 9 Texas before Texas faded, taking some luster off that win. And the committee’s consensus is that Auburn is the No. 12 team, but if the Tigers lose to Alabama and Georgia this month, they’ll have suffered defeats against those two teams plus LSU and Florida, the four powers devaluing one another’s wins.

Since 2009, Alabama has the greatest difference between winning percentage against top-10 opponents pregame and top-10 opponents postseason.¹ Nick Saban’s program is 7-2 on the road against teams that enter in the top 10 and 1-3 on the road against teams that finish there, and it’s 16-4 against pregame top-10 teams anywhere versus 4-7 against top-10 finishing teams. This is no fault of Alabama’s. Perhaps because the Crimson Tide have been known to deliver some season-crushing losses, its opponents don’t hold up well enough to furnish Alabama’s resume.

Alabama bargained in 2015 for a high-stakes showdown to start 2017 and ended up with a so-so victory. That kind of highly anticipated nonconference matchup isn’t going away, so the top teams will have to hope that they continue to win them — and that their defeated opponents are well enough to walk away afterward.

Looking ahead: Week 12

Although neither game is quite as earth-shattering as last week’s LSU-Alabama clash was, two big matchups figure to reshape the playoff landscape in Week 12: Oklahoma-Baylor and Georgia-Auburn. In the latter, UGA is simply fighting to remain relevant in the College Football Playoff chase — the Bulldogs’ odds (currently 33 percent) would fall to just 18 percent with a loss. But in the case of the former, it could be another one of those natural elimination games for each team’s playoff hopes, where the winner stays alive and the loser is essentially finished.

Despite being undefeated, Baylor is just 13th in the official playoff rankings, and has only a 14 percent chance of making the playoff according to our model. (The Bears have played a paper-thin schedule to this point.) If Baylor loses to Oklahoma, any playoff dreams are obviously dashed. But if it wins, Baylor’s odds would rise to 29 percent, potentially putting the Bears right in the conversation for the final playoff spot. As for the Sooners, their playoff prospects already teeter on the edge of disaster; we give them a 28 percent chance, but there are very legitimate questions about whether OU would actually get the nod if it came down to them against Alabama for the final slot, with each carrying one loss. Oklahoma can help its case here, however, with a win — we think the Sooners’ playoff odds would rise to 40 percent if they beat Baylor, and we give that outcome a 63 percent chance of happening even though the game is in Waco.

Check out our latest college football predictions.