clock menu more-arrow no yes

Filed under:

Let’s Talk about the Playoff

New, 54 comments

If Baylor’s undefeated, the Bears are in.

NCAA Football: Pac-12 Media Day Kirby Lee-USA TODAY Sports

Two things are undoubtedly true: an undefeated Baylor will make the playoff, and Baylor is very unlikely to go undefeated.

I’ve called the playoff correctly every season. I thought the committee should have Baylor or TCU over Ohio State in 2014, Penn State over Ohio State in 2016 and Ohio State over Alabama in 2017. But I was able to put aside what I would do and guess what the committee would ultimately do.

Let’s start with showing why an undefeated power five team will never miss the playoff, turn to why Baylor is very likely to lose a game and then discuss what Baylor’s situation would be with one loss.

No undefeated power five team will ever miss the playoff:

We now have five years of the playoff. We’ve had six undefeated teams in the playoff (five power five teams and Notre Dame). All of those teams have been at least the No. 3 seed in the playoff.

In 2014, Florida State wasn’t one of the four best teams. The Seminoles won seven games by less than a touchdown. In their final regular season rankings, the committee even moved TCU up to No. 3 and dropped Florida State to No. 4. But after TCU walloped Iowa State, and Florida State barely beat Georgia Tech, the committee moved Florida State to No. 3 and dropped TCU to No. 6. Almost everyone who watched Florida State thought Baylor, TCU and Ohio State were better, but with Florida State winning, the committee elected to not make a subjective judgment and discount a team that won every week.

The committee has valued losses the most. The committee has never put a two loss team in the playoff over a one loss team. In 2017, the committee was prepared to put two loss Auburn into the playoff if they had knocked off Georgia in the SEC Championship game. But that Auburn team wasn’t competing against an undefeated team. Rather, they were competing against one loss Alabama—a team they’d just beaten by 12. That Auburn team also had a win over Georgia, and if they’d won the SEC title game, they’d have a conference title and another win over Georgia.

Losses have been the committee’s favorite metric because the committee has been able to avoid controversy by falling back on wins and losses. Let’s turn back to 2014. That season the committee committed the fallacy of A>B, B>C, but C>A. Daniel Kahneman won a Nobel Prize for identifying that humans continuously fall victim to that fallacy. That year, the committee thought that TCU was greater than Ohio State, Ohio State was greater than Baylor, but Baylor was greater than TCU. When the committee just uses losses, they can avoid the disaster Jeff Long had to explain in 2014.

Sure, but can we be certain past behavior is predictive of future results?

In this case, yes. It’s almost impossible to finish a power five season undefeated. Again, only five teams have done that in five seasons. In the Big 12, a team has to win 10 Big 12 games (nine league games and the title game) to do that. Nobody has gone undefeated in Big 12 play since Texas in 2009.

The committee’s protocol says that:

The committee will select the teams using a process that distinguishes among otherwise comparable teams by considering:

The first point in that process is “otherwise comparable teams.” The committee has felt like undefeated teams are not “otherwise comparable” to teams with a loss. Take Penn State and Ohio State in 2016. Although Penn State had two big reasons to be selected over Ohio State—head-to-head and conference title—the committee did not feel like those teams were comparable. Ohio State had one loss, while Penn State had two. Ohio State also beat Big 12 Champion Oklahoma in Norman. “Otherwise comparable teams” is a condition precedent to the other factors. And if a team is undefeated, the committee isn’t going to even turn to those other factors.

If teams do become comparable though, then these criteria matter:

Conference championships won,

Strength of schedule,

Head‐to‐head competition,

Comparative outcomes of common opponents (without incenting margin of victory), and,

Other relevant factors such as key injuries that may have affected a team’s performance during the season or likely will affect its postseason performance.

An undefeated power five team is going to always have a conference championship won. Usually the one loss team we’re comparing to the undefeated power five team will not have that factor. Their conference champion should usually be in the playoff already too. For example, this year if LSU goes 13-0 and Alabama finishes 11-1 with only a loss to LSU, then 13-0 LSU will already be in the playoff and have the SEC title. Alabama—if they’re deemed comparable to 13-0 Baylor—will not have a conference title.

The 13th game and conference title also gives a big boost to strength of schedule. Baylor will get another top 25 win against Oklahoma (or possibly another Big 12 team that goes wild and solidly establishes themselves as a top 25 squad). To go 13-0, Baylor will also have wins over a couple of other Big 12 squads that finish in the top 25. Right now, Kansas State and Oklahoma State look like they could be ranked. While Alabama receives a ton of credit for being in the SEC and their brand, if the Tide finish 11-1, they’ll have one top 25 win (Auburn) and maybe one other (Texas A&M). That’s not going to be enough to win, even assuming the teams are comparable.

So Baylor’s in the playoff, right?

We all fall victim to the status quo bias. We assume things will continue as they are. Wisconsin looked like a legitimate playoff team two weeks ago. Illinois and Ohio State ended that illusion. Oklahoma seemed destined to go 13-0. Kansas State went on a 48-13 run to end that dream.

Baylor’s going to be favored in a bunch of games, but let’s take a look at those odds. Let’s even give Baylor great odds against all their opponents. If Baylor’s an 85% favorite against West Virginia, an 85% favorite against Kansas, a 65% favorite at TCU, a 65% favorite against Texas and a 55% favorite against Oklahoma, then a 50% favorite on a neutral field against Oklahoma, Baylor would have an 8.4% chance of finishing undefeated. Again, those odds are probably too high for several of Baylor’s games, but that still puts Baylor at less than 10% to go undefeated.

What about a one loss Baylor?

There are so many permutations, it’s difficult to figure that one out. Baylor would be below any undefeated power five champions. They’d finish below a one loss SEC champ or Ohio State with one loss.

But we don’t know how the committee will view a non-power five champion or Oregon/Utah against Baylor. Does Baylor lose a close game to Oklahoma in Waco? That loss wouldn’t be disqualifying like Purdue’s evisceration of Ohio State in 2017. But does Auburn keep losing, which makes Oregon’s loss look worse? Does Texas A&M go on a run to give Alabama another top 25 win? Does Texas keep collapsing to make LSU’s strength of schedule and non-conference win look worse? Does Clemson drop a game, and then what does the committee do with their schedule—they’re not going to have a top 25 win.

I’d say that Baylor’s odds don’t look great with a loss, but it really depends on so many factors. 2014 was a terrible year to have one loss because Ohio State and TCU were also there. In 2015, every one loss power five champion made the playoff.

Baylor doesn’t need any help, but it takes a lot to go undefeated in college football. If Baylor can do that, then they’ll be in the playoff.