Best vs. most deserving: top-4 teams will often miss the Playoff

In 2018, Georgia narrowly missed the College Football Playoff field, slotting #5 in the selection committee’s final ranking before the big event.

Selection Sunday came a day after Kirby Smart’s team went toe-to-toe with top-ranked Alabama in the SEC Championship. Within a few minutes of that loss, Kirby Smart was at a microphone arguing for his 11-2 team’s inclusion.

“Give that coach across the sideline a vote who he doesn’t want to play,” he said, meaning his ex-boss Nick Saban. “It will start with us. I promise you he don’t want to play us. It’s not our decision, it’s their decision. But if you’re going to put the best four football teams in ...”

Indeed, Saban said immediately after the game he didn’t want to play Georgia again.

Smart was also right that UGA was a top-four team. S&P+ had the Dawgs #3, almost 4 points better on a neutral field than #4 Oklahoma. ESPN’s Football Power Index had the Dawgs #3. The Massey Composite, Sagarin ratings, and many Vegas power ratings agreed with the eye test on UGA’s capabilities.

The problem for 2018 UGA and a few others in the Playoff era: the committee has failed to say clearly if it wants the best or most deserving teams.

Technically, the committee works off a protocol that calls for the “best” teams to get in, and the “most deserving” thing is an annual debate that goes nowhere.

“Best” is forward-looking. “Most deserving” is backward-looking. The truth about what the committee wants falls somewhere in the middle, with no precise landing point.

How do we know the committee’s never been sure about this balance? A few big things.

For one, the committee has regularly noted it might take into account injuries when picking teams — i.e., if a team has a great case but just lost its QB, that might hurt its chances. Playoff executive director Bill Hancock has claimed the committee only uses injuries to handicap what a team’s done so far, but if that player’s still hurt, how could it not still factor into a decision? (It’s also best not to put a ton of stock into Hancock’s public statements.)

But it’s the committee’s record of actually picking teams that makes the strongest case here.

Time and again, it’s prioritized clean resumes over teams Vegas would take to win future games.

The Playoff has to reward winning, so this makes at least some sense.

In 2014, the committee took (in order) Alabama, Oregon, Florida State, and Ohio State. The Buckeyes leapfrogged TCU for a spot at the last minute, because they trounced Wisconsin in the Big Ten Championship while the Horned Frogs idled. But the computers generally thought TCU was a better team than not just Ohio State, but also FSU. Keeping out the unbeaten Noles, of course, wasn’t going to happen. But the Frogs sitting at #3 in the Massey Composite, which features dozens of computer ranks, didn’t help them.

In 2015, the committee took Big Ten champion Michigan State instead of Ohio State. The Spartans had beaten the Buckeyes in a (per Spartans standards) weird, hideous game. OSU was #4 in that year’s final S&P+ ranking, and MSU was #14. The Buckeyes were 4th in the Massey on Selection Sunday, and the Spartans were 5th. But the committee valued MSU’s head-to-head win and conference title. No one considered keeping out a one-loss Big Ten champ, even though few mistook the Spartans for an actual top-four team. MSU got shut out in the CFP, and Ohio State won the Fiesta Bowl.

In 2018, UGA missed out despite all the analytics chops. Oklahoma, which had Kyler Murray but also a Pop Warner defense, had to play #1 Bama in the Orange Bowl semifinal and did not come as close as UGA had. This was extremely foreseeable, but the committee didn’t want to take a two-loss non-champ (UGA) over a one-loss team that won the Big 12 (OU). It was a clear sign “most deserving” meant at least as much as “best.”

This conflict between “best” and “most deserving” predates the Playoff. The BCS also likely excluded numerous top-two teams, talent-wise.

Most of the time under the BCS, the two terms wound up being synonymous. But not always. Three times between 1998 and 2013, the two highest-rated teams on the closest thing to a computer consensus (Massey Composite) after Championship Weekend were not the two the BCS (which included the solo Massey, not a consensus ranking) spat into the title game.

In 2002, the Composite had #1 Miami, #2 USC, and #3 Ohio State. The BCS took the Buckeyes, and that was the end of controversy that year! It wasn’t.

In 2006, the Composite had #1 Ohio State, #2 Michigan, and #3 Florida. The BCS either robbed the world or spared it from a rematch of The Game, and Florida beat OSU.

In 2013, the Composite had unbeaten unbeaten FSU #1, Alabama #2, Stanford #3, and Auburn #4. The Tigers were probably not better on the season as a whole than Alabama — the Kick Six is hardly a repeatable feat — but they were still an easy choice after beating the Tide head-to-head. And no one was lining up to make a strong case for Stanford, which had two losses to mediocre teams, even if one of them (USC) crept into the top 25 on Selection Sunday.

Computers don’t always agree with other computers, let alone do human selectors always interpret the word “best” the same way. Anything other than being an unbeaten champ of a power conference (or Notre Dame) is at risk of falling on the wrong side of that fence.

As always, the simplest path to Playoff inclusion is to join a power conference and then not lose games.

Take 2018 Georgia for example again.

If the Dawgs only had one loss on Selection Sunday, things would have been different, even if they lacked a conference championship. The committee could’ve easily slipped the Dawgs into the field at #4, as they did with one-loss Alabama in 2017.

But the Dawgs didn’t only have one loss. They also had a 20-point defeat at the hands of LSU, which went on to lose three games. That put UGA in league with 2016 Penn State (a 49-10 loss to Michigan) and 2017 Ohio State (a 55-24 loss to Iowa) as teams to lose exactly two games, one of them by a massive margin, and then finish #5 on Selection Sunday.

It would be nice if college football had a system that could guarantee the best teams play in the biggest event. But there’s no way to do that without divorcing wins and losses from the formula, and that’d make more people even madder than the best teams missing out. It’s a hazard of a sport with 65 Playoff-eligible teams and 12-game seasons.