College Football Playoff: Evaluating Top 25 teams against the spread

A dispassionate way to evaluate a team is how they fare against the spread. Let’s look at how College Football Playoff Top 25 teams perform versus the odds.

With the completion of the College Football Playoff semifinals, we now know that Alabama and Clemson will meet to play for the national championship for the third time in the past four years.e For the fifth year, we will see the season culminate in a plus-one opportunity for the two teams that emerged through the mini-bracket to reach the title game.

The return of the Tigers and the Crimson Tide is hardly any surprise, given the fact that the two teams have been mainstays of the tournament for pretty much its entire history to date. But have they really been deserving of their place on every single occasion, or has this become a self-fulfilling prophecy and a positive feedback loop that just keeps feeding upon itself to grow stronger and stronger?

It is easy enough to say that this team or that team finished with a perfect record or won a given conference. But a truer indicator of how well a team is really outperforming expectation is by looking at how they perform against the point spread set each week by the oddsmakers that look to balance out the betting on each game.

Their entire job is to get the line right so that there is equal money on either side. The object is to let the house collect its vigorish on each transaction and for the losing side to cover the winnings of the other bettors. As a result, they are the impersonal arbiters of the sport that help reveal how the general public perceives each challenger in a given week.

For that reason, we are going to take a deeper look at point spreads for every College Football Playoff Top 25 team dating back to the beginning of the new system in 2014 in this week’s edition of the Sunday Morning Quarterback.

Why evaluate College Football Playoff teams against the spread?

Before we look deeper into the numbers, there are a few things to note about the layout of the dataset and how teams are being evaluated. First and foremost, the odds are hardly an infallible or even entirely objective way to evaluate teams. What they are, however, is a consistent measure of how teams are perceived by those whose only stake in the affair is trying to ensure equal betting on both sides of the line.

In that way the line is a matter of showing where public perception sits between two sides. And it is a sliding scale, meaning that the better teams do, the higher the expectations become in subsequent contests. In that regard, beating the spread becomes a somewhat Sisyphean task. To beat it with consistency and finish with a winning record against that spread is no insignificant thing.

The following dataset offers a look all 125 teams that featured in the College Football Playoff Top 25 since the selection committee started releasing a final regular-season set of rankings in 2014. The first two tabs offer an evaluation, There are also tabs for each of the 125 teams that break down the game-by-game data against the spread compiled from the historical game log data at Odds Shark.

Several columns are especially critical, namely the aggregate and per-game ratings. These are calculated from a combination of each team’s average point spread, their winning percentage against the spread, and their point differential against the spread. Multiplied against one another for the aggregate score, these are then divided by the number of games played by each team to normalize for different schedules and for teams that reach conference championships.

The figures in the table are all from only the results prior to the bowl game, representing the records that the selectors would have seen when evaluating who to include in their final Top 25 rankings of the season.

Looking especially at the rankings sorted by the per-game averages, what we find is that the College Football Playoff selection committee is not always picking teams that prove most dynamic against the weight of expectation.

Reading into how teams perform against the spread

There are several things to take away from how each College Football Playoff Top 25 team has fared against the spread. Let’s look at the three biggest stories in each season:

2014 season

• TCU remains the most overlooked team in College Foottball Playoff history, with the Horned Frogs kept out of the system despite winning a share of tehe Big 12 title.
• Florida State was given too much credence as an undefeated defending BCS champion. The Seminoles were perhaps the lowest-rated team to make the field in Playoff history to date.
• The one team that ranks in the top four in both adjusted performance per game against the spread and in the College Football Playoff Top 25 is Ohio State, which made it back in at No. 4.

2015 season

• Like the 2014 season, the one team that made both the list of top teams against the spread and top teams by the committee’s rankings is No. 4 Oklahoma.
• Similarly, the lowest-ranked of the top four teams that reached the College Football Playoff field in 2015 was No. 3 seed Michigan State, the Big Ten champion that was later blown out by Alabama.
• Houston and Navy ranked second and third in performance against the spread, showing the Group of Five was able to produce a pair of top teams.

2016 season

• Topping the list of team performance against the spread was unbeaten Western Michigan. The Broncos, under P.J. Fleck, went 9-4 against the spread with an average margin of victory against the spread of 8.8 points.
• The next three teams on the list were ranked No. 3, No. 1, and No. 4 in the final College Football Playoff Top 25.
• This year, No. 2 Clemson was the worst-rated of the four playoff teams against the spread. They were 13th against the spread in the per-game average.

2017 season

• No. 1 Clemson and No. 3 Georgia were relative locks for the College Football Playoff, ranking in the top four of the committee rankings and in performance against the spread.
• No. 2 Oklahoma and No. 4 Alabama were less obvious choices to reach the playoff field, as they struggled to put teams away against the spread.
• According to their performance against the spread, undefeated UCF proved themselves worthy of inclusion in the playoff field in relation to how other contenders performed.

2018 season

• This year, Alabama and Clemson stood head and shoulders above the rest of their Power Five counterparts, rating more than a point and a half per game better than next-placed Washington State against the spread.
• UCF, once again, was glossed over by the College Football Playoff despite being one of the top two teams in the country against the spread.
• This year’s Big 12 champion, Oklahoma, was just the second team to make the College Football Playoff (alongside 2014 Florida State) despite finishing with a negative mark in the per-game average.

What should we take away from this? Most importantly, point spreads are not hard and fast rules of how a team should perform but rather a measure of the expectations of how the public things a team might perform.

Teams that make the playoff field are not always the most impressive performers from week to week against the spread. That is obviously not the only measure that the committee brings to the table.

Without adding a strength-of-schedule component to the evaluation, there is a certain skepticism that is warranted when approaching these against-the-spread numbers. But even on their own, they provide a powerful window into how perception can shift over time and how certain teams (such as Alabama) are consistently playing against tougher expectations than others.