Only 13 undefeated teams remain in the FBS after Week 6. SMQ looks at each unbeaten team’s College Football Playoff adjusted margin of victory.
College football offers a platform for endless arguments about both individual and team greatness. One of the most fun debates is determining who really deserves to snag one of the coveted spots among the top quartet. This has become more critical than ever during the College Football Playoff era.
How does one quantify skill level among teams that play disparate schedules, though? No two college football schedules are equal. Even within the same conference, two contenders can play vastly different groups of opponents.
For some, the eyeball test is enough. But this Sunday Morning Quarterback is always inclined to dig a bit deeper into the numbers to try to separate the wheat from the chaff. There are various methodologies by which one can create normalizing models. Let’s try to build a new model today to determine adjusted margin of victory for each of the contenders.
What is adjusted margin of victory?
Every week, teams win and lose by a certain number of points. Looking at those numbers straight up, though, can only tell us so much about a team. That is due to variations in quality of opponent and expectations. In past attempts to normalize the numbers, I have worked out several systems to try to flatten the data across the board.
Some have utilized extant computer rankings to normalize figures. Others have involved more elaborate calculation of offensive and defensive numbers. Combining weekly stats with coaching, recruiting rankings, and margin-of-victory data, I came up with a proprietary composite figure at a previous site.
Each system obviously has its own strengths and weaknesses. No normalization procedure will ever be perfect. Nor will any single system satisfy fans whose teams are left looking in from the outside. But they do provide different ways of looking at the sport of college football that go beyond merely assessing wins and losses and relying on subjective poll figures.
How are we looking at the numbers today?
On this Sunday, let’s take an approach that melds some of these earlier attempts at normalizing margin of victory and give it a bit of a twist. Week after week, oddsmakers at betting establishments across the country provide point spreads for every college football game. Betting odds are also far from perfect. What they offer, though, is a relative understanding of quality gaps between opponents that account for far more than just home-field advantage.
For today’s exercise, I have gone through to determine whether teams beat the point spread in a given week over the first six weeks of the season. The first calculation is simple enough — by how much did a team beat the point spread in a given week? Only two of the 13 undefeated teams left in the FBS have won against the point spread in every game so far this season.
But that too says only so much. Once I calculated the initial figure, I also chose to incorporate the strength of schedule numbers found at Sports Reference as a way of further incorporating a look at opponent quality into the equation. When added to the formula, this provides an additional level of normalization that accounts for opponent quality.
In this week’s Sunday Morning Quarterback, let’s look at the last baker’s dozen standing using this formula to see who is a contender and who is a pretender in the race for a College Football Playoff and/or New Year’s Six berth. Read ahead as we separate the Pretenders, the Sleepers, and the Contenders. (The full dataset is available by clicking here.)