EPA model has Texas A&M Football as top SEC team
Despite an early stumble at Miami, Texas A&M football has showcased some impressive play (albeit in fits and starts) throughout the first four weeks of the college football season. Many detractors would claim the eye-catching numbers both the offense and defense have put up are simply a result of inferior competition as their opponents; however, opponent-adjusted metrics seem to show otherwise. Take, for example, this opponent-adjusted EPA model:
Let’s interpret this chart. So first of all, it’s based on EPA/play, or Expected Points Added per Play. This is essentially a statistical model that assigns a point value to each yard line on the football field, considering the down and distance to go for the offense in question that has the ball at that yard line. In this way, it recognizes that a team that has a 2nd and 1 from the opponent’s 45 is much more likely to score than a team that has a 2nd and 28 from the opponent’s 45. The first team is much more likely to convert and continue their drive into the opponent’s territory, which correlates to more points at the end of the drive. There are different EPA models which weight the factors differently , or assign different point values to different yard lines, based on the architect of the model’s perception of what is appropriate for each of these things; however, they all work off of this basic premise. So when you are looking at EPA per play, you are seeing how consistently a team’s offense is creating positive plays, accounting for down, distance, and the explosiveness gained, as well as how consistently a team’s defense is stifling opposing offenses in these same categories. It’s sort of an “all-in-one” statistic to build a model around.
So now let’s look at this specific chart. As you can see, it’s divided up into tiers, indicated by those diagonal lines. The only SEC team in that top tier (as well as the only team from the state of Texas in that top tier) is Texas A&M football. Notably, Miami is one of the top teams as well, which helps explain that week 2 performance.
Obviously, this is not an all-encompassing chart. It doesn’t account for specific positional matchups. You can see that Ole Miss is still ranked as just a tad bit better than Alabama; likely because they stack up better by this one predictive statistic—panoptic though it may be—but fall short in key on-field matchups (such as their offensive line vs. Bama’s pass rush).
In any case, though, it’s encouraging to see that there is such a positive outlook for Texas A&M football moving forward. If Conner is out for an extended period, I wouldn’t be surprised to see a slight offensive regression, of course, but I similarly think the defensive rankings will continue to climb. I’m high on the future of that unit.