Texas Longhorns football came to an 8-5 close in the 2022 season that left most fans wanting more. In head coach Steve Sarkisian’s second year, there seemed to be many shortcomings: no bowl win, no conference championship appearance, clutch-time issues, and struggling to find the next great Texas quarterback. However, it was a significant improvement from the 5-7 finish from 2021 and Pete Kwiatkowski’s defense seemed to develop as well. And I believe that even with the roster turnover, Longhorn fans can be hopeful for the next couple of years.
Granted earlier this season, I wrote a couple of articles pointing to how misused Texas talent has been in the past decade, with programs like Florida State and Jimbo Fisher’s recent Aggies mess being the only worse instances. But I decided to take a different perspective that utilized the Fremeau Efficiency Index (FEI) ratings. It is “opponent-adjusted possession efficiency data representing the per-possession scoring advantage a team would be expected to have on a neutral field against an average opponent” as per the footballoutsiders.com. With statistics such as EPA/PPA that we have been looking at throughout the season, we must keep strength of schedule in the back of our mind while interpreting them. FEI allows us to gain a better understanding of these teams in the college football landscape regardless of who they played. This first graph compares, total FEI, talent score as per 247Sports, and the number of wins for the 2022 season.
Out of the 131 teams across the nation, Texas is already an apparent anomaly.
But it is all about PERSPECTIVE. If we only take a two-dimensional approach for this graph, neither of the images below would prove that Texas is a stand-out team.
When solely looking at talent versus number of wins, Texas appears as a disappointment yet again. Visualizing plot FEI versus number of wins, Texas can be viewed as unlucky at best. If anything, it would almost uphold the idea of the needed culture shift for the Longhorns. However, from a top-down view, plotting FEI and talent score, Texas does indeed look hopeful.
Combining the insights from all three perspectives tells a story that the Longhorns had what it took to compete this year. Their total FEI ranked sixth this year with a value of .89 behind only Michigan, Tennessee, Alabama, Ohio State, and Georgia (in ascending order).
Another key insight here is that if a random team in the nation were given the Texas schedule this season, they would have been expected to finish 4-9. Even more so, there is only a 15. percent chance that team would have had less than five losses and a three percent chance they would have finished with one loss and been playoff eligible. I am not trying to throw a pity party for the team, but this season needs to be taken with a grain of salt. I agree that Texas needs to be able to win its most difficult games, but the fact remains they had a lot of them.
Looking ahead, I decided to separate the above graph by conferences. Next year, BYU, Cincinnati, Houston, and UCF will join the Big 12, taking the conference to 14 teams. We do not know yet what schedules will look like, whether it will be a random selection of nine teams to play for in-conference schedules or whether divisions will be made like the SEC or Big Ten just for this one year. In 2024, we anticipate Texas and Oklahoma’s move to the SEC and there is another graph comparing how they would have performed relative to those teams this year.
Even if we wish to look at PPA for the year, Texas still ranks in the top quartiles on either side of the ball. The graphs below display offensive and defensive PPA, for both passing and rushing. I have only included labels for teams that will either be in the 2023 Big 12 or 2024 SEC conference for the sake of simplicity. There are guiding lines that showcase percentile as well: red is 20th, blue is 50th (average), and green marks the 80th percentile.
Lastly, I ran a k-means cluster analysis on only teams in the Power Five conferences. Using the elbow method, I determined that the best number of clusters to use was six (k=6) to retain the most significant distinction between them. The model considered net FEI, offensive FEI, defensive FEI, talent composite score, number of wins, explosive rate produced and allowed, success rate produced and allowed, and PPA produced and allowed.
The results prove what we were hoping for. Despite the lack of wins, by every other metric, Texas is grouped into the top echelon of teams. Over a dozen variables went into the model, but with the ability to only visualize three, I chose to display Offensive FEI, Defensive FEI, and number of wins.
So even though, Texas might seem like it doesn’t belong in one visualization, a model combining several of the most popular advanced metrics proves otherwise. Based on last year’s performance, Texas has what it takes to be a top-tier team. Other teams that are in this cluster (same light green color as Texas) include Georgia, Ohio State, Alabama, Tennessee, Michigan, Penn State, LSU, and Clemson.
Texas ranks third in recruiting for the incoming 2023 class, adding No. 1 QB Arch Manning, No. 2 LB Anthony Hill, and No. 1 RB Cedric Baxter to an already insanely deep running back unit. The last time Texas ranked this high was in 2019 when they landed five-star recruits Bru McCoy and Jordan Whittington. Texas has led the Big 12 in recruiting for the past decade years, and this season it seems we finally saw a glimpse of that potential being used. The next step for Texas now is dominating the win column, not just the advanced stats.