After this year's bowl games, the narrative that developed among the sports media was the Big Ten, after what seemed like a historically strong regular season, proved to be weak after a terrible bowl performance. But how bad was that bowl performance?
To find out, I compared the actual performance of the conferences to a set of 10,000 simulations of this year's bowl games, excluding this Monday's National Championship Game. To create the simulations, I used log5 and Pythagorean expectations, as I did previously for the NFL and MLB. However, because of the variability in strength of schedule in college football, using actual points scored for the Pythagorean expectations would be inaccurate. Instead, I'm using the Football Outsider's S&P+ ratings of offense and defense (referenced on 1/4/2017), which adjust for strength of schedule and account for efficiency as well as performance.
To rate the performance of each conference, I found the percentile of the conference's wins in reality in my set of simulations. The results make me feel OK Purdue did not participate in making this happen.
Yup, that was bad. Really bad. As one would expect, the expected wins were vaguely Gaussian; and with an average of 5.35 wins, 3 wins is a disappointment. A disappointment to the tune of being in the 3% percentile of simulations. Of course, four of those losses resulted from upsets: Utah (who despite having a better record was considered the underdog based on the rankings I used) over Bloomington Junior College Indiana, Boston College over Maryland, USC over Penn State, and Florida State over Michigan.
On the flip side, the ACC had an incredible bowl season, with their 8 wins being in the 91% percentile of simulations. With the distribution of wins appeared much like the Big Ten's, they managed to be be a mirror image. Some was due to head-to-heads with the Big Ten (Clemson was favored over Ohio State and won, Florida State upset Michigan, and BC upset Maryland), but down the line against every conference, the ACC out performed expectations.
With a rather middling performance this season was the conference with the most arrogant and annoying fans SEC. Their 6 wins was slightly below the simulation average of 6.5032, putting them in their 27% percentile, although still not a poor performance. Just about the average performance, and not quite proving anything, despite what some hot takes may claim. Especially in terms of comparisons to the Big Ten. None of the Big Ten's upsets came at the hands of the SEC, and in the match ups of the two conferences, the SEC team was the favorite: Florida over Iowa and Tennessee over Nebraska.
This bowl season had a lot of great games (Virginia Tech had an incredible upset of Arkansas, the Rose Bowl was one of the greatest games I've ever seen), and one big winner: the Atlantic Coast Conference. Even if Clemson loses to Alabama (my model is giving Alabama a 87.97% chance of winning, although my guess is when comparing very high winning percentage teams, log5 becomes inaccurate due to an 100% winning percentage team always having a log5 of 100%), it is still one heck of a bowl season for the ACC, and not exactly proving anything about the SEC (Alabama is a great team, but the conference as a whole acted as expected). If the Tigers do pull the upset, the year will be filled by hot takes about the ACC's new found greatness, which would not be totally inaccurate.
Comments
Post a Comment