Which coaches have added the most value to their Power Five programs over the past three years, ADJUSTED FOR RECRUITING?
Have a look:
Have a look:
Observations:
1) Not shocking that Swinney comes out on top. Clemson has outperformed its on paper (i.e. per recruiting rankings) each of the last three years.
1) Not shocking that Swinney comes out on top. Clemson has outperformed its on paper (i.e. per recruiting rankings) each of the last three years.
2) Chryst should probably have an asterisk next to his name. He is definitely doing a good job, but both Gary Andersen and Brett Bielema had similar scores in Madison. And after they left, they didn't. So clearly Wisconsin's program is more the star here than they head coach.
3) Petersen took two teams with statistically average P5 talent (on paper) to the CFB Playoff and to a New Year's 6 Bowl game.
4) Saban has the best talent in the country and yet still manages to get it to play better than could reasonably be expected. When you consider that and factor in the fact he is the guy who put the structure in place to GET that talent, it confirms his all-time greatness.
Gundy, Ferentz, Patterson, Whittingham, Snyder, and Leach are all long-term "doing more with less" guys. One wonders what they could do with more talented rosters.
Franklin stands out as a guy who is both overachieving with the talent on hand AND significantly upgrading the talent at the same time, much as Saban and Meyer did years ago.
If you want to read the nitty gritty of how this analysis works, have a look at the article I wrote on it five years ago. Some of the inputs have changed slightly, but the concept is identical.
footballstudyhall.com/2013/8/19/4635…
footballstudyhall.com/2013/8/19/4635…
Final thing: That headline on the article always bothered me because it says "Apparent Quality" and the idea was to compare objective quality on the field to "Apparent" talent (i.e. recruiting rankings).
For the real stats nerds: I used a simple linear regression model here, even though a mixed effects model or machine learning algo would probably be more appropriate. Going to update it with the latter at some stage, but doubt the results change too much.
If you are interested in the coaches that have been standing out on the opposite end of the spectrum, here is a link to the thread on the bottom tier performers:
