Something to consider when evaluating coaches: it's not just about the record, but about how the record was made. Even for good coaches, improvement isn't always linear as reflected in the overall record (ie, 6-6, 9-4, 11-1, etc.). If we look at how many close games the team played, the record against ranked opponents, and the number of blowouts (wins and losses), that should take some of the noise/luck factor out of records.
For example, let's compare the best individual seasons for three head coaching candidates that have been discussed here:
- Coach A: overall record 10-2
- against ranked opponents: 1-1
- in one possession games: 3-1
- wins by more than two touchdowns: 6
- losses by more than two touchdowns: 1
- Coach B: overall record 9-3
- against ranked opponents: 3-1
- in one possession games: 4-2
- wins by more than two touchdowns: 5
- losses by more than two touchdowns: 1
- Coach C: overall record 10-4
- against ranked opponents: 5-2
- in one possession games: 7-3
- wins by more than two touchdowns: 2
- losses by more than two touchdowns: 0
Coach A blew out half of his opponents and only played four close games, winning three.
Coach B played more close games, probably because he played four ranked opponents to Coach A's two, but he blew out his remaining opponents.
Coach C played almost every game on his schedule within one possession. He played ton of ranked opponents, but played even his other opponents close, going 7-3.
Let's assume for the sake of argument that winning close games is replicable skill, instead of luck. Do we prefer playing close games or blowing out opponents? I would think that by year 3 or 4, we want to be blowing out lesser opponents like Vandy, South Carolina, and Mizzou.
If you go through coaching records game by game, I think it's easier to distinguish the mirages (Coach C, aka Tom Herman) from the actual program builders (Coach B, aka Matt Campbell and Coach A, aka Lane Kiffin).
View attachment 342946View attachment 342947View attachment 343068