top of page

A good coach tends to have good teams, whereas a bad coach tends to have bad teams. And an elite coach generally has elite teams. Of course, that isn't always the case. A bad coach at an good program can find success, while a good coach at a bad program can struggle. So, in order to tell how good a coach is, not only do you have to know how good the team is, but how good the program is (and how good the past coaches were). Controlling for said variables, a value can be calculated for each season that defines its success/failure relative to its history and attributed to the coach (who recruited the players, ran the practices, called the plays, etc). Some mathematical findings things to consider...

  • ~80%, how much of a season's success can be attributed to the program

  • ~67%, how much of a season's success can be predicted by the program's prior season

  •  ~5%,  how much a coach's performance depreciates (in terms of predicting his future value)

Then, a coach's value is the average of his seasons' values - adjusted for sample size - with units of points per 100 possessions better than average. This is useful as you can not only rank coaches, but see how much better your team is with your new coach in place of that guy you fired.

Like coaches, but more so, a good program tends to be good, and a bad program tends to be bad. A program's rating is simply its average effective margin over the years less the ratings of its coach(es).

The hiring value is relative to the quality of the program. It can be calculated by comparing the ratings of a program's coach(es) to one-fourth of the program's value (because 20% of success is the coach and 20/80 = 1/4).

These are the biggest outliers. The season val is the effective margin of the team, while the val+ is the effective margin adjusted for the quality of the program. It should be pretty self-explanatory. The only thing is the season column. That signifies how long the coach had been at the program when the magnificent/disastrous season happened, and whether or not he should be blamed or his predecessor.

bottom of page