Site icon McIllece Sports

2022 Coach Ratings

When it comes to college football coaches, especially assistant coaches, most CFB fans have little objective information or metrics by which to evaluate them. The same is true of sports journalists and athletic directors (search firms often put forth dubious metrics to rate coaches). Head coaches can generally be measured by their overall records and power ratings, but that excludes the impact of nine or ten other full-time staff members.

Despite the ever-spinning coaching carousel, and all the winter articles written about FBS coaching changes, the vast majority of people attached to the sport have little idea about how good any of these coaches really are. Those who do tend to hoard their knowledge for purpose of capitalization, which is okay but does not serve the general college football fanbase well.

To help fill this chasm of analytical coaching information, we are publishing first-of-its-kind, free FBS coach lookup tables that include a battery of objective metrics of coaching performance. These tables are drawn from all FBS coaches, games, and plays from the 2005 – 2021 seasons:

wpDataTable with provided ID not found!

To use the table, simply type in an FBS coach’s name in the Coach search box, or type in his unique CoachID in the CoachID box. (CoachIDs help avoid misspellings and duplicate coach names if you want to come back and repeat a search later.)

Each coach is rated across all units he shares responsibility for, such as QBR (quarterback rush), WRP (wide receiver pass), etc. Responsibility levels are determined by coaching roles, and ratings are measured as the average percentage increase or decrease in game-by-game win probabilities. The final column, standardized wins (StdWin), aggregates all these unit ratings into a single +/- net wins value for the season. This StdWin number estimates how many wins (or losses) the coach contributed that season, adjusted by recruiting level and strength of schedule. Theoretically, if all the coaches on a team had a StdWin = 0.0, then their expected record versus teams with equal talent and coaching would be 6-6 over 12 games, or a .500 record overall. Standardizing by recruiting and strength of schedule puts all coaches on a level playing field.

All metrics are conditional on the team’s 4-year composite recruiting baseline; i.e., a team’s expected power rating (ExpPwr) is compared to their actual power rating (Power) to compute differential power (DiffPwr). This allows the measurements to look beyond the record and get a true sense of how good the team really was. By constructing ratings in this way, all ratings are unbiased for a given recruiting level–otherwise, the system would be unfairly biased toward the big recruiting programs. Averages (μ) are given for each of the unit ratings columns. Note that these averages include empty cells (meaning the coach had no responsibility for that unit) as zeroes, so keep that in mind when considering those summary statistics. Career/total StdWin (∑) is given in the rightmost column.

METRICS LEGEND

ROLE LEGEND

NOTE: 2021 OL/blocking ratings not yet included. All the 2021 unit ratings will change slightly once the blocking data is incorporated into the coaching metrics. Blocking data is incomplete for the 2005 – 2006 seasons. This data may be incorporated at a later date.

 

Exit mobile version