If you have read ‘About the CFBMatrix’, you already know it started with the idea of using recruiting ranking to establish trends and predictability in college football teams and games. With I have shown that our national composite recruiting rankings can be directly correlated with over 68% of all CFB games. The rest, as was discovered, was a tied into coaching ability and scheduling. There are multiple other ‘little things’ that are part of a team’s success or lack thereof, but those a proven to be isolated and not correlated to football outcomes as a whole.
In this page, I want to breakdown the net coaching effect for each SEC coach in total, at home and away. While these numbers do not reflect the true ‘value’ or ability of a coach, it certainly provides a comparison and perspective any other ‘ranking’ article lacks. These are not my opinions or ‘expertise’ on the subject, as they are simple numbers provided to add to the information gathering for every fan. They may confirm your emotions about a coach, open your eyes to a new angle of interpretation or just make you more angry.
The biggest caveat to these numbers are chances and opportunity. For teams like LSU or Alabama and Georgia, they have very few chances in a season to beat someone that the model says should beat them and therefore, negative game effects are more likely. For teams like Vanderbilt, nearly every game is a chance to beat someone that they are supposed to lose to in the model. A team that recruits poorly and doesn’t upset many teams is either really bad and/or really poorly coached. The other part of your evaluation of the coaches should be their recruiting. As I always say, you recruit to win, not win to recruit. You need to be able to maximize recruiting and then your wins. A coach may be good on game day with less talent, but without recruiting they are never going to move up too far.
The most interesting facet to the chart is the clustering of teams. Unlike some other conferences, the modeling has done very well in determining wins and losses for SEC schools. The number I focus on is the Total Annual Average. This is the annual value a coach brought after recruiting and home/away advantages were factored into their predictions. For all but just 2 teams, the model was +/-1 or less in predicting games won or lost by each team. It makes the hugely positive number of Petrino and the negative number by Richt stand out even more.
| Team | Coach | Coaching Effect (games) | Coaching Effect Home (games) | Coaching Effect Road (games) | Coaching Years (4Yr Max) | Total Annual Average | Home Ave/Yr | Away Ave/Yr | Recruiting Class Rank ’08-’11 |
| Arkansas | Bobby Petrino | +7 | +3 | +4 | 3 | 2.33 | +1 | +1.25 | 25 |
| Kentucky | Joker Phillips | +1 | +1 | 0 | 1 | +1 | +1 | 0 | 45 |
| Mississippi | Houston Nutt | +1 | -1 | +2 | 3 | .333 | -.33 | +.67 | 20 |
| Alabama | Nick Saban | +1 | -1 | +2 | 4 | +.25 | -.25 | +.5 | 1 |
| South Carolina | Steve Spurrier | +1 | -1 | +2 | 4 | +.25 | -.25 | +.5 | 19 |
| Mississippi State | Dan Mullen | 0 | 0 | +2 | 2 | 0 | -.5 | +.5 | 30 |
| Auburn | Gene Chizik | -1 | -2 | +1 | 2 | -.5 | -1 | +.5 | 10 |
| LSU | Les Miles | -2 | -3 | +1 | 4 | -.5 | -.75 | +.25 | 4 |
| Tennessee | Derek Dooley | -1 | -1 | 0 | 1 | -1 | -1 | 0 | 13 |
| Georgia | Mark Richt | -6 | -5 | -1 | 4 | -1.5 | -1.25 | -.25 | 6 |
| Vanderbilt | James Franklin | NA | NA | NA | New Coach | – | – | – | 61 |
| Florida | Will Muschamp | NA | NA | NA | New Coach | – | – | – | 8 |
The College Football Matrix 