Breaking down CFP qualifiers statistically Before I break this down, I need to answer some inevitable questions up front.
As a proxy for CFP qualifiers, which we only have since 2014, I’m using the Final AP polls. I knew there may be a few differences, but it allows an apples to apples comparison over multiple years.
These scoring averages only include conference games. By doing that I can eliminate the Alabama vs Mercer cupcake games.
In some cases, I used an average of averages. Mathematically, this is a bad idea, but for a couple of reasons I have confidence in the answers nonetheless. For instance, each team plays pretty much the same number of games, so any weighting problems are minimized.
Okay, with that out of the way, let’s look at a statistical breakdown of college football play contenders and how Scott Frost’s teams compare to them.
BLUF:
Frost’s offenses at UCF and Oregon are competitive with the offensive performance of CFP contenders but his defenses at UCF as well behind where they need to be to be competitive for the college football playoff.
The average scoring offense for a CFP contender is 38.8 ppg and the average scoring defense is 18.4 ppg.
Frost’s offenses averaged 41 ppg and his defenses at UCF averaged 25.6 ppg.
Frost’s performance as a head coach is both limited and kind of a mixed performance bag.
His offense made exceptional improvements from 2016 to 2017 while his defense actually regressed, giving up just less than a field goal more per game in 2017 than in 2016. Getting very granular into the defensive performance in 2016, his defense game up 25.5 ppg in October 2016 but only 23.0 ppg in November. This may be evidence his teams were learning and improving. Or maybe not. It’s a small sample size.
Contrast that to Nebraska in 2017, which was learning a new defense under Coach Bob Diaco. In October 2017 Nebraska’s scoring defense averaged 39.3 ppg but jumped to 49.3 in November. I understand that Nebraska’s November schedule was more difficult than it’s October conference schedule but this is a significant regression regardless.
There is some variance in contender’s performance in scoring offense and defense.
The blue box covers the middle 50% of scoring offenses and the orange box is the middle 50% of scoring defenses. The top and bottom bars illustrate the extremes of each data sub-set. Frost’s offense and defense are shown by the red footballs. His offense is in the upper middle of CFP contenders, where as his defense, while not completely off the graph, is approaching the upper limit of contenders.
Seen as a scatterplot, there is a clear relationship between scoring offense and defense.
The dotted line is the regression trend line. Frost’s defenses at UCF are above the line, meaning they are probably giving up too many points for the number of points they score. The further above the line, the bigger the problem. Only four of the 40 teams in this analysis have a greater variance from the prediction line.
This final chart might indicate that things are moving in a positive direction for Frost. It’s a busy chart, but focus on the dotted trendlines. The scoring offenses show a slight increase over the 10 years, but there has been a larger increase in average scoring defense.
So what does this all mean?
Frost has only 2 years as a head coach and 3 years as an OC on which to compare his to CFB contenders, so we have to be careful drawing too many conclusions. Still, I think it’s fair to state that Frost’s defenses have not been competitive for entry into the CFP.
General trends reflect a movement toward higher scoring defenses, but there is still a significant gap there. Offensively, Frost is where he needs to be when he has the players he needs to execute his system and game plans. Granular analysis shows an ability to teach and motivate teams to get better over the course of a year, and this alone may be the most significant insight I found.
To answer the question of how far away is an Frost-coached NU team from the CFP…quite a ways. There is some evidence that it is easier to improve and offense is quickly than to improve a defense quickly. The challenge, then, for Frost is to improve NU’s already pretty-good offense enough to outscore enough Big Ten opponents to keep the recruiting momentum going.
GBR

Source: Corn Nation