Has Dan Mullen truly done more with less as a HC? A statistical breakdown.

A common belief is that Dan Mullen will do well at Florida because he did well at Mississippi State. With MSU having considerably less resources than Florida, and assumedly, it is much harder to attract top talent at MSU than at UF. It is quite logical to think that if Mullen did well at MSU with a talent deficit, then he should do even better at UF where that deficit should be considerably less. So, did CDM do more with less at MSU? In taking a look at the data, it is clear that he did.

To conduct this study, I took the time period that Mullen was at MSU (2009-2017) and gathered the talent levels for all of his teams based upon a 4-year Rivals rating average. For example, the 2009 MSU team had a talent rating average of 83.74. This rating was obtained from the average Rivals class rating for 2006, 2007,2008, and 2009. This represents the four classes that could have participated in the 2009 campaign. I then took the 2009 win percentage for that MSU team (41%). I repeated this process for each year Mullen was at MSU. Then I did the same for every other SEC team* for the same period. This way, I had the same data for Mullen’s peers in the SEC. Then each score was standardized. This is labeled ‘z-win’ and ‘z-talent’. I then subtracted the z-talent score from the z-win score to determine the talent deficit against winning percentage.

*= includes Texas A&M and Missouri, even though they weren’t in the SEC the entire range of years.

Here is how it shakes out:

Team Win Talent Z wins Z talent Effectiveness
MSU 0.60 85.11 -0.06 -0.89 0.83
Mizzou 0.60 85.44 -0.03 -0.77 0.73
Alabama 0.90 92.00 2.37 1.70 0.67
USCe 0.63 86.78 0.20 -0.26 0.46
Arkansas 0.54 85.56 -0.51 -0.72 0.21
UK 0.41 83.21 -1.57 -1.61 0.03
LSU 0.75 90.47 1.14 1.13 0.01
Auburn 0.66 88.56 0.38 0.41 -0.03
aTm 0.62 87.92 0.13 0.17 -0.04
Vandy 0.41 83.47 -1.57 -1.51 -0.07
Ole Miss 0.53 86.68 -0.61 -0.30 -0.31
UGA 0.70 90.45 0.76 1.12 -0.36
Florida 0.63 90.56 0.22 1.16 -0.94
Tenn 0.50 88.47 -0.82 0.37 -1.20

 

The larger the positive number, the more a team accomplished in terms of wins versus talent level. The lower the number, the more underachieving a team is. Not great, Tennessee and Florida.

It might be a surprise to see Alabama at the ‘overachieving’ end of things. However, they played some very talented teams in this period, and they typically beat them. Take a look at their ‘z-wins’ score. This means they are 2.37 standard deviations above the SEC average in winning, but “only” 1.7 standard deviations above the SEC average in talent rating. Over this period of time, Florida, LSU, and Georgia have all had very talented teams, but did not produce anywhere near Alabama. This pushes Alabama’s score up.

Back to the point. MSU under Dan Mullen overachieved the most in the SEC, relative to other SEC teams, during the period of 2009-2017. His teams were almost one full standard deviation below the SEC mean in terms of talent, but they won at about the mean level for the conference over that time. Had Mullen had Florida’s talent over that period, he would have been 1.99 standard deviations above the mean for wins, which would have him winning 87% of his games, or 10.44 games per year. So, yes, Mullen did more with less. More so than any other coach in the SEC during the same time period.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s