Understanding the Degree of Florida's Underperformance in 2013, even after Accounting for Injuries

I wanted to analyze the degree of underperformance in 2013 relative to the talent we have on the roster.

We seem to be understating the degree of how bad this season is, and merely lumping it in with other bad seasons. Probably because we are overstating the impact of injuries.

I did some simple research that hopefully highlights the degree of crap that 2013 is, and that a good chunk of that is attributable to coaching (and not injuries).

Some basic premises:

1. Recruiting is highly correlated to actual results.

Its not a rock solid correlation, but generally a pretty strong relationship. I haven't gotten to the data yet but as an example Bama was the #1 recruited school in the study, they are 1st in the BCS, Auburn 2nd-4th in recruiting and 4th in the BCS, Ohio State 2nd-4th and 3rd, FSU 6th and 2nd, and on the flip side Massachusetts 123rd in recruiting and 119th, Army 114th and 108th, AUB 111th and 106th, etc.

2. True busts should be evenly distributed.

Holding player development constant there is no logical reason why 1 school should have more busts than another school.

The data:

123 teams in the sample.

I took the last 4 years of scout class rankings (pessimistic on the Gators I may add), took the average and re-ranked. This is how I determined "talent" on a particular team. The top 10 are Bama, Auburn, LSU, Ohio State, Texas, FSU, Florida, Oklahoma, Georgia, and Michigan. Auburn through Texas are actually tied. Florida is 7th here, but if you used a difference service with the same method Florida skyrockets to the consensus #2 most talented team.

I took the re-ranked sample and subtracted out the current BCS rankings, and got the difference between what the talent suggests and the actual production on the field. Florida is currently 71st in the BCS rankings, which may be a bit optimistic right now.

After some statistical rejiggering we can truly get to the degree of over/underperformance relative to talent.

If we took the entire sample (123 teams): we are underachieving by 2.25 standard deviations relative to what our talent suggests.

Only 3 other teams have bigger performance differentials than that. NIU and Fresno on the positive, and Cal on the negative.

If we only took the Top 50 in "talent" the data gets a tick worse. We want to reduce the sample because we want to get to our true peers and frankly the teams with worse recruiting/bcs end up having wider differentials simply by sheer math. Plus logically there is a bigger difference between the #1 and #25 than between #98 and #123. So what happens if we do that? Our degree of underperformance drops to nearly 2.5 standard deviations. Cal is still worse than us in the sample, so we have the 2nd biggest extreme value. Tennessee, Kentucky, and Arkansas all have recruited relatively well and are major underperformers here.

What happens if we restrict it to the top 25 talented teams? After all we should be judged more on how we compare to FSU, Bama, Ohio State, ND, Oregon, Clemson, Miami, etc than vs Rutgers, Kentucky, Virginia, Maryland, etc. Our degree of underperformance skyrockets to about 3.25 standard deviations away from what we'd expect.

What this is telling us that we are underachieving relative to our talent in an extremely statistically significant way. Yes it is talent "on paper", but "on paper" talent is translating to somewhat predictable results for every other team. No reason, other than player development/coaching, should our "on paper" talent produce a result less predictable than other teams.

Okay, one reason. Injuries. Fair enough. Lets drop Florida from the #7th most talented team to the 25th most talented team. It seems reasonable to me. That means Florida is now less talented than TAMU, South Carolina, Miami, Oklahoma St, Mississippi, yet slightly more talented than North Carolina, Mississippi State, Pittsburgh, Texas Tech, and Rutgers. Given how many players we will have playing on Sundays, even with our injuries, I think that is a pretty fair slot. Let us not forget Skyler was a top 20 QB recruit, only a couple slots below Maty Mauk.

Making that adjustment for injuries our degrees of underperformance change to 1.6 standard deviations for the whole sample, 1.75 for the top 50, and 2.1 for the top 25. Still those are some staggering statistically significant degrees of underperformance. Out of the top 25 'talented' teams this adjustment would still leave us as the worst performing team relative to talent, tied with Tennessee.

What does this mean:

1. Florida is underachieving its talent by an extremely significantly significant amount.

2. Even accounting for injuries Florida is still underachieving by a significant amount. You'd expect more in line with a 35th ranked team vs a 71st ranked team

So when someone cites injuries as an excuse for our poor play please kindly tell them that while injuries are a portion of the poor play, the vast majority of it is attributable to coaching. The math bears that out.

This study isn't meant to be perfect. Using all the teams doesn't make tons of sense because logically Florida and Florida Atlantic aren't true peers. However, the more your restrict the sample the more skew enters the picture. The more talented you are the easier it becomes to underachieve than overachieve (IE Bama at #1 most talented couldn't overachieve).

Please be kind and use good grammar.

Log In Sign Up

Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

By becoming a registered user, you are also agreeing to our Terms and confirming that you have read our Privacy Policy.

Join Alligator Army

You must be a member of Alligator Army to participate.

We have our own Community Guidelines at Alligator Army. You should read them.

Join Alligator Army

You must be a member of Alligator Army to participate.

We have our own Community Guidelines at Alligator Army. You should read them.




Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.