FanPost

What should we now expect from Clark Lea’s first year?

George Walker IV / Tenneseean.com via Imagn Content Services, LLC

Before the start of the 2020 season, I tried to make the case that Derek Mason’s record at Vandy demonstrated that not only was he a cromulent head coach but (importantly), after adjusting for differences in strength of schedule, his best teams were as good as those fielded by James Franklin. Yes, his bad teams were worse, but the data also hinted at the cyclical nature of VU football "success" going back to the Bobby Johnson era, and Mason stuck around long enough to experience both the peaks and the troughs while Franklin did not.

While these findings (and my opinions on this topic) haven’t changed, the circumstances surrounding Vanderbilt football certainly have. After the circus and sideshows that comprised the 2020 season, the new Chancellor and Athletic Director have shown unprecedented commitment to athletics in the form of a $300 million investment aimed at improving both the student-athlete and fan experience. Given these promising attempts to shed the SOV impression and the need to regain buy-in from a fan-base that had all but given up on VU football, I came to terms with the decision to part ways to Coach Mason. Moreover, I was legitimately excited about the possibility of a new era of VU football where the team was coached by an alum who understood the uniquely demanding nature of fielding a competitive football team at Vanderbilt as opposed to one who would attempt to replicate the model for success at Stansbury or similar institution.

However, after yet an offseason of growing optimism and expectations for the upcoming season, we were once again injected with a dose of cold hard reality, this time in the form of a shocking loss to ETSU, which represented the first time that the Dudley ‘Dores have been bested by an FCS opponent. This loss led to frustration at Clark Lea and comparisons to Mason’s 2014 Temple debacle, with many in the commentariat writing off Coach Lea almost as quickly as they had Mason. While there’s no real basis to think that anyone can accurately predict Clark Lea’s potential for success based on one game (despite the FCS opponent), historical data could provide some valuable insights.

So, given the loss against ETSU and dreams of a potential bowl season having gone traveling with Jeff Green, the question for me became, "what should we now realistically expect from Clark Lea’s first year?" I used the data I gathered for my piece last year to shed some light on what reasonable expectations should be for this year’s team.

A couple of disclaimers:

1) I left 2020 out of these analyses since VU played an all-SEC schedule (which will not be the case this year) and let’s face it, no one really wants to acknowledge 2020 happened.

2) I only collected data going back to 2005 because that’s as far back as ESPN’s FPI numbers went back to at the time. Not sure if they have published more now, but I don’t really have the motivation to track these down.

The margin of victory/loss over the past 15 years is most strongly correlated with opponent FPI.

This should be a no-brainer, but over 15 years (2005-2019), the strongest determinant of whether VU football wins or loses a game is how good the opponent happens to be. For this analysis, the measure I used to gauge the strength of the opponent was their end-of-season FPI (or after Week 1 for 2021). Yes, these measures will be partly confounded by the fact that they beat or lost to VU that year, but this should be a relatively small effect. From 2005-2019, there was a statistically significant correlation between FPI and the margin of victory/loss (p<0.01) with R2=0.44:




We should still reasonably expect a 1-2 win season.

Taking the line-of-best-fit from the regression with FPI and assuming average performance compared to teams over the past 15 years, we should expect that they comfortably win games against 2 remaining opponets (Colorado State and UConn) and play toss-ups (predicted margins of win/loss <= 3 points) against Stansbury, the Game Penises, Clanga, Missouri and the Chuggers, which (assuming a 50/50 split) brings us to 4.5 wins:

Opponent

FPI

Predicted Margin

Win/Loss

Colorado State

-12.9

16.588

Win

Stanford

-0.8

3.036

Toss-up

Georgia Bulldogs

20.6

-20.932

Loss

UConn

-28.6

34.172

Win

Florida Gators

15.7

-15.444

Loss

South Carolina Gamecocks

2.6

-0.772

Toss-up

Mississippi State Bulldogs

4.3

-2.676

Toss-up

Missouri Tigers

3.2

-1.444

Toss-up

Kentucky Wildcats

7

-5.7

Loss

Ole Miss Rebels

12.9

-12.308

Loss

Tennessee Volunteers

3.3

-1.556

Toss-up

But given that this team just lost to ETSU, it’s safe to assume they are definitely not average and likely well below that. The closest example we have is the data from 2014 (Mason’s first season during which we almost lost to the FCS opponent), which seemed more appropriate to use in this situation. Using this season alone, there was still a strong (R2=0.47) and significant (p=0.01) relationship between FPI and margin of victory/loss. Based on these data, it appears that it would be more realistic to expect a comfortable win against UConn, a less-comfortable win against Colorado State, and basically no chance against anyone else:

Opponent

FPI

Predicted Margin

Win/Loss

Colorado State

-12.9

4.6918

Win

Stanford

-0.8

-7.9164

Loss

Georgia Bulldogs

20.6

-30.2152

Loss

UConn

-28.6

21.0512

Win

Florida Gators

15.7

-25.1094

Loss

South Carolina Gamecocks

2.6

-11.4592

Loss

Mississippi State Bulldogs

4.3

-13.2306

Loss

Missouri Tigers

3.2

-12.0844

Loss

Kentucky Wildcats

7

-16.044

Loss

Ole Miss Rebels

12.9

-22.1918

Loss

Tennessee Volunteers

3.3

-12.1886

Loss

This is not what most of us (at least in the sunshine pumping camp) anticipated this time last week, but I found this slightly reassuring that, despite the debacle from last week, a 0-win season still seems unlikely thanks in large part to a ridiculously (and Franklin era-esque) soft non-conference schedule:

It will be interesting to see how these predictions change as FPI numbers change over the course of the season, but since the non-conference opponents are front-loaded, we will probably know soon enough anyway.

FanPosts are most often submitted by users. The views and opinions expressed in FanPosts do not necessarily reflect the views and opinions held by the editorial staff of The Anchor of Gold or SB Nation. Unless they are awesome.