What Advanced Stats Tell Us About UGA After Week 5

Ok, so you might be wondering to yourself, “Why did Nathan take the week off after the biggest game of the season?” First, I want to assure you that the rumors of my death and ascension to the supernatural realm are greatly exaggerated. Ghosts can’t work with statistical scripting languages, that’s a known fact. Sure, I could tell you that this is the side-hustle to my third side-hustle, or that I had a high school band competition to prep about 75 teenagers for, or that I had 3 classes worth of essays to grade. But those are excuses, and the honest answer is I was just tired and I want to stay married. ¯\_(ツ)_/¯ So, I’ve come bearing gifts today as an offer of supplication to all of you. I have charts, graphs, and (hopefully) the correct RGB color codes for Georgia Red. So let’s get down to What Advanced Stats Tell Us About UGA After Week 5.

In past columns and podcasts, I’ve gone on at length about the difficulty of determining a teams quality given a small sample size. Now that we sit here, with a little under a quarter of the data we’ll have for the season, I think it’s appropriate to focus on a different stumbling stone to our understanding: human bias.

Now, when I say bias, I don’t mean the “Bama ain’t played nobody PAAAAWWWWLL!” kind. Nor am I talking about the, “Kirk Herbstreit secretly hates UGA” (true) kind. Rather, I’m using bias here to refer to its least partisan meaning: favor towards one interpretation of events. Really, the bias I’m talking about here has less to do with favor towards interpretations, and more to do with the way humans perceive events in general.

Take, for example, the Notre Dame game. UGA won in closer than expected fashion, and while their were a few spectacular plays, the bulldog offense felt stilted and uncoordinated for most of the day. This has lead many red and black faithful to question James Coley’s abilities as a coach (as of this writing, I can’t find anyone who questions his abilities as a fashion icon, because tip-to-toe bae got the look on lock.)

My thoughts on Coach Coley’s prodigious fashion talents notwithstanding, the doubt cast on the UGA offense over the past two weeks, in my eyes, seems a little bit silly. Certainly, I’ve been on record decrying the UGA coaching staff’s (*cough cough* Kirby’s) proclivity towards conservatism, but this is an example where the aesthetics of what we’re seeing with our eyes simply do not match the efficacy demonstrated by my numbers. Let’s check out the UGA offense’s radar chart as of week 5:

Nathan Lawrence

UGA’s offense is, unequivocally, in the top ten nationally. By pretty much any metric, this unit has been efficient, explosive, and deadly in nigh-on any situation. They are equally effective running and passing the ball, and even their least superlative top-line stat, havoc rate allowed, is still in the top 15. And when we start to break down these numbers with a little bit more granularity, the Dawgs offense is still far from offensive:

Nathan Lawrence

UGA runs a successful play a staggering 65% of the time that they possess the ball inside of their opponents 40, a rate that puts them 2nd in the nation. Further, the red and black score a TD 75% of the time they cross the opponent’s 20, which puts them squarely in the top 20 (15th) in that category.

These graphs profile an offense that, against even above average defenses, has been able to score at will from most anywhere on the field. So far as I can tell, the word “will” is the key modifier in the previous sentence. Despite entire quarters of - what to my eyes seemed to be - lackluster play calling and uninspired game-planning, this offense has been able to score when it needed to in every game this season. Even their performance against Notre Dame has not significantly impacted the UGA offense’s season-long statistics in comparison to their peers. While it remains to be seen what this unit can do against a truly elite collection of defensive play makers, there is very little statistical support for the idea that UGA’s offense is regressing under Coley’s leadership.

Instead of fretting about style points, I think the real worry for Dawg fans should be the offense’s performance when they must score on several consecutive possessions (i.e. when playing Alabama, Auburn, or Clemson). I don’t worry about the talent or the playbook, but rather the staff has the aforementioned will to win a shootout. In the Kirby Smart era, UGA has thrived in games where it could dictate the pace of play. This generally means abusing the opponent’s defensive line with a deep RB core, and passing the ball when the rushing attack has trained LB’s and safeties to come into the box. But when the Dawgs have either not been able to run with impunity (LSU ‘18) or have been drawn into a shootout by a high-powered offense (Alabama every time we’ve played them in the last decade), UGA has been unable to close out games. The landscape of college football today dictates that any team with title aspirations will find themselves in one of these aforementioned shootouts. Is Kirby’s “Manball or Death” philosophy a clever attempt to exploit a market inefficiency caused by smaller, more agile front sevens? Or is this weakness the last vestige of mid-2000’s Alabama thinking still clinging to our beloved head coach? I don’t know, but it’s the one question I can’t find numbers to answer, and that bothers me.

Anyway, the defense is also pretty good (at everything except for havoc rate):

Nathan Lawrence

Look, I was the first one swooning when Kirby Smart started talking about an advanced statistic this summer. “ Havoc rate?! Oh, I have uh case of the vahpahs!” But UGA’s defensive profile supports the axiom that many successful defenses are only average at creating havoc. When your Success Rate surrendered and YPP surrendered are both inside the top 15, you’ve got a pretty good defense, regardless of any other statistic.

I wonder if the emphasis on havoc rate this year from the coaches has more to do with bringing that side of the defense up to just “Average” from “Pretty bad” last year. In team building, it’s often true that replacing a bad starter with an average one yields better results than replacing a good starter with a great one. I’m starting to suspect that Kirby’s focus on havoc was the statistical version of bringing in a 3 star kicker to replace a 1 star one. (A lesson Alabama could still learn.)

To this point on the season, Georgia’s D has been great but not elite. In the context of the burgeoning youth movement throughout the the unit, and in particular the emergence of blue-chip freshman like Nakobe Dean and Nolan Smith, that is enough to put UGA squarely in the CFP conversation. Ultimately, I believe the continued emergence of these players, and the trust that UGA’s coaching staff feels they can put in them, will determine the ceiling for this defense, and maybe the team as whole. Recent history has taught us that teams with one truly elite side of the ball can reach the CFP, but not that they can consistently win it with a glaring weakness. The roll the Bulldog defense plays in that equation remains to be seen.

More stats and such are attached below for the truly tortured souls among you. I’ll catch you in the Classic City this week, and in Knoxvegas on the weekend. But until then, as always, GO DAWGS!!!

Nathan Lawrence

Nathan Lawrence