*Author’s note: I’ve been informed that I made the lines for ND’s radar charts orange instead of gold as I’d intended. I’m very colorblind, a fact you all now know about me. In the spirit of this column’s ethos “Chase your bliss,” I’m not changing it. I think I got red right for UGA so that’s cool, I guess? (Thanks to my dear friend Austin, who likes me enough to click on my sports articles even though he’s not a sports person.)*
My normal approach to this space - insofar as you can establish a pattern after doing something three times - is to treat it as a place of learning. I generally want to keep everything focused on just one or two concepts, and slowly guide readers through my thought processes. However, as we approach this weekend, and the most-hyped home game at Sanford since at least 2007, I need to slightly deviate from my previous approach. I always want to match my energy level to the occasion, SO BUCKLE THE F*** UP. PUT YOUR GAMEFACE ON. NOW SLAP YOURSELF, DO IT, DO IT. NOW SLAP ME. I SAID SLAP ME. YEAH, NOW WE’RE READY. IT’S TIME FOR: What Advanced Stats Tell Us About UGA’s Matchup Against Notre Dame.
In the past couple of weeks I’ve received a lot of gentle, constructive, and fair criticism of what I’m doing here. The biggest issue readers have is sort of the problem with all statistical analysis, at any level: reams of numbers quickly become overwhelming. This isn’t a roadblock to just layman, as even professional researchers can become lost in the veritable seas of data needed conduct scientifically rigorous studies. To that end, I’ve spent the past couple of days experimenting with visual representations of the concepts I’m talking about, and settled on one popularized by the granddaddy of #sbnanalytics, Bill Connelly: the radar chart.
Radar charts, or spider-web charts, are an efficient way to quickly compare two sets of data across multiple criteria. So, for instance, if you wanted to compare the relative strength of two football teams using the stats that correlate most often to winning, you would make something like this:
So let’s take a moment to orient ourselves with this kind of graph. Each axis measures a different stat based on national ranking. The interior of the graph is 140th ranked, and the outside edge is 1st ranked. So you want your team’s line to be closer to the edge. I’m very colorblind, but I still think I’ve managed to set it up so that each team has its own color (gold for ND, red for UGA). So whenever UGA’s line is outside of Notre Dames line, that’s a positive sign for the good guys.
The image above gives us a comparative look at the game when Notre Dame has the ball. The usefulness of this chart, IMHO, is that it allows you to quickly discern what areas a given team may be successful in. Intuitively, that kind of awareness lets us paint a picture of what the flow of the game will look like. So let’s craft that particular narrative now using the chart above.
At the most shallow level, we can tell from this image that UGA’s defense faces a challenge, but they are certainly not going to be overwhelmed. That’s an observation of no small value because it lets us set the basic parameters of how we think the game will look. When ND has the ball, we can expect them to have some success, but UGA matches up well across the board where the Irish are strong.
On a more granular level, UGA has performed well against passing attacks to this point this year. We can see that there is a significant gap between the SR surrendered by the Bulldog D, and the efficiency attained by ND’s passing offense. That tells us that, if these numbers are accurate, ND should struggle in consistently moving the ball through the air.
On the ground, however, this chart shows us that ND is actually slightly more efficient in running successful plays on the ground than UGA is at preventing them. This does not necessarily tell us that Notre Dame will run wild on the ground, but it should give us a hint that they may have success moving the ball. They will probably not, however, be able to consistently real of 8 - 9 yd. runs. UGA is currently surrendering a far lower Opportunity Rate (the percentage of runs that go for more than 5 yards).
So how is all this helpful? It helps us shape what we expect the rough contours of the game will be when ND has the ball. It doesn’t mean that, for instance, Notre Dame will never have an efficient passing drive. It just means that consistent success from the Irish through the air is aberrant based on our numbers. And if it happens, that’s probably a bad sign for UGA.
So when ND has the ball, we have an outline of how the game may look. Two relatively evenly matched sides, each having some slight to moderate advantages over the other. The picture when UGA has the ball, however, is starkly different.
First of all, let’s just look at UGA’s plot here. Holy s***. The dawgs are formidable on offense to this point in the year. When your radar plot on one side of the ball is basically a regular polygon, you’re doing everything right. There’s been a low-boil narrative developing across Kirby Smart’s career that he, to a fault, wants to “impose his will” or play “man ball.” This has manifested and is driven by many, many predictable offensive calls in short-yardage and goal-to-go situations. I know that this narrative exists because I’ve been one of the main people driving it, along with Senator Blutarsky over at Get the Picture. To this point in the year, however, Coley and Smart have constructed a well-oiled death machine of offense. It’s hard to complain about the performance of a unit when they’re ranked in the top 15 of every major statistical category. That’s “runaway AI that only feeds on human sadness and blood” level of offense, and I, for one, can’t wait to see the violent path they reap on the way to sentience.
As for the matchup, despite my general preference that UGA evolve into a modern, pass-happy offense, this is definitely a moment where the Dawgs would do well to Run The Damn Ball, Bobo. We talked about this on our two preview episodes over at Chapel Bell Curve, but by far the biggest weakness on either side of this ND team is it’s run defense. It’s a genuine mismatch according to basically every data-point we have.
So what kind of narrative can we construct from these numbers? When UGA has the bal, they should be able to impose their will in the running game. Notre Dame’s strong secondary means that not only is this the most convenient strategy, it also moves the action of every play away from the the Irish’s best players, who are concentrated in the secondary. If UGA can have success on standard downs, and run when they want, I have a hard time imagining ND pulling out the win. Those are “if’s,” I’ll grant you, but they’re conditionals based on what we know UGA both wants to do, and can do. And when want and ability line up, success often follows.
So what should we, as Dawg fans, be most concerned about entering this game? I think the answer to that is found in ND situational stats on offense:
One of the very few situations wherein Notre Dame has a decided statistical advantage is on first downs. Not only has there been a large margin in the performance of these two units, the Fighting Celtic People have been a top-ten offense in first-down situations. Most of the the advantages ND has over UGA are in situations where both teams are, at best, average, 1st down SR is the only area where ND is excellent, and there is a significant gap between the two teams. One of the most worrying signs possible in the early stages of this game would be to see consistent success on first down. This would open up the playbook for Ian Book, ND’s senior QB. As more of a scrambler than a pure pocket passer, Book has shined when he can move in the pocket, keep his eyes upfield, and run against zone defenses. The unequivocally best way to put him that situation is for the ND offense to be successful on early downs. If we see that unfold on Saturday, particularly in the first frame of the game, that’s when we need to start worrying.
Additional stats and charts attached below. I’ll catch you in the Classic City this weekend (maybe literally, find me, I’ll be with the band), but until then GO DAWGS!!!!