Graphing the SEC, 2023 Week 1

In this season’s first Graphing the Tide yesterday, I mentioned a new format I’ll be trying out for this season: Graphing the SEC. In this column, I’ll gather all of the SEC team’s games that week, break out their topline efficiency (SR) and explosiveness numbers (XR) into tables, and sort those from best to worst.

Some data definitions

Success Rate (SR) is the rate of an offense’s snaps that can be considered “successful plays” according to a metric often used in football stats circles. It’s the same one we use in the Graphing the Tide series, and you can see the breakout in my primer from last season:

These are fancy metrics, based on simple foundations that inform Bill Connelly’s SP+ system. The primary metric is Success Rate (SR), which judges each play by how much it gained vs. how much yardage was needed for a first down or score. Offenses tend to do better when they gain 50% of needed yardage on 1st down, 70% on 2nd down, and 100% on 3rd and 4th down. There’s some simplification involved, but Connelly did a lot of analysis to develop the model and I find it informative.

For success rate, league averages tend to hover around 42-44%, so keep that in mind when judging these rates.

Explosiveness (Xplos) is a little different than in Graphing the Tide, in that here we’re looking at a version of isoPPP (Isolated points per play) instead. It’s a bit complicated to conjure up, and the point value looks a little abstract, but basically “higher is better,” and ranges tend to fall between 1 and 2. (Closer to or below 1 is bad, closer to 2 is very good). This is often calculated from an entire game’s worth of stats, given the risk of noisy data from small samples, which is one reason I don’t use it in our usual Graphing game breakdowns.

Now, some disclaimers

In this SEC series, I’m pulling data from an external source instead of calculating these numbers from scratch myself. That means a few things:

• I can’t vouch for specifics; though I trust they’re generally accurate from having tested against my own numbers.
• I’m not positive how they’re calculated; though they seem to use conventional methods and get to reasonable results.
• I believe these metrics exclude garbage time; which is really nice if that’s the case. I don’t do this in my usual series because you get less data and it’s complicated to apply.

To all of those points, Alabama’s SRs may look different than the ones I calculate in our GTT series, but that’s probably OK given the garbage time exclusion and potential nuances in calculation.

Overall, the numbers work just fine for comparison purposes between the teams. Let’s get into it!

So that’s fun! Week 1 is full of cupcake games, but it’s still an interesting glimpse into what these conference teams might be in 2023:

• Ole Miss leads the pack far and away. That 68% SR is absolutely sky high, and is a full 9 points ahead of the other leaders (Alabama and Tennessee). It’s probably some mix of a genuinely sharp offense ... and a coaching staff that doesn’t know when to let up vs. an obvious cupcake opponent.
• Alabama is more than respectable, with a 59% SR from a big win. That said, the Tide fell to #3 here because the Vols had the same SR, but held their opponent — a “technically Power 5” Virginia squad — to a much lower SR. I wonder to what degree that’s the Vols putting up a “real game plan” vs. a P5 opponent, but it’s a little disconcerting regardless.
• The SEC cupcake-munchers rank high, of course, while those with “real” opponents — LSU, Florida, and South Carolina — sink to the bottom of the list.
• Yes, some teams show up twice because of week 0. That won’t be the case next week.
• LSU actually put up a league-average efficiency, and held their opponent to only a 5% point SR advantage, despite getting pancaked in the 2nd half by the Noles. (For those of you that didn’t watch the end, this was basically a 17-45 loss, as that last long LSU TD was very well into garbage time against some backups in the secondary). This result was hilarious, but it might be fair to say LSU played better than the scoreboard suggests.
• Florida and South Carolina were bad, which is really too bad considering that Georgia will apparently have an even easier schedule than usual; we won’t find out their mettle until maybe late November (at Vols).
• Speaking of Georgia, they won big and held their opponent’s offense to a pathetic SR. But the Dawgs had a pretty middling Success Rate to show for it. New QB, eh?
• Back to that Florida Utah game ... Utah actually lost on efficiency by a few points, and neither team really did much on offense. But the Utes won by multiple scores. How strange. Reminds me of Iowa’s games.

And then let’s take a look at the explosiveness table: I separated these out to make it easier to read each. In this one, we’re ranking by Explosiveness (isoPPP).

Some notes on explosiveness:

• Vandy rising to the top is hilarious, but of course it’s vs. Hawaii ... games vs. the Rainbow Warriors always seem to turn into circus shows. That said, 1.6 is a very high isoPPP and a surprising result from the ‘Dores
• A&M and LSU posted up high ones too, which is something to keep an eye on (especially given that LSU’s was against an apparently legit P5 contender).
• Arkansas and Kentucky posted high numbers too, but I haven’t heard even a whiff about either game so I don’t know what to make of it.
• Alabama fell pretty far on this list, despite the long TD passes we were celebrating yesterday. I guess when you recall the game, the running game wasn’t particularly explosive and the passing game wasn’t either outside of a few bombs.
• Funny enough, Ole Miss is down there with us. You’d usually expect Ole Miss to have lower efficiencies but high explosiveness, but the Kiffin version so far in 2023 is the opposite.
• And in another surprising flip, the Vols are literally at the bottom of the list. They who put up such high efficiencies against Virginia apparently weren’t throwing ‘em long like they did last year. I wonder if the change in QB will inform a very different philosophy for them this season. I would say “one has to hope” after the result last year, but really you’d hope that they weren’t explosive or efficient.
• Auburn wasn’t explosive at all either. I didn’t watch their game and wonder what is going on with their QB situation. It’s been consistently shoddy for, what, 16 years or so now?

Anyway, there’s a fun little romp around the conference. Let me know what you think of the new series ... per the disclaimers, I don’t have much control over this particular dataset, but it’s all new data we’re looking at compared to seasons past.