Weekly Thoughts Week 3

A look at Bama/Ole Miss, Louisville, and SEC Quarterback play

One of the more interesting things to me is the way that possessions screw with our understanding and interpretation of statistics. A major theme this week was the offensive “dominance” that Louisville put on against Syracuse. But that’s really deceptive; Louisville, due in large part to a big-play offense, scored extremely quickly. That meant they had an absurd 19 possessions against Syracuse. Most teams try to avoid that number of possessions against inferior opponents due to injury concerns. But it also means the numbers get over-inflated.

Here is a comparison of Louisville against three SEC teams versus low-end FBS teams:

Team Scoring On Possessions Points/Possession
Louisville Scored 10/19 possessions 3.26
Alabama Scored 4/12 possessions 3.16
Vanderbilt Scored 8/14 possessions 3.58
Auburn Scored 9/11 possessions 4.64

* possessions exclude game-ending kneel downs

There are three very interesting stats here:

First, Louisville scored often, but they didn’t score touchdowns all that frequently. Alabama, despite only 4 scoring possessions, averaged almost as many points/possession due to defensive scores and touchdowns. Auburn, despite scoring over twice as often as Alabama, was about 50% higher.

Second, Louisville didn’t score that often. Only about half of Louisville’s possessions resulted in points of any kind, compared to over 80% of Auburn’s possessions.

Finally, the closest SEC equivalent to Louisville’s performance is arguably Vandy. Vanderbilt scored slightly more points/possession, and they also had a higher percentage of scoring drives (57% to 53%).

Louisville had a serious “wow” factor because they generated over 10 yards/play. However, that is largely a product of style, as Petrino throws downfield on a high percentage of plays, and they generate a lot of big plays running their quarterback. However, their offense isn’t overly consistent. Of their first 13 possessions, Louisville drives went as follows:

  • Touchdown – 6 times
  • 3 and Out – 4 times
  • Fumbled – 2 times
  • Interception – 1 time
Louisville had a serious “wow” factor because they generated over 10 yards/play.

While there are a lot of points there, Louisville was actually more likely to either go 3 and out or turn it over than they were to score. That indicates a volatile boom-or-bust offense, and one that is extremely susceptible to a bizarre loss on a bad day.

One of the other reasons it looked so much better on TV is in large part due to the fact that Louisville had 12 possessions in the first half. That’s as many as or more than Alabama and Auburn had in their entire games. Put differently, by the time Louisville had scored 42 points (with 3 turnovers), Alabama had scored 38 (with 1 turnover) and Auburn had scored 51 points (with 1 turnover). There is a strong argument that Louisville, despite the high yards/play, performed worse than Alabama or Auburn from the perspective of winning the game. Thus, I would be very critical of early pundits who are going to go gaga over Louisville’s offensive production. It isn’t a mirage (they are still having crazy yards/play averages), but the production doesn’t actually translate all that well into winning ballgames.

SEC Quarterbacks

Fun fact: there are currently 3 quarterbacks in the SEC ranked in the top 5 in both yards/play and QB Rating. They are Eason, Hurts, and White: two true freshman and a redshirt sophomore, with Eason actually leading in both categories. As usual, all the pundits picked well-known returning starters as their All-SEC players (Kelly and Dobbs). However, while Kelly was a top-5 rated guy at both positions (and quite possibly will end there again, as FSU is skewing his numbers), Dobbs was only 8th in the SEC in QB Rating and 10th in yards/play last season. Simply put, Dobbs wasn’t particularly good last year, and it was almost inevitable that some of the first-time starters would end up being better. Further, it was extremely misguided to assume that, because 4 guys ahead of him were leaving, Dobbs would suddenly improve. Even if he were to jump from 8th to 4th as a result of the departures, it would have been because other teams got worse, there simply wasn’t much reason to expect a multi-year starter (throwing to the same players behind a worse line) to suddenly get that much better.

Some Teams May Not Be What You Expect

Here are the bottom-five SEC teams in two major categories:

(Worst 5 in yards/attempt allowed)

  • Miss St. – 6.8 yards/attempt
  • LSU – 6.9 yards/attempt
  • Auburn – 7.9 yards/attempt
  • Ole Miss – 8 yards/attempt
  • Kentucky – 9.5 yards/attempt

My main takeaway here: It is very surprising to see LSU and Auburn on this list. Going into this season, LSU was expected to have one of the best secondaries in the conference, and Auburn was supposed to be greatly improved. However, LSU (who was only 9th last year) has regressed to 11th, and Auburn has actually fallen from 6th to 12th. Some of that may be due to the opponents, but I note that Clemson actually averaged less yards/attempt against Troy than they did Auburn. Both of these pass defenses may have some cracks that could turn into real issues later in the year.

Worst 5 in yards/carry (on offense)

  • Tennessee – 4.11 yards/carry
  • Ole Miss – 4.08 yards/carry
  • Arkansas – 3.45 yards/carry
  • Kentucky – 3.22 yards/carry
  • South Carolina – 2.76 yards/carry

It’s really, really interesting to see Arkansas on this list. Even though they faced a TCU defense that gave up over 10 yards/play against South Dakota State, Arkansas is still the 12th worst rushing offense in the conference (in comparison, they were 4th last year). Allen has a live arm, which was able to generate offense against poor TCU and LT secondaries, but this should be a huge red flag for anyone that is expecting Arkansas to be a sleeper pick in the SEC this season.

Game Of The Week: Alabama @ Ole Miss

First off, the quarterbacks. There is really one reason Jalen Hurts has won the job. It isn’t his arm (he’s probably not quite the passer that Barnett is at this point), and it’s probably not even the athleticism. The reason he won the job is that he’s more composed. Barnett still looks shell-shocked when he steps onto the field, and spends most of his time looking to the sideline for instruction. Hurts, by contrast, is directing the players to get lined up and clearly seems in control. Whereas Hurts lined up Robert Foster to avoid a negative play, Barnett is motioning a man into the snap (creating one). Given the talent on offense, there is no question that Saban wants the quarterback who will help them avoid negative plays. Strangely enough, it’s Hurts.

I really am of the opinion that Chad Kelly is a first-round talent. He is very mobile, has adequate size, and has a cannon of an arm. Honestly, he’s probably the best pure passing talent that the conference has seen since Stafford. His issue, which still isn’t corrected, is that he is also every bit as liable to make poor decisions as Bo Wallace. In part due to his arm talent, Kelly has never really learned to hold the ball, he’s going to throw it whether his guy is open or not. Alabama wins this game if they force the throw early so that Kelly is throwing to a covered target, which should result in multiple turnovers. The Ole Miss offense isn’t going to go silently into the night; their offense is volatile by nature, and they are almost as likely to cause themselves to be blown out as they are to put up big numbers.

On the other side of the ball, the game will be dictated by how Ole Miss plays defense. There is really no question here: Ole Miss’s weakness is the secondary. There are two ways they can play:

I really am of the opinion that Chad Kelly is a first-round talent.

The first is to play tight and take away the quick passing game. This seems doubly-beneficial: it 1) reduces the ability for our young quarterback to easily get the ball in space to our elite receivers while 2) buying time for their pass rush to come home. The issue is that if our quarterback, who is mobile, can buy time in the pocket, they will be extremely susceptible to big plays down the field. This is exactly the gamble that USC took and lost.

The second approach is to give a big cushion and let us have the short throw. This takes away from their pass rush, and it allowed Francois to pick them apart. However, it also requires our quarterback to make smart decisions and take what’s available. If they take this approach and Hurts tries to go for the deep play, he will be throwing into double coverage and/or will face heavy pressure. I honestly think this is the smarter play and what we will see: they will give us the easy play and simply challenge us to execute early. Only if Alabama starts succeeding will Ole Miss start playing press coverage and gamble with the blitz.

I’m guessing that Alabama executes relatively well with the pass, though the run may struggle a bit. After a few drives where things are close, Ole Miss is forced to play the receivers tighter, which ends up leading to big plays (and people suddenly declaring that the Alabama offense is “clicking” as a result). I also believe that this Alabama defense is tremendously better suited to generate negative plays and turnovers. As a result, I think Alabama wins by multiple scores.

Alabama - 34

Ole Miss - 17