It’s been a tenet of this blog that ever since Gus Malzahn arrived on the Plains, the best statistical measure of Auburn’s defense isn’t total yards allowed or scoring defense or any other “counting stat” that depends just as heavily on how many plays the defense faces–which should always be a substantially higher number than average, thanks to Malzahn’s preferred tempo and love of the big play–but on yards allowed per-play. Yards per-play isn’t a perfect measure of performance (a long series of 4-yard gains can obviously be very productive even if a 4-yards-a-play-allowed average is an excellent number for the defense), but it still ought to be a more accurate than total defense.
But as it turns out, Auburn’s per-play statistics and accompanying total play numbers (as highlighted by commenter John and Auburn fan JD at TSK) are even more interesting than just “hey, the defense isn’t that bad.” Those numbers:
Offense: 193 total plays, 74th in D-I; 7.2 yards-per-play, 9th in D-I. (Total offense isn’t bad, either: Auburn’s 21st.)
Defense: 234 total plays, 118th in D-I; 4.4 yards-per-play, tied for 20th in D-I. (Total D: 61st.)
So … definitely some food for thought there. The questions I think are raised, and my best attempts at answering them:
If the offense isn’t running at Malzahn’s full-tempo pace and the defense is holding opponents to the 20th-best mark per-play in the country, how has the defense been on the field for that ridiculously many snaps? I think there’s a couple of factors, aside from the handful of snaps Clemson took in overtime:
1. Despite the slower pace thus far this season and relative success on offense, Auburn’s far from monopolizing the ball; the Tigers rank 95th in time-of-possession. Why? Most of Auburn’s possessions have ended one of two ways: a huge play and quick score, or a three-and-out. Neither keep the defense on the sideline for long.
2. We’ve discussed Ted Roof’s bend-but-don’t break approach ad nauseum, and it’s done its job on the scoreboard, but this is where the price is paid; Auburn’s actually done decent work on third downs (36th through three games, a spot ahead of Ohio State), but big gains against soft coverage on first and second down, especially early in drives, has resulted in a lot of long drives.
Put the two together, and you get two games already with 80-plus snaps for the opponent. It’s also worth noting that Auburn’s D has been (like the offense) somewhat boom-or-bust: they’re tied for 6th in TFLs and 25th in sacks.
Should Auburn be worried? How can they restrict the number of defensive snaps in the future? I think it’s far from time to panic; between the extremely high number of quick scores vs. Arkansas St.. and the equally odd number of three-and-outs against Clemson, I don’t think 80-snap games are going to be a regular occurrence. The numbers should regress to the mean a bit; if Auburn’s still in the country’s bottom 10 after, say, the Kentucky game, then it’ll be time to start worrying about the D slipping into the same second-half woes that plagued it down the stretch in 2009.
But all the same, now that Clemson’s big-play-dependent offense is behind us, it really might behoove Roof to start playing more aggressively on first and second down. If Auburn’s not going to substitute liberally (and at linebacker and corner, they’re not really substituting at all), the long-term benefits could outweigh the short-term risks.
I would not, however, advise slowing down the offense, since on the basis of the Clemson game, it works better the faster it goes. If it could just work better in the first half–building a lead and then slowing things down later with the game (theoretically) in hand–that should be enough assistance from the offense’s standpoint.
So does that 4.4 and 20th spot mean that Auburn’s defense isn’t all that bad? Three games isn’t much of a sample size, especially when one of those games is against Arkansas St., but it’s encouraging, yes, and without question a better indication of Auburn’s defensive quality than the misleading total defense numbers.
If Auburn’s offense was really the ninth-best in the country, shouldn’t they have put up more than 41 points in regulation the last two weeks? Well … yes.
To this point, as mentioned, the defining characteristic of the Auburn offense has been its extremely high level of variability. After three games, Auburn’s touchdowns have covered 3, 36, 2, 48, 71, 38, 2, 39, 12, 12, 8, and 78 yards … meaning that Auburn’s average scoring play has covered 29 yards, and that’s not even discussing the other long passes Auburn’s completed that haven’t resulted in immediate touchdowns. This is a dangerous offense, at the least.
But when Auburn hasn’t been putting up those huge gains, of course, they’ve been giving the ball back to their opponent with a quickness, either via turnover or the three-and-out. With the exception of a couple of those second-half drives against State, it’s been all-or-nothing, smashing success or total failure.
So what does that mean for the future? It seems unlikely that Auburn’s offense will continue to remain so manic-depressive; one way or another, they’re going to come back to the statistical mean.
But there’s two ways for this to happen. The depressing possibility is that the big plays dry up while Auburn sees only a modicum more consistency down-to-down, resulting in neither long drives nor huge scores … and a rather terrible offense. The other, though, is for Malzahn to continue finding the plays down the field while ironing out the kinks in the effort to maintain possession … giving Auburn the best of both worlds and the truly complete offense we expected entering this season.
Now, I may be admittedly viewing things though a pair of orange-and-blue colored glasses, but I think Auburn will edge towards the latter outcome. The Tigers have been plagued by negative, drive-killing plays so far; lots of penalties, lots of tackles-for-loss (86th in the country in TFLs allowed), lots of sacks (56th there, and that’s with Auburn only throwing the ball 20 times or fewer a game). Maybe I’m just wrong, but I think with an offensive line this experienced, those problems are fixable. We’ve settled on a tailback rotation; Newton should continue to improve as a passer; the line should be better than it’s been. There should be more extended drives, and we know Malzahn’s not going to stop going downfield. I think if Auburn’s already at 7 yards per-play, some improvement on the scoreboard should be on its way.
And overall? I’ll point you towards this link again, where Auburn’s per-play margin of +2.8 ranks fourth-best in the SEC … despite the fact that Auburn’s played a tougher schedule than anyone in the conference other than Georgia (and maaaaaaybe Florida). It’s not all the Arkansas St. game, either; Auburn finished at +1.6 against Mississippi St. and +2.0 (!) against Clemson. Again, per-play numbers aren’t totally accurate … but that’s still a pretty damn good sign that things aren’t nearly as bad as “17-14” and “27-24 in OT” might indicate.
I think. Probably. If the defense doesn’t get some rest … and the offense can’t keep up its rate of home runs … things might unravel in a hurry. But for the time being, I think there’s a lot more encouragement here than not.
You’ve got a typo in there.
“If Auburn’s defense was really the ninth-best in the country, ”
I think you meant offense.
It would be nice to see a standard deviation number for our offense per play. I’d have to imagine it would be in the top 5 in the country. The fact that our ypp differential is 2.8 and South Carolina’s is 0.7 gives me hope for Saturday night’s game. Going into the last game our ypp differential was 3.2 and Clemson’s was 3.1 (although against a weak schedule). Looking further into USC’s numbers shows that they have a tough run defense (#2 in the conference with 2.24 ypr) but a soft pass defense (#10 in the conference with 7.3 ypa and a 126.42 passer rating against). Unfortunately, that means we need to get Cam going early to loosen up the defense. If he comes out to the awful start he had against Clemson, it could be another tight one. If he can find his inner Manning and open up lanes for Dyer et al. to bust off good runs, our defense should be good enough to hold Lattimore and USC in check (remember, we do better against power backs, which it seems Lattimore is).
The official stats showed that defensive players who didn’t start accounted for just 20% of the tackles. And most of those were McNeil and Blanc.
Not much encouraging depth there.
Thanks, Neal, fixed.
JD, I’m planning a post on the “Aubur ndoes well vs. big backs” theory. Stay tuned.
Luke, I dunno, 20 percent is probably 10 times as high as that number last year. Backups probably weren’t playing more than 20 percent of the snaps, right? I think we’re OK at DT, safety, and OLB; it’s Bynes, the corners, and the ends I’m worried about.
Id love to see Gus introduce the slant into his O. It is a pretty safe play and can get the QB confidence level up early in the game. Also is it me or are we not throwing the 10 yard out route to the sideline that was a pillar of the O last year?
Subtracting the long TD pass to Zachery, Auburn was only +0.8 on Clemson. I think it’s tough to draw conclusions on Auburn’s offense given the high variance involved with running only 61 plays compared to 82.
Blake W says
WorkDontRun, I completely agree with you. I have been wondering over the past few days why we haven’t seen those routes to take advantage of Cam’s strong bullet passes rather than trying to make him throw touch passes that he doesn’t appear ready for.
CStiger, but is it really fair to subtract an entire 78 yard play? I mean, the play happened…
CST, if Malzahn didn’t try to hit those plays with such regularity–or if the 78-yarder was, say, a straight-ahead run that just happened to bust long–I’d agree with you. But it doesn’t strike me as a fluke when it’s such a major part of Malzahn’s philosophy, if that makes sense.
WDR, yeah, I don’t know why we can’t try to hit some shorter throws _before_ asking Cam to go long.
it really is feast or famine it seems like, but when we are moving with pace and in a hurry we seem to do better. When there is more motion in the back field, we seem to do better. when we run Malzahn’s intended offense we do better.
also, i agree with the short passes being a big factor to Auburn moving the ball better. that is how the other teams have moved the ball against us so well. once you see a couple of drives where we run to the outside, run up the middle, go for a deep pass, it gets pretty easy and predictable. thats what the offense has looked like more often than not this year. I have never played a down of football in my life. If i know what the play is going to be before the ball is snapped, there is a problem. Still, i trust the coaches, im just saying.
That said, iagree that we have played and beaten better teams so far than we did last year. i agree that both the miss st and clemson games would have been losses last year, so we must be better….we just don’t look it. i would rather win looking bad, than lose looking good.
We thought our defense was going to be a lot better this year because we finally have depth.
I looked at last year’s stats to compare how many tackles non-starters accounted for. Above I noted that backups made 20% of the tackles against Clemson. Over all of last season the backups made…
Aside: Wow, we were a mess at that 3rd linebacker position.
CST – so when we subtract out our biggest play of the night (BY FAR) we’re still almost a yard per play better than Clemson? I’ll take it!
Hey, maybe we have improved in D depth after all.
Subtracting out the 3 cupcake games from last year, the backups made 15% of the tackles in the “real” games. They made 20% versus Clemson and 16% versus MSU this year.
Blanc, Bell and McNeil are the only backups with more than two tackles in those two games.
But assuming Stevens applies a lot of Brillo, maybe we’ll have depth in all three areas! (Knocking on all handy injury-prevention wood.)
As a point of reference, the non-starters for the vaunted B*m* defense made 25% of the tackles in non-cupcake games in 2009. That’s less than I expected since they presumably had the most depth and most blow-out games of anybody.