TENDENCIES: 2023 First-Half Defense
More Charts About Football And People
Glossary of Terms1
This is the second in a series of midseason evaluation posts, focusing on Nebraska’s defense. The offensive post can be found here.
Bye weeks are often a chance for programs to detach from the week-to-week grind of game prep, take big-picture analysis of what’s worked on the field, and make adjustments to scheme or personnel. That NU’s staff probably spent last week in evaluation mode — and that Nebraska’s bye week comes at the exact halfway point of Matt Rhule’s first season — makes it a good time for the newsletter to also take a step back and draw some conclusions from the data collected from the first half of the season.
I’ve blocked out some percentages on both the usage and success of Nebraska’s overall performance, personnel, operation, fronts/box counts, pass rush, coverages, and safety alignment. To the right of that, where I have the data I’ve also provided the same for Nebraska’s defense last season under coordinators Erik Chinander and Bill Busch for comparison to what both the Huskers have done in the past.
Some of the charts/numbers had to get pretty small, but if you click on the individual graphic it should bring it up for closer viewing.
Overall Performance
The major advanced metrics all view NU as having a top-40 unit at worst, with ESPN’s SP+ considering them a borderline top-20 defense. The underlying numbers are also strong, with under 5 yards per play allowed in non-garbage time. The points per drive is a little worse than that, indicating the Blackshirts might me due for some positive regression in terms of teams scoring against them.
NU’s success rate on defense at the halfway mark would be the best for a Nebraska team since 2012, in the heyday of Bo Pelini’s strong units. The one concern so far would be the amount of explosive plays this defense is allowing, which has been pretty consistent across the board, even against some not great offenses. Minnesota and Colorado were both able to generate eight explosives each against them, and Illinois got seven. NU has been great down-to-down but has given up some back-breakers when they’ve gotten beat.
Personnel
The numbers here look starkly different from last year but are, in essence, identical. The 3-3-5 is what’s known as a “base nickel” defense, meaning the standard look for the defense features five defensive backs, the same number as NU’s nickel personnel in 2022. So, basically, the 2022 Huskers and 2023 Huskers are both playing with five defensive backs on the field for around 75% of snaps; this year’s defense just categorizes it as “base” and last year’s defense considered it “nickel.”
What’s different is how successful the personnel groupings have been: Nebraska’s five-defensive-back alignment has been almost seven percentage points better in overall success rate than last year’s group, and 16 percentage points better against the run. So the improvements to the defense haven’t largely come from what player-types have been on the field as much as how they’re being deployed.
One other notable thing in this section: 2023 NU has played 22% of its snap in its version of “nickel” personnel, with three corners on the field instead of three safeties. But almost all of those snaps have come since safety DeShon Singleton was injured two games ago. The Huskers had played just eight total snaps of three-corner personnel groupings in the first four games, but used it for 22 snaps against Michigan and 34 against Illinois. It has been less successful than NU’s base personnel, but they might have to use it more in the second half of the season if Singleton’s injury is long term.
Operation
Turns out, it’s a lot easier to play defense when you can stop the run. Nebraska’s opponents were able to get a successful play on a run or RPO 51% of the time last season, leading to an even 50-50 split on runs and passes. No sense in throwing the ball if you can pick up yards on the ground, etc.
But this year, Nebraska has been stuffing over 63% of its opponents’ runs and RPOs, forcing teams into more pass attempts and unbalanced offense. The overall data would say the Blackshirts’ pass D has been worse this year than last, but that inefficiency has largely come off the Michigan game: NU’s success rate on pass plays was 71.6% against Minnesota, 69.7% against Colorado, 55.1% against Louisiana Tech, and 73.1% against Illinois — but just 33.3% against Michigan. The Wolverines game is skewing that sample quite a bit; let’s see where it rests after the final six games of the season, which include some pretty dreadful passing offenses.
Fronts/Box Counts
Here’s where we get to our first major difference in scheme. Nebraska played in true three-player fronts (not including pressure packages) on 7% of its non-pressure package and non-goalline snaps last year; it’s using them at a nearly seven-times higher rate in 2023. And not only that, it’s been dominant out of them: Nearly 65% of its snaps out of three-player fronts have been successful. There was a lot of hand wringing in the offseason over if the 3-3-5 could stand up in the Power 5; so far, it’s passed with flying colors. NU has also just been plain better out of four- and five-player fronts, too, with success rates over five percentage points better in each.
And again for the offseason talk about the 3-3-5, NU has spent about the same amount of time in light boxes as it did last year — and been more effective with it. Where you really see the difference has been in heavy boxes, with NU cutting its usage in almost half from last year and now matching heavier sets with more standard boxes. Last year’s team also just had to use heavier boxes a lot to stop the run, so that number is probably a little high.
The other piece of data here is that Nebraska’s defense has been almost 10 percentage points more effective when it uses some sort of motion or shift in its front before the snap.
Pass Rush
One of the advantages of the 3-3-5 was supposed to be its ability to confuse offenses on who was rushing the passer and who was dropping into coverage. You can see that playing out in the diversity of rushes Nebraska has used so far this year. NU in 2022 was pretty much always bringing four or five players every snap, with those two being used on over 80% of snaps. This year, the Blackshirts are at about 57% four- and five-player rushes, while also utilizing nearly equal three-player rushes, and also bringing six rushers on a not-insignificant amount of snaps, too. That diversity is a lot harder for offenses to prepare for.
NU is blitzing about seven percentage points more than last year so far, but success on plays blitzing and not blitzing has been virtually identical, as has the use of sim pressures (snaps where it “simulates” a blitz before the snap but either drops out of it or brings pressure from a different spot). Where it’s blitzing is pretty interesting, though. Nebraska’s blitzes get progressively more effective the wider they are, with A-gap (between the center and guard) and B-gap blitzes (between the guard and tackle) being less effective than the defense’s overall success rate but pressures through the C gaps (between the tackle and tight end) and D gaps (outside the tight end) really getting home at a high rate while being used less than the interior pressures. If you wanted to point to one thing White might do differently in the second half, it’s maybe bringing more heat from the edge.
NU’s overall plays with stunts have been less effective than average, but that’s a data point that is again skewed by the Michigan game; White used them more frequently in that game as the UM line ran the ball on Nebraska with a lot of success. I’m not taking anything away from that, really. Let’s see where it’s at at the end of the season.
Coverages
The other big change in the defense from last year has been the increase in zone coverage. Nebraska has dropped its man-coverage rate nearly 13 percentage points — while doubling its usage of Cover 0 (aggressive all-man across the board with no safety help). The means the loss has largely come from Cover 1, which has fallen from 43.6% usage under Chinander to 28.7% under White.
What’s replaced it has been a more diverse zone coverage package. Offenses more or less knew NU in 2022 would be in one of Cover 1, Cover 3, or Cover 4; those three coverages combined to be used on 86.7% of Nebraska’s snaps. No other coverage has a usage above 4.3%. This year, however, six coverages have usage rates above 6%, with four over 10% usage. Cover 3 has now become the most used coverage, with Cover 2 and Cover 6 also getting big jumps. Being able to use more potential coverages makes it harder on opposing quarterbacks — a six-answer multiple choice question is harder to get right than a three-answer one. Cover 2 has been an especially deadly curveball early in the season, with its plays successful over 65% of the time. NU is using a Tampa 2-style coverage most of those reps with Rover Isaac Gifford often dropping down into the deep middle of the field. That variation is destroying offenses right now.
Safety Alignment
The coverage shells from this year and last year are somewhat similar, with near-identical use of a single-high safety but with NU eating into its two-high coverage rates to run more three-high. The three-high shell has been NU’s most effective alignment by far, but is also a smaller, more situational sample size.
The data on rotation was a little surprising; anecdotally, it felt as if the plays with rotation has been really effective, but in actuality they’ve been about identical to no-rotation plays.
The 3-3-5 treats its three safeties on the field as more interchangable than NU’s 2-4-5 structure did in 2022, allowing it to do some more cool stuff with coverage structures. NU’s best alignment has been having all three lined up deep, allowing it to morph into unexpected coverages after the snap. But it’s also been really good in a single-high alignment with the Rover deep and the field and boundary safeties both rolled down to cover slots and tight ends that it’s used on about 21% of its snaps.
The field safety (primarily Omar Brown so far this year) has played down in the box the most out of the three positions, spending over 70% of his snaps rolled down near the line of scrimmage over slot receivers as a more traditional nickel back. The boundary safeties (primarily Singleton and last week Phalen Stanford) have spent less time down, but when they have have also spent most of their time in the slot. Gifford at Rover has spent about 40% of his snaps down, though his has predominantly been in the box as another linebacker with some other slot work mixed in. All these alignments have worked at pretty much the same high rate.
Success Rate measures how often NU prevented a gain of 50% or more of the yards its opponent needed to convert on a first down, 70% or more of the yards its opponent needed on second down, or 100% or more of the yards its opponent needed on third or fourth down. An Explosive Play is any designed run that gains more than 12 yards and any designed pass that gains more than 16 yards. A Havoc Play Allowed is any tackle for loss, sack, fumble, interception, pass break-up or batted ball.