OFFENSIVE/DEFENSIVE RECAP: 2023 Wisconsin
Did Nebraska find a quarterback?/Defense gets RPO'd to death
Glossary of Terms1
Link to Offensive Charting Sheet2
Link to Defensive Charting Sheet3
OFFENSE
At the start of last season, Brock Purdy was the San Francisco 49ers’ third string quarterback, an afterthought on the bench. But after an injury to the Niners’ starter in an early game, and then an injury to the second-stringer during a key late-season stretch, Purdy found himself thrown into the fire in games with dire stakes late in the year. He responded by becoming the most efficient passer in the league and winning more games than any other quarterback over the same span.
Now the same thing is happening with his younger brother for Nebraska.
…
Just kidding! Or maybe I’m not? I probably am.
But however it turns out, no one can deny the new Chubba Purdy era4 (if it’s even an era) got off to an exciting start: The sophomore scrambled 55-yards for a touchdown on his sixth play from scrimmage by outrunning the entire Wisconsin defense; threw a 58-yard touchdown pass six plays later; hit a few more impressive, accurate throws; and, on the whole, seemed to be operating the offense better by the eye test than either of Nebraska’s other two starting quarterbacks this year. In a season of what’s been the worst NU quarterback play of my lifetime, that’s all the bar someone has to clear to get me excited.
The two long early touchdowns were largely just backyard play/fortunate breaks, but the efficiency numbers do somewhat bear out Purdy playing better than the other two quarterbacks NU has used this year. Nebraska had 41.6% success rate in regulation Saturday, which isn’t particularly good, but it’s a marked improvement on the 37.2% season average efficiency and the highest efficiency game NU has had since Louisiana Tech. It’s also a lot better than the putrid stretch of offense Nebraska had been mired in for its last three games, when it averaged a 33% success rate under Heinrich Haarberg and Jeff Sims. And Wisconsin is a much better defense than what the other two QBs played — at least one of the three major statistical services has the Badgers in the top 10 nationally in defensive efficiency.
There were a couple big underlying factors on why things functioned better under Purdy. The offense’s efficiency on first down (42.3% success rate on first downs) was seven percentage points higher than season average; doing even marginally better on first down makes it way easier to function well on the later downs. Second, Nebraska’s success rate on true dropback passes (non-RPOs or run-play tags), was also 11 points higher with Purdy on Saturday than it had been under Haarberg or Sims (40% to 29.5%). In essence: Nebraska actually got some yardage on first downs and was able to complete some dropback passes. Who knew that would make your offense work better?
It’s hard to tell a ton about quarterback decision-making without knowing how NU teaches certain concepts or having the All-22 film, but with Purdy I could at least see him doing some progressions and following some process, which is more than I can say about the other two passers’ panicked styles. And I think the coaches agree with that, as we saw more downfield passing concepts and more complex passing designs called Saturday. The staff really only trusted Haarberg to run Stick, some quick-slant/post stuff, and an occasional downfield shot. But with Purdy, coordinator Marcus Satterfield was calling a lot more layered downfield concepts and downfield breaking routes that require anticipation and accuracy, as well as sort of counters off Nebraska’s most-used concepts that are more advanced.
Another element Purdy brought to the table Saturday was a willingness to actually throw the ball out on run-pass options and tags for running plays. RPOs and run tags are supposed to be handoffs, but if defenders align certain ways or behave certain ways after the snap, the quarterback has the option to pull the ball out and throw some quick routes into wide open space. Nebraska didn’t run a ton of these with Haarberg or Sims — especially not with Haarberg — with about 6.3% of its calls being RPOs and 5.5% or its run plays having a tag on them.5 The previous quarterbacks were also essentially not getting the ball out to the receivers or tags when they did run these concepts, with just 6.6% of these plays being thrown out to the receivers.
That was different under Purdy. He got the ball to a receiver on an RPO or tag on 28.6% of his plays, over four times higher than the rate the other QBs had been. Purdy threw six RPOs/tags on Saturday; the previous QBs had thrown eight in the previous 10 games all season. Nebraska’s coaches also trusted Purdy to run more of these plays, with a 13.3% RPO rate (double what it had been for the other two QBs) and a 28% rate of tags on running plays, almost six times higher than what the other QBs had been given. This is likely because he was actually throwing them as intended; coaches are a lot more willing to call a play if they know you’re going to try to fully execute it. So if you were watching Saturday and thought, “Wow; it sure seems like Billy Kemp is catching a lot of bubble screens,” you weren’t wrong.
They also just trusted him to run some more complex variations of RPOs. This is one rep from early in the game that Nebraska would end up running a couple of times:
This is a strongside pull sweep as the run element of the RPO with two slant routes behind it as the pass element. Purdy has the option to pull the ball and throw one of the slants based on the leverage of the edge defenders and general composition of the box. This is the first time Nebraska has run an RPO like this, with most of the RPOs called for Haarberg and Sims being little slide or flat routes to the tight ends or a very occasional downfield post route where they were reading one linebacker. This double-slants concept called for Purdy is a bit of a higher degree of difficulty and requires a quarterback you trust to deliver the ball with timing and accuracy, as a misplaced ball on a slant route has a high chance of being an interceptable pass. There’s a reason they didn’t call this for Haarberg.
Outside of the quarterback play, Saturday was probably the Huskers’ most “spread” game of the season. Nebraska was in 11 personnel (one back and one tight end) on a season-high 76.7% of its snaps; the season-average entering had been 55.2%, and the previous high had been 63.3% against Illinois. It played with multiple tight ends or a fullback on the field on just 14 total snaps, a large decrease from the 43% it had been using entering the game.
The desire to go spread and outside was also evident in the run game. NU used Outside Zone, Pin-Pull, Toss, or outside option concepts on a combined 44.3% of its snaps Saturday, way above the 20.8% it had run those same plays entering. Satterfield also called none of the deep-shot play-action out of heavy personnel he had been calling for Haarberg and Sims, with just two play-action attempts period for Purdy.
Some of this was an injury to tight end Nate Boerkircher putting them down a starter and the two drives at the end of each half up against the clock messing with the numbers, but I also think NU wanted to attack this defense by getting it wide. Wisconsin doesn’t align that differently on defense than Nebraska’s defense does, with a lot of flexibility and players in the spine, pre-snap movement and dropping into single-high shells. So wanting to avoid the chaos and confusion in the middle and beat them to the edge makes sense.
Notes:
The last four games have seen Nebraska really try to condense its formations. Entering the Michigan State game, NU was using “condensed” formations — or alignments with the receivers in tight to the linemen — on about 14.8% of its snaps. But over the last three games, the rate has been more than doubled, at 33.8%. Condensing the formation is a way to get opposing corners and safeties more involved in the running game and tackling, with the hopes of breaking big runs to the outside. It seems as if Satterfield has made a bet that his best offense is hoping Nebraska is able to pop big runs in traffic. The rate of condensed formations was only at 26.7% Saturday, as I mentioned things were a bit more spread with Purdy than they had been.
Nebraska on Saturday used its second-highest rate of motion on running plays, at 51.2% of its snaps, trailing only the Colorado game. Motion attached to runs is a largely used to try to influence linebackers. So Satterfield didn’t think these Wisconsin LBs were much better than their Colorado counterparts.
The NU offense has been bad in a lot of areas, but has low-key been excellent in short-yardage situations this year and was again Saturday. The Huskers entered converting 68.2% of the time in third- or fourth-and-3 or less, and they were 4 of 5 in those situations Saturday. The one time they weren’t was the first-half fourth down attempt on the edge of field goal range that cost them points, a pretty unfortunate big of sequencing.
Satterfield called two trick plays Saturday, one the double pass of the Toss action — the same play that scored the long touchdown against Minnesota in the opener — and a reverse off of an I-formation Load Option play. Those are the first two trick plays he’s run since that Minnesota score. He was trying to pull out all the stops.
DEFENSE
Nebraska’s defense lost Saturday on second down.
The Huskers were just ticks worse than their dominant season averages in efficiency on first and third down against Wisconsin, but the Blackshirts gave up uncharacteristic big plays on the middle down, letting the Badgers set up a ton of third-and-shorts that should have been third-and-longs and kept drives alive.
Nebraska faced 20 second downs Saturday and was successful on just 9 of them. Many of these unsuccessful plays came in the form of 6-, 7- or 8-yard gains that didn’t seem huge in the moment, but the bar for a “successful” play for the offense is gaining 70% or more of the yards you need to convert. When you rephrase it that NU gave up over 70% of the yardage it needed to defend on over half of the second downs it played … it sounds worse. Nebraska’s average success rate on second down entering had been about 65%. The lack of success on second down also led to harder third downs for Nebraska, with Wisconsin only needing to gain 6.7 yards per play on average to convert, the second-worst performance by Nebraska on the season, behind only Michigan.
And it wasn’t just about efficiency: Among the 10 successful plays for the offense on second down were gains of 26, 17, 13, 11 and 10, among Wisconsin’s biggest plays in the game.
Overall, this was probably the defense’s worst performance of the year. The Blackshirts finished with a 55.17% success rate, which was eight percentage points lower than its season average and its second-worst mark of the season. It was far less efficient against Michigan (a horrid 35% success rate), but Michigan is an excellent offense. There’s not a lot of shame in getting steamrolled by this Wolverines attack. This Wisconsin offense, though, is not a good unit and has been one of the least efficient teams per play in the Power 5 and has almost no explosive plays on the year. Some of the lack of success has been driven by an injury to a quarterback who has now returned, but this is, at best, still a mediocre offense. So to perform at the level NU did — especially when staked to a 14-0 lead early — was pretty disappointing. It’s hard to be mad at this defense for, essentially, one bad game, especially considering the unit on the other side of the ball its dragging along, but with an average performance Saturday from this Blackshirts group, the team probably has a bowl locked up.
Part of the lack of success was just in how Wisconsin plays. The Badgers ran RPOs on about 38.9% of their plays, the most this season of any Nebraska opponent. The three other worst efficiency performances by the 2023 defense have come against teams that use a lot of RPOs: Colorado (59.7% overall success rate), Michigan (35.5%), and Maryland (62.0%). It makes some sense, as the 3-3-5 is a defense built around disguise and doing unexpected things after the snap; if your offense is built to react to what the defense does post-snap, you’re going to perform better than a more train-on-the-tracks scheme. And, to be fair, NU has performed well against some RPO teams, as Northern Illinois and Purdue each ran a lot of RPOs and Nebraska finished with high efficiency against each. Saturday was just an example of the scheme getting burned by an offensive scheme that uses optionality.
As for coordinator Tony White’s gameplan, we saw predominantly four-player fronts again (51.8% usage), for the fifth straight game, though he did deploy some more three-player fronts (23.2%) and Drop 8 coverage (15.25%) after largely eschewing those over the same stretch. White started the game with about an average amount of pressures, blitzes or line stunts in the first half, and then brought way more heat in the second half. He brough pressures on 57% of second-half snaps, and used a defensive line game on 39.4% of second-half snaps. He used line games on just four snaps in the first half.
I imagine this is largely because Wisconsin was having success with its running game getting solid yardage and pushing the pile. In an attempt to generate more chaos, White tried to bring more blitzers and have box defenders shoot gaps from unexpected spots.
It wasn’t a bad plan, but Wisconsin was able to burn it on a few plays, especially late in the game on quarterback draws with its quarterback jitterbugging to find the seams where defenders were vacating. Those draw plays were devastating; Wisconsin used four in regulation Saturday, and three were successful for the offense, gaining a combined 36 yards.
This was also Nebraska’s second straight game with very elevated Cover 3 usage. Nebraska entered the Maryland game using Cover 3 on about 27.9% of its snaps but used it at a 43.9% rate and a 38.9% rate over the last two games. It hasn’t been super effective, with just a 50% success rate over those two games, worst among all of Nebraska’s coverages.
One other coverage thing I notices was a cool new variation on Cover 2. I’ve talked in recent weeks about Nebraska running it out of a four-down, two-high shell with its linebackers vacating in the middle and a safety coming down:
Then last week, it showed some new variations with the deeper safeties coming down to play the flat areas:
Saturday, it showed a new disguise, this time out of a three-player front:
NU had previously only played this Cover 2/6 package out of four-player fronts, but Nebraska aligns here in a three-player front before the snap, making it look like a usual Drop 8 play. But afterwards, one of the linebackers shoots in to form the four-player front, and the C2 look plays out as normal. Just a designer disguise that shows you coordinator Tony White is concerned about this stuff.
White mixed in a lot more personnel groupings Saturday. NU had entered playing base defense on about 86% of its snaps with about 7% Nickel usage and 7% usage of a pass-rush package. Most of that Nickel usage had come in an emergency against Michigan, too, when starting safety DeShon Singleton went down with an injury. Factoring that out, Nebraska had been about 87.9% base D and 3.8% Nickel in its other games. But White was a lot more willing to take a linebacker off the field Saturday, deploying 8% Nickel usage on passing downs and also running 5% usage of a “heavy” package with an extra defensive lineman on the field in short yardage. I had had him doing this on only a few snaps total throughout the season.
Happy Thanksgiving to those who celebrate! Combined post this week as I’m flying and didn’t have time for two separate in-depth ones. Will be back to normal schedule for Friday’s finale against Iowa.
Yards Per Play measures how many non-penalty yards NU gained on a possession divided by its non-penalty snaps, a measure of its efficiency. Success Rate measures how often an NU play gained 50% or more of the yards it needed on a first down, 70% or more of the yards it needed on second down, or 100% or more of the yards it 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 NU allows.
Purdy made a handful of unimpressive spot starts last year for an injured Casey Thompson, but I’m not counting those as real.
A “tag” is an add-on to a nominal running play, like a quick bubble screen to the slot receiver.