2020 Defensive Personnel: Bills and Packers Nickel and Dime the League
Yesterday, we looked at which personnel groupings were most popular on offense in 2020. Now, it's time to flip that around and look at how defenses responded.
The story of offensive personnel was the lack of story—the degree to which personnel usage did not change in 2020. That raises an interesting question defensively. With no significant change in the composition of opposing offenses, would we see any change in defensive personnel groups? After crunching the numbers, we can come with an absolute, unqualified … maybe. It's complicated.
In 2020, defenses played in nickel on 59.7% of all snaps, essentially erasing the dip we saw in 2019 and continuing the trend we have been following for a decade. Nickel remains the default defensive package in the league, by a margin that only seems to be getting larger as time goes by. It's not an enormous year-to-year jump, but it's certainly there.
|Defensive Personnel Groupings, 2020|
|Personnel||2019 Pct||2020 Pct||Difference||2020 DVOA|
A few quick notes before we continue:
- For the record, 51% of base snaps were in 4-3, and 45% were in 3-4; the two tend to bounce back and forth from year to year. That doesn't add up to 100%; there were 153 snaps of 2-5 (mostly from Arizona), 150 snaps of 5-2 (generally from Washington or Tampa Bay), three snaps of 1-6, and one each of 6-1 (Philadelphia) and 0-7 (New York Giants).
- Fourteen teams stuck to either 3-4 or 4-3 at least 75% of the time, with the Colts notably running all 206 of their base defense snaps in a 4-3. That means that the majority of the league is more flexible, as the distinction between defensive ends and linebackers continues to diminish. In 2019, only 10 teams had less than a 3:1 ratio between 3-4 and 4-3, which was an increase from seven teams in 2018. We're now up to 18, and 12 of those teams saw neither 3-4 nor 4-3 hit two-thirds of their 2020 base snaps. Base is becoming less frequent, but it's becoming more flexible.
- "Dime+" includes any package with more than five defensive backs. That includes all your dime packages, as well as the 278 snaps in quarter and 14 snaps with eight defensive backs on the field, including one snap where the Giants trotted out the half-dollar package in the red zone. They allowed a Mitchell Trubisky touchdown pass on the play, for the record.
- "Big" defenses are 4-4 or 3-5 lineups, while "Goal Line" includes all other personnel groups with fewer than four defensive backs. Despite the name, there's no requirement for goal-line defenses to be played on the goal line; Dallas even had a snap of 6-4 against Washington with the Football Team facing a first-and-10 inside their own 20.
A 3.8% bump in nickel is significant, if not landscape-altering; we're talking an average of three extra snaps of nickel defense per team per game. An extra 1,500 snaps of nickel defense is nothing to scoff at, but it falls on the outer edge of what you might expect to see from scheduling differences and year-to-year variance. Some defensive playcallers attempt to heavily match personnel; others stick with one unit regardless of what the offense brings to the table. Get the right set of matchups, and you'll see nickel usage rise, especially if one or two teams are really driving the train.
But then again, you could make the argument that 2020 was evidence of the all-nickel trend continuing to increase, above and beyond year-to-year and situational fluctuations. This is the first year since 2016 when both base and dime+ usage have gone down at the same time. Nickel usage went up in every split—against 11, 12, and 21 personnel; against the pass and against the run; you name it. The average snap in 2020 had 4.89 defensive backs on the field, a new record in our data (going back to 2011). Nickel was everywhere.
Defenses Against Top Offensive Personnel Groups
|Defensive Personnel DVOA Breakdown, 2020|
|Split||% Used||DVOA||% Used||DVOA||% Used||DVOA|
There are three main takeaways from this table.
First, the reason why defensive DVOA in base is still better than DVOA in nickel is almost entirely play selection. Nearly 60% of snaps in base are running plays and running remains less efficient in general than passing, and an extra linebacker tends to help more against the run than a nickel corner or extra safety does. It's the inverse of the "don't run into loaded box" advice; on defense, you should load the box if you know they're gonna run.
Secondly, you can see why offenses may be interested in experimenting with 12 and 21 sets. Base defense performs better than nickel against these heavier sets regardless of the play call. It has a better DVOA against both the run and the pass from heavier sets (though you start running into selection bias with base being used more in short-yardage situations). As the league's defenses continue to zig towards using nickel more and more, the offensive zag might be to do what the Patriots are doing and grab a pair of pass-catching tight ends to try to take advantage of smaller defenses. Especially if, say, one of your division rivals is in nickel 90% of the time. More on that later.
Thirdly, base against 11 personnel has basically vanished. It has been going the way of the dodo for years, but there has usually been someone stubbornly trotting out third linebackers against slot receivers, most notably Gregg Williams and the 2017 Browns of 0-16 fame. 2019's base-on-11 champs, Seattle, saw their nickel rate increase significantly, in part because of more trust in their depth corners and in part because newly acquired Jamal Adams could slide up and play pseudo linebacker in nickel and dime formations. More than half of Seattle's 'nickel' against 11 personnel had Adams in the box, with 2020 corner Ugo Amadi replacing 2019 safety Bradley McDougald in coverage; not exactly a massive paradigm shift from 2019. Your 2020 champs, the Steelers, ran just 116 snaps of base against 11; no one else had over 40.
In fact, it seemed like the league as a whole smartened up about base personnel in 2020, which lets us segue nicely into the team-specific table.
Team by Team Frequencies
|Defensive Personnel Frequency, 2020|
Jacksonville led the league by using base 39% of the time. The previous low-water mark for a league leader was the 2018 Broncos at 45%, and there has been at least one team above 50% in every other year in our dataset. Not only did no team have a majority of snaps in base, but no team even had a plurality for the first time ever. The increasingly misnamed base is now a changeup defensive front throughout the league.
Abandon the Base
In 2019, the Ravens became the first team in our dataset to use base less than 10% of the time. In 2020, we had four such teams, with the Panthers and Patriots tying by playing just 46 snaps apiece in traditional fronts. For Carolina, a lot of that comes from Jeremy Chinn, who we have listed as a defensive back but really plays a hybrid linebacker/safety role. As a rookie, he was basically the player the Panthers hoped Shaq Thompson would be, and it's not like Thompson has gone anywhere; Carolina has arguably the most versatile linebacker-type players in the league.
As for the Patriots, well, those 46 base snaps don't get across the extreme efforts the Patriots used to avoid base defense. 23 of those 46 snaps came in Week 7 in a blowout loss to a run-heavy 49ers team; those were the only base snaps New England had in the first nine weeks of the season. The average snap of base for the Patriots came with a league-low 6.7 yards to go. When the Patriots did use base, opponents ran the ball 70% of the time; Bill Belichick did not want his base defense out there unless he was damn sure you were going to run. The loss of personnel last season likely dictated some of that; you can expect the Patriots' base percentage to go up with the additions of Dont'a Hightower, Matt Judon, Kyle Van Noy, and others. But the Patriots have been among the vanguards of abandoning the seven-man front, ranking in the bottom three in base percentage in each of the past four years.
The Patriots ran 458 snaps of dime+ and 456 snaps of nickel. So, by the thinnest of margins, they're one of only two teams to not have nickel be their primary defensive formation in 2020. The Packers are the other; they're the first team to be over 50% dime in back-to-back seasons and have been either first or second in dime usage in each of the past four years. It hasn't paid great dividends; they haven't had a top-10 defense since 2015. They used dime+ 15% of the time that season, a rate that has risen every year since, but only really entered hyperdrive when Mike Pettine took control. Well, Pettine is gone, replaced by Joe Barry. Barry worked with Brandon Staley with the Rams last year, and Staley comes out of Vic Fangio's system currently in use in Denver; both of those teams had significantly lower rates of dime usage, and I would expect the Packers to go the same way in 2021. It's an interesting clash of personnel and style; Green Bay now has former Arizona/Atlanta linebacker De'Vondre Campbell but if the Packers want to go with two inside linebackers, that puts Krys Barnes or Kamal Martin on the field, so we'll see to what degree talent pushes Barry to use extra defensive backs in 2021.
Changes from 2020
The two teams that saw the most dramatic personnel changes in 2020 saw new defensive playcallers. In Carolina, Matt Rhule and Phil Snow took over for Ron Rivera, and the Panthers went from a nickel-first 32/66/1 base-nickel-dime split to a dime-heavy 5/54/41; the Baylor boys came in and radically altered how Carolina played defense. A similar transformation happened in Los Angeles, where Staley replaced Wade Phillips. The Rams' base percentage dropped from 34% to 15%, and they went from last in the league in nickel usage to firmly middle-of-the-pack. With Staley now across town with the Chargers, you can likely expect a similar transformation, which would mean a heavier dose of dime in 2021.
Personnel changes explain some other moves. The Seahawks still played more snaps in base than most other defenses, but after the acquisition of Jamal Adams -- as noted a few paragraphs above -- they started to use more defensive backs from time to time, with their base percentage falling from 69% to 38%. If you're going to spend multiple first-round picks on a guy, he'd better alter your defense. Meanwhile, the Ravens backtracked from their base-light 2019 season, shooting up from 9% in 2019 to 21% in 2020. Slot corner Tavon Young tore his ACL in Week 2 and Earl Thomas was let go before the season began, so Don Martindale's crew had to adjust on the fly; we saw plenty more snaps from Tyus Bowser, Yannick Ngakoue, and first-round pick Patrick Queen alongside Matt Judon and a returning Pernell McPhee in 2020. I imagine the Ravens will be looking to use more nickel and dime packages in 2021 with Judon and Ngakoue out of town, but we'll see.
The final two teams we have to mention both play in the AFC East. Both the Bills and Jets played more than 80% of their snaps in nickel, with Buffalo setting a new high-water mark at 91%. We'll somewhat set the Jets aside as A) they were bad and B) they'll almost assuredly be playing more base with Robert Saleh taking over the reins, but the Bills are really interesting. They only had four games all season where they played double-digit snaps in something other than nickel, and at least three of those can be chalked up to injuries to Matt Milano, Tremaine Edmunds, and Micah Hyde. Over the back half of the season, with Milano healthy and playing every week, those numbers only rose. When Milano started, Buffalo averaged 4.3 snaps of base defense, 0.4 snaps of goal line defense, and zero snaps of dime+ or big per game—and nearly all of those base snaps came in Week 8 when Milano was injured and Week 16 where the Bills ran out to a huge lead and basically rested everyone for the fourth quarter.
Milano doesn't jump out as someone you build a defense around, but he has ranked third, second, and second in pass coverage success rate over the past three seasons. When you have a linebacker who can cover well, it frees playcallers up to do so much more with their defense—or so much less in Buffalo's case, as they were content to just sit back in Cover-3 with the same basic personnel on the field at all times, counting on their ability to play their scheme better than an offense's ability to scheme against them. With Milano healthy over the last six weeks of the season, the Bills had the sixth-ranked defense in the league. A nickel rate of 90% or more still feels very, very high, and the Bills drafted a couple of edge rushers who could push that number down in 2021, but you can see why Sean McDermott was comfortable rarely going to sub packages last season.