Interception regression likely to hit Packers defense in 2020

When discussing the Green Bay Packers’ 2019 defense, most discussions revolve around the run defense. The unit was gashed viciously by the 49ers in the NFC Championship Game, and was alternating between barely good enough and a legitimate problem all season. In these discussions what is often lost is the more important part of defense: pass defense.

The NFL is, of course, a passing league, and while passing offense does a lot of the dictating, passing defense certainly plays a major role in the equation. The Packers passing defense was good in 2019, ranking 9th in DVOA and 8th in EPA/play. That improvement was welcomed, and should be expected after the Packers spent significant cap resources at EDGE with the Smith Bros. and at safety with Adrian Amos. Green Bay also added safety Darnell Savage in the draft and had Kenny Clark, Jaire Alexander, and Kevin King entering their prime years.

While Green Bay’s pass defense was good last year, it was certainly not great. While they did rank 9th in DVOA, they were far behind the league’s top units. The Packers’ -1.3% pass defense DVOA (negative on defense is good) was as close to 20th in the league as it was to the top five. The pass defense was more at the top of a group of pretty good pass defenses than truly being a difference maker. If that level of play was to be improved upon in 2020, they could jump into that higher tier of pass defense.

For the first time in a half-decade, Green Bay had a legit EDGE duo with Za’Darius and Preston Smith, a dynamic interior rusher in Kenny Clark, and a secondary that combined the potential of Jaire Alexander and Darnell Savage with the veteran consistency of Tramon Williams and Adrian Amos. Williams is gone, but much of the unit returns, with many of the key players heading into their primes. While some regression can be expected from Za’Darius Smith’s career year, the rest of the group didn’t seem to play over their heads in 2019, or else they are heading into their prime years. However, there are things working against Green Bay improving in 2020.

Green Bay’s defense produced quite a few turnovers in 2019. According to the NFL’s play-by-play data, Green Bay ranked third in interception rate at 3.34%. Getting interceptions is great, and it’s always going to play a part in having a good pass defense and will be necessary to have a truly great one. A lot of research has been done in the sustainability or repeatability of turnovers year-over-year and even within a year. The tidiest show of how volatile turnover creation is is this table taken from a Josh Hermsmeyer piece from January 2019 (Note: Format has been changed):

Football Outsiders and FiveThirtyEight

While offense, and particularly passing offense, is quite predictive of future passing offense, the two least-predictive things going forward were fumbles and interceptions. This is problematic for Green Bay, because they benefited heavily from interceptions last season.

According to the NFL’s play-by-play data, only the New England Patriots garnered more EPA value from interceptions than Green Bay last year, as you can see in the graph below (negative EPA is good on defense).

NFL play-by-play data

Green Bay garnered a lot of value from interceptions in 2019, and interceptions are not very stable year-over-year across the league, so how can one try and find an idea of how Green Bay’s defense would look in 2020? With interceptions being very noisy year-over-year, what may work better is looking at all defensive plays where an interception did not occur. This would encompass less randomness and probably better capture how good a defense is at limiting yardage. Football Outsiders had Green Bay ranked 9th in pass defense DVOA, and they ranked 8th by EPA-per-play, also slightly in the negative at -.0022 in total passing defense. Once interceptions are taken out of the equation, Green Bay drops from 8th in EPA-per-play to 15th.

Green Bay falls from a good pass defense to an average one. This passes the smell test as well. It’s also important to note that Green Bay had quite a few legitimately bad quarterbacks on their schedule last year. Thanks to injuries and general opponent incompetence, they played half of their regular season games against a player that was either a backup (Matt Moore and David Blough) or ranked 21st or worse in DVOA. DVOA does attempt to adjust for opponent, but EPA does not. The fact that Green Bay got to beat up on so many bad quarterbacks yet was still so reliant on interceptions to be anything more than average is concerning.

Compare that easy 2019 schedule with 2020 and the situation looks much worse. Not accounting for potential injury, Green Bay has ten quarterbacks on the schedule who ranked in the top 14 in DVOA last year. The remaining group includes names like Philip Rivers, Tom Brady, Carson Wentz, Nick Foles, and Gardner Minshew. Three of those are hardly pushovers, and Minshew and Foles, while not good, are certainly not disasters in the way that Kyle Allen, Mitch Trubisky, and Dwayne Haskins were last year. The road will be tougher for the Packers pass defense in 2020.

It’s not all doom-and-gloom though. Jaire Alexander should be coming into his prime. Darnell Savage is heading into year two. Kevin King finally got a full off-season. Maybe Josh Jackson does something. Chandon Sullivan looked good in limited reps last year. Maybe someone other than Kenny Clark will be productive in the interior line. Perhaps Rashan Gary takes his strong physical gifts and turns them into production.

Green Bay is going to need a lot of these to hit in 2020 just to stave off the regression gods and hold serve. In order to take a leap into the upper echelon, they may need a miracle.