The 2019 New England Patriots defense just completed one of the most shocking half-seasons of football most of us have seen in our lifetimes. You might think that's hyperbole. I disagree. After dominating opposing offenses across the first seven weeks of the season, the Patriots hit 8-0 by stifling the Browns in a 27-13 victory. In the process, they became the first team since 2012 to force takeaways on three consecutive plays. Linebacker Dont'a Hightower's fumble return marked the fifth time they have scored on defense or special teams this season.
What makes this so surprising is that there was no rhyme or reason to see this coming. This wasn't the 2007 Patriots offense, which rebuilt its receiving corps by adding Randy Moss, Wes Welker and Donte' Stallworth. It wasn't the 2018 Chiefs, who swapped out Alex Smith and Albert Wilson for Patrick Mahomes and Sammy Watkins. We didn't expect those teams to be as dominant on offense as they turned out to be, but there was a plausible path for them to make massive improvements.
This Patriots defense had -- and really has -- no obvious claim to getting significantly better. It was great in Super Bowl LIII, but it finished seventh in points allowed per drive and 16th in defensive DVOA last season. During the offseason, the Pats lost their best pass-rusher, Trey Flowers, and replaced him with Michael Bennett, who is no longer on the team. Their top defensive pick was second-rounder cornerback Joejuan Williams, who has played 21 defensive snaps. They lost defensive playcaller Brian Flores to the Dolphins and still haven't actually named an official defensive coordinator or playcaller.
And yet, after eight games, the Patriots' defense has scored as many touchdowns (four) as it has allowed to opposing offenses. In an era in which the rules have made it easier to score than ever before, the Patriots are historical outliers. This is a defense that has been about as dominant as that legendary 2007 Patriots offense was 12 years ago.
As I've written about in the past, that 2007 Pats attack eventually rewrote football in its image. I'm not sure this defense will be quite as influential, if in part because there's no single clear factor driving its success. The Patriots aren't doing anything revolutionary. As easy of an excuse as it has been for detractors, they haven't just faced an easy slate of opposing passers, either. They are out-executing the rest of the NFL. Their version of the future is to be a smarter, faster version of the present.
Let's run through what the Pats have done and how they've pulled this off, as well as what might change in the weeks to come:
Jump to a question:
How has this defense been built?
How dominant is it now?
Wait ... who is in charge?
Is Belichick ahead of the curve?
NE has played bad QBs ... right?
Will it make a long-term impact?
Can it really keep this up?
How dominant is this defense?
Start with keeping things as simple as possible. When you strip out the three return touchdowns the Patriots' offense and special teams have allowed this season, this defense has allowed 43 points over eight games. That's 5.4 points per game. The data can get a bit iffy moving backward, but if we look back toward the AFL-NFL merger in 1970, no NFL defense had ever allowed fewer than 6 points per game through the first eight weeks of the season.
The Patriots are also doing this in a very scoring-friendly era. If we standardize New England's performance, its defense is three standard deviations below the mean in terms of points allowed per game. That's the simplest definition of a historical outlier, and it's the best any defense has ever posted since the merger over the first eight weeks of the season. No team has outplayed the league's other defenses more than these Patriots have through Sunday's victory.
In my blind comparison column from Thursday, I used standardized score to compare the Pats' defense to the 2007 Patriots' offense. Week 8 was arguably the best game of the year for those Pats, a 52-7 win over Washington. After that game, the 2007 Patriots were 3.1 standard deviations above the mean on offense. Only three post-merger offenses outperformed that offense through eight weeks, and two of them came after 2007.
I'm not going to spend the entire column running the numbers on the Pats' defense. Here's the last thing I'll throw out there, a short list of some of the many categories in which New England's defense leads the league this season. I'll also include the league average for the category:
Passer rating: 40.6 (average: 92.1)
Interception rate: 7% (average: 2.3%)
Pressure rate: 32.1% (average: 27.1%)
Yards/attempt: 5.1 (average: 7.4)
Points/drive: 0.5 (average: 1.9)
Score percentage: 7.6% (average: 35.1%)
Red zone TD percentage: 14.3% (average: 54.9%)
Third-down conversion percentage: 15.6% (average: 39.3%)
Sacks: 31 (average: 18.8)
Interceptions: 19 (average: 5.9)
What has spurred the success on the personnel side?
One often-overlooked factor for team success is health. The Patriots ranked in the middle of the pack for defensive adjusted games lost across four of the past five seasons preceding 2019. The one exception was 2016, when they were the league's healthiest defense. In the process, they held opposing offenses to just 15.6 points per game. Among New England defensive regulars, the only players to miss time via injury are Hightower and safety Patrick Chung, who have combined for three absences. Hightower alone had missed 25 games in his eight-year career before 2019. (Linebacker Kyle Van Noy missed the opener because his wife was giving birth.)
Some teams would use their healthy start to the season to justify keeping the same players on the field for more than 90% of the possible snaps on their side of the ball. The Rams, as an example, basically wanted to run the same 11 players out on every offensive snap in 2017 and 2018, goal-line situation aside. Most defenses around the league will run their secondary and top linebackers onto the field for every single snap or close to it. Buffalo's excellent defense, as an example, has had five players suit up for 93% or more of the snaps this season.
Owing in part to their leads, the Patriots haven't let a single defender play more than 90% of the snaps. Only two defenders have topped 80% of New England's snaps, star cornerback Stephon Gilmore and veteran safety Devin McCourty. Twenty-one players already have racked up 100 or more defensive snaps for the Patriots. Most teams would only hit that figure by midseason if they were riddled with injuries. The Patriots are doing it by choice, and their emphasis on rotations is working.
The team's third-most-used defender has to be the biggest surprise for this Patriots defense. He also has been their most valuable addition of the offseason by a considerable margin. Jamie Collins was unceremoniously shipped out of town by the Pats in 2016 amid reports that the linebacker was prone to bouts of freelancing and wanted a significant contract. Collins got that deal with the Browns, struggled to stay healthy and then lost his role as a three-down linebacker before being released this spring. He spent two months on the free-agent market before New England signed him to a one-year deal worth up to $3 million if he hits all of his incentives, which includes making the All-Pro team. The 30-year-old might very well be on his way.
Collins has turned around his career in New England, excelling at virtually everything they have asked him to do. They have employed Collins as a blitzer and thumping run defender, and he has responded with six sacks and three tackles for loss against the run. He already has a career-high three interceptions, and as the nearest defender in pass coverage, NFL Next Gen Stats have Collins allowing a passer rating of just 41.7 to opposing quarterbacks, the second-best mark in the NFL -- behind San Francisco's Kwon Alexander -- among linebackers with 100 or more coverage snaps. Alexander has been phenomenal, but the 49ers are paying the former Bucs starter just over $14.5 million this season. Collins has been one of several Defensive Player of the Year candidates on this roster at a fraction of the cost.
Continuity and experience within the scheme have really helped the Patriots. Each of New England's 15 most-frequently used defensive players has spent multiple seasons with the team. Ten of those 15 have spent at least three years in New England. Most of the newcomers were along the defensive line, where the responsibilities of playing in Bill Belichick's complicated coverage schemes are minimal. The only defensive back new to the organization who has played more than 5% of the defensive snaps is backup safety Terrence Brooks.
Who is in charge, exactly?
Well, you know the answer. Belichick is in charge, although the Patriots seem loath to admit it. After Matt Patricia was hired as Lions coach during the 2018 offseason, the Pats promoted Flores to defensive playcaller. Flores lasted one season before volunteering his services to whatever the Dolphins are doing this season. It seemed likely that the Patriots were going to hire longtime Belichick confidant Greg Schiano and install him as their defensive coordinator, but Schiano resigned from the team's staff in March.
After Schiano's decision, the Patriots never ended up announcing who, exactly, was going to be their defensive coordinator. That's not necessarily a surprise, given that Belichick has gone through seasons in the past without naming an offensive or defensive coordinator. Usually, though, it's pretty clear to see who is calling plays for a team on either side of the football.
The Pats have gone to great lengths to obfuscate their playcaller's identity, with Belichick and his players repeatedly refusing to answer the question in interviews. The two defensive coaches with play sheets on Sunday against the Browns were safeties coach Steve Belichick, Bill's son, and former standout linebacker Jerod Mayo, who is in his first season as Patriots inside linebackers coach.
Players admitted Mayo made some of the calls during the preseason, and both the younger Belichick and Mayo have relayed calls into the defense at different times. Mayo could turn out to be a fantastic coach, but it's not likely that Bill Belichick has turned over the defensive decision-making to Mayo in his first year as a coach at any level. My guess -- and the most common opinion I heard in asking both coaches who have played against the Patriots and media members who cover the team -- is that Belichick is making most of the decisions here.
The easy story would be to imagine that he has focused his efforts on building a dominant defense, turned the offense and the personnel decisions mostly over to longtime confidants Josh McDaniels and Nick Caserio and built some absolutely breathtaking force of nature in the process. There's probably some slim truth to the premise, and any defense is going to be better with Belichick solving its problems, but if his attention were all the Patriots needed to field the best defense in football, it would have happened a long time ago.
Has Belichick installed concepts to get the Patriots ahead of the curve?
I reserve the right to be told I'm wrong by one of the many people on this planet who know more about defense, but I don't think there's anything about this defense that hasn't been seen in the annals of professional football. The talent, experience and communication Belichick has been able to rely upon has allowed him to be more creative and daring with his decisions, and that has helped this Patriots defense create takeaways.
A great example here is Belichick's ability to threaten opposing quarterbacks with Cover Zero looks, which means that the Patriots will be in man coverage with no safety playing zones behind. They have been able to crowd the line of scrimmage, fill up the box to take away the numbers advantage for running opportunities and then either send a big blitz or drop off into coverage while sending selected pressure to where the opposing protection is weakest. If a defense convinces the opposing quarterback that it is sending six rushers and then sends three he was expecting and one he wasn't, the defense ends up with a quarterback who might just be seeing ghosts.
As ESPN's Louis Riddick pointed out during the beatdown of the Jets, though, Belichick isn't reinventing defenses here. Simulated pressures and Cover Zero blitzes are nothing new to the NFL. Belichick isn't even using them as frequently as some other teams around the league. I can use NFL Next Gen Stats as a proxy here and note that New England has lined up without any high safeties on 75 pass plays this season, which is the fourth-highest total in the league, behind the Ravens, Chiefs and Vikings.
The difference between the Patriots and those defenses is that they almost never get burned without safety help deep. The Pats have allowed a passer rating of just 23.5 in zero-high coverage in a league in which no other team is below a rating of 50.0 and the league average is 104.4. No team has blitzed more frequently without safety help, which has led to eight sacks and 17 hurries out of the look, second behind the Vikings.
Ryan Clark, who picked the Browns to win the AFC North, admits that he at least expected the Browns to perform at baseline, whereas instead they're coming in below their averages.
Belichick can do all of this, of course, because he has supreme confidence in his secondary. No other team has Gilmore, who is the consensus best cornerback in football; he's allowing a passer rating of just 35.8 in coverage this season despite often covering the opposing team's top receiver. Only three cornerbacks have been targeted more frequently, suggesting that Belichick tries to steer opposing quarterbacks into throwing at his best defensive player. It doesn't go well.
Cornerback Jason McCourty has improved in his second year with the Patriots, while Jonathan Jones -- who was on the bench for most of the second half of 2018 -- has surprisingly turned into one of the league's best slot cornerbacks. In Chung, Devin McCourty and Duron Harmon, the Patriots have a trio of safeties they trust in coverage. More than anything -- and I realize how soul-crushingly on-brand it is to say this -- Belichick trusts his defensive backs to do their jobs without blowing any assignments. Snap after snap, they do.
The biggest play they've allowed this season was the 64-yard touchdown pass to Golden Tate in the Giants game, and that was on solid coverage from Jones in the slot. Rookie quarterback Daniel Jones made a great throw, Tate did the veteran thing of waiting for the last moment to try to catch the football, and Jones just didn't get turned around quickly enough.
If a team doesn't have Belichick's secondary or some close approximation, it can't trust its defensive backs to hold up without safety help 10 times per game without getting burned frequently enough to regret it. What we're likely to see, though, is defenses continuing to blend and blur the line between man and zone coverage to try to trick quarterbacks who are depending on differentiating between one or the other pre-snap to make post-snap decisions. Belichick, unsurprisingly, was at the forefront of building pattern-matching coverages that can play like zone or man depending on the route combinations the offense tries to run, having built one such tactic with Nick Saban when the two were on the Browns staff in the 1990s.
Everything has to work together. Take Sam Darnold's first pick in Week 7. The Jets come out in an empty set, and the Patriots respond by showing man coverage across the board with no safety help while crowding the line of scrimmage. Darnold tries to diagnose the blitz that's coming, but a late move by Collins inside of John Simon blows up the Jets' protection and ends with Collins coming in as a free runner.
Darnold knows he wants to go to his hot read, which is going to be Jamison Crowder on a quick out against Jason McCourty. Throwing a quick out against a big blitz in man coverage isn't a new idea; the Packers hit Marquez Valdes-Scantling for a long touchdown on a hot route against a big blitz the day before. If you watch Devin McCourty, though, he's never even playing man coverage. He doesn't run with Le'Veon Bell on a go route because he knows Darnold's going to face an unblocked pressure and throw to his hot read. When Darnold's throw is off target, McCourty has an easy pick.
Without the well-timed pressure, Darnold doesn't have to throw hot and might not sail the throw. Without McCourty's experience, he's going to run with Bell, and that throw falls harmlessly incomplete. Put everything together and you create an interception. The Patriots have put everything together 19 times in eight games.
Well, aren't they just beating up on bad quarterbacks?
Yes, the Patriots have played an easy slate of opposing passers. They've faced, in order, a possibly compromised Ben Roethlisberger, Ryan Fitzpatrick, Josh Rosen, Luke Falk, Josh Allen, Matt Barkley, Colt McCoy, Jones, Darnold and Baker Mayfield. Any defense is going to look better against that group of quarterbacks. If you're using that list to mildly discount the Patriots' performance, that's one thing. To disqualify the Patriots from contention as a dominant defense because of that schedule, though, is naive.
Plenty of teams play bad quarterbacks. This is probably not the worst slate of quarterbacks any team has seen since the merger, and yet, the Patriots have topped every other defense over that time frame by standard score.
To prove that the Patriots aren't simply taking advantage of bad quarterbacks, I built a simple example. I took every one of those quarterbacks above, measured their performance against teams who aren't the Patriots this season, and then plugged in that prorated performance to serve as a baseline for how they should have performed against the Pats.
Instead of Allen's 28 dreadful pass attempts for 153 yards and three picks against the Pats, the baseline for Allen's non-Patriots performance across the rest of the season includes a much more efficient performance, with him going 17-of-28 for 192 yards with a touchdown and an interception. Each quarterback has a similar line thrown into the "Expected" column below. In the case of Barkley and McCoy, who have only suited up against the Patriots this season, I just stuck with their numbers against Belichick's defense.
The difference shouldn't require much interpreting:
Against everyone who isn't the Patriots, these quarterbacks have been pretty bad, playing something like Jay Cutler's final season with the Dolphins in 2017. Against the Patriots, they've collectively turned into Nathan Peterman. These two things are not the same. If you're writing off this defense because of the quarterbacks it has faced, you're not paying attention.
Will this defense be as influential as the 2007 Patriots offense?
It will be hard to really shift the league as much as that Patriots offense did in the following years, if only because teams are smarter now. Even a bad offense could benefit from passing the ball more frequently and trading inefficient running plays for low-risk completions, something that was obvious to both the nascent analytics community and anyone who watched college football. There isn't the same sort of obvious trade-off at hand with this New England defense.
In a few ways, though, I wonder if this will end up either as an example of trends that already exist or as an indirect argument for other ideas. The league is already becoming more and more trade-happy, and I wonder if other teams will forgo the traditional draft-and-develop model to focus more on alternate paths of player acquisition. Belichick's propensity for using draft picks on defensive backs who don't develop is a punchline at this point, but he has gathered talent on defense through other routes. Van Noy, Jason McCourty and nose tackle Danny Shelton were acquired via trade. Gilmore was a big-ticket free agent, but Collins, Chung, Simon and defensive lineman Shilique Calhoun were signed on the cheap after struggling elsewhere.
Studies have also begun to suggest that pass coverage is a more reliable and productive way to stop opposing passing attacks than the pass rush. My own research has suggested that a pass rush's success rate in the first half of the season has little relationship to its performance during the second half of the season.
The Patriots have repeatedly been comfortable moving on from stars like Flowers and Chandler Jones while simultaneously investing in their secondary, both with big contracts and through the draft. The edge-rushing market has continued to rise in recent years, but the top end of the cornerback market has been relatively flat since Josh Norman signed with Washington in 2016. Teams could follow the Patriots' lead and focus their investments in the secondary as the league continues to grow more and more pass-happy.
With those passing games becoming more successful, the NFL is different than it was 10 years ago. Even while some coaches remain stuck in the old days, field position has never meant less and possession of the football has never meant more. Punting from midfield to try to force Patrick Mahomes to go 95 yards is a waste of time and just gives Tyreek Hill and Mecole Hardman more space to run go routes.
In 2006 -- the final year before the 2007 Patriots rebellion -- offenses went three-and-out 36% of the time and scored on just under 31% of their possessions. In 2019, those numbers have flipped; offenses are going three-and-out 31.1% of the time while scoring on 35.1% of their drives. Many of the league's more analytically inclined teams have noticed the shift and begun to go for fourth downs deep inside of their own territory, even before their backs are against the wall.
While there's an impact on game management, the changing league might also influence how teams plan their defenses. With interceptions harder to come by and more valuable -- given that they flip possession -- than ever before, teams might value players differently. The Rams built their secondary with a similar focus, and the Patriots already have been tagged as a bend-but-don't-break philosophy when that's a myth, but I wonder if New England will encourage teams to pay for defensive backs with ball skills.
Can this defense keep it up?
It's difficult to imagine any defense being this good over a full 16-game season. Even the 2007 Patriots offense wasn't quite as dominant during the second half of the season; Tom Brady & Co. averaged 41.4 points over the first eight games and fell off to a lowly 32.3 points per game over the final eight contests.
The Patriots will deal with injuries, because those are inevitable. I mentioned that pass rushes are wildly inconsistent from one half of a season to the next, and the Pats rank first in pressure rate (32.1%) and second in sack rate (9.9%). Both of those marks are likely to decline over the second half. History suggests they also probably won't intercept 7% of opposing passes. Everybody drops interceptions, including Gilmore, who said he wouldn't be able to sleep Sunday night after letting a would-be Mayfield pick slip through his hands.
The quarterbacks will also get tougher. Over the next few weeks, New England will go up against some of the league's best young quarterbacks. Belichick will have to field a more physical defense against Lamar Jackson and the Ravens. After the bye, it will go up against Carson Wentz, Dak Prescott, Deshaun Watson and Mahomes. That's a brutal five-week span, and even if you think the Patriots will hold their own and drag those quarterbacks down from lofty heights, their baseline is still a lot higher than those of Fitzpatrick or Falk.
The Pats probably won't be extreme outliers over the second half of the campaign on defense. They're still likely to rank among the league's best defenses, though, and the offense might be able to make up for any slips. The Pats' offense ranks a middling 20th in points per possession over the past five games, and there doesn't appear to be any second gear coming from Sony Michel, but Belichick addressed a point of weakness by trading for Mohamed Sanu, and receiver N'Keal Harry and offensive lineman Isaiah Wynn should be coming off injured reserve in the weeks to come.
Brady has not been impressive over that stretch -- his passer rating (83.8) and Total QBR (45.3) are both behind that of Gardner Minshew since Week 4 -- but the Patriots have a habit of figuring things out on offense as the year goes along. Wynn's return is critical for an offensive line that hasn't been able to reliably protect the 42-year-old Brady. The Patriots might also dip back into the trade market before Tuesday's deadline.
With competitors such as the Chiefs and Texans slipping up in recent weeks, the 8-0 Patriots are basically locks to stay home throughout the AFC playoffs. With no one-loss team in the conference, ESPN's Football Power Index (FPI) gives the Pats a 92.5% chance of finishing with the top seed in the AFC.
According to FPI, the Patriots are more likely to go 16-0 (9.3%) than fail to come away with the No. 1 seed in the conference (7.5%). And after years of relying on Brady and a transcendent offense to carry them through the postseason, Pats fans have been able to enjoy one of the best defenses in league history for the first half of the 2019 season.