Skip to main content
Menu
The New-Look Patriots Are Still Finding Ways To Win

Editor’s note: The Steelers-Titans game, which we originally highlighted for Week 4, has been postponed because of positive COVID-19 cases.


Three weeks of play are in the books for the post-Tom Brady New England Patriots, and the early results have been … well, still relatively Patriots-like. New England is 2-1, winning against the Miami Dolphins and Oakland Raiders and losing a nail-biter against the Seattle Seahawks; the Pats are also tied for the eighth-best point differential in the league and tied for second in turnover margin. No, coach Bill Belichick’s team has not fallen apart yet after the GQBOAT took up residence in Florida. But with a huge matchup against the Kansas City Chiefs looming for the Pats on Sunday afternoon, we’re about to truly learn how much of the usual championship business will be conducted in Foxborough this season.

Peeking under the hood, we can see both encouraging signs and potential trouble spots for the Patriots looking forward. On defense, I wondered before the season how much the Pats might slide in 2020, given that regression is natural for great defenses (New England ranked first in schedule-adjusted expected points added per game last year) and the Patriots sustained unusually heavy personnel losses over the offseason. But so far, if we adjust for schedule with a mix of 2019 and 2020 EPA stats for opponents,1 New England’s defense ranks ninth overall this season in EPA per game and fifth against the pass — down slightly from its dual first-place finishes in 2019, though far from a collapse.

Individually, the Pats have struggled to replace some of their best performers from a year ago. Last season, 12 of the team’s 14 qualified defenders2 carried a ProFootballFocus grade of 67.0 or higher; this year, only four of 14 are above that threshold. And teamwide, that shows up with New England allowing a higher completion percentage, more yards per pass and a higher Total Quarterback Rating (QBR). But although opponents are moving the ball more effectively, you could argue the Pats have not suffered a drop-off in defensive execution so far this season.

According to ESPN Stats & Information Group’s win rate metrics, which use tracking data to judge how often players “won” their individual matchups, New England defenders are succeeding more often on pass rushes (up from 40 percent to 43 percent) and run plays (up from 32 percent to 35 percent) than in 2019. Stylistically, the Patriots have compensated for their upheaval by playing more zone coverage and blitzing less — yet they’ve managed to generate pressure on a greater share of opposing pass plays. And despite the 72 percent accuracy rate they’re allowing on passes, the Patriots have forced opposing quarterbacks into throws that have a lower expected completion percentage this season (62.9 percent) than last (64.0 percent), according to the NFL’s Next Gen Stats.

The Patriots are executing better this season

Win rates and opponent expected completion percentages for the 2019 and 2020 New England Patriots

Win Rates*
Season Pass Rushing Run Stops Pass Blocking Run Blocking Exp. Comp% Allowed
2020 42.5% 34.7% 60.6% 72.8% 62.9%
2019 40.0 32.0 57.7 68.3 64.0

*ESPN’s win rate metrics use NFL Next Gen Stats to track how often players “win” their matchups with opposing players.

Source: ESPN Stats & Information Group, NFL Next Gen Stats

Of course, Patrick Mahomes’s Chiefs currently rank No. 1 in schedule-adjusted passing EPA per game, just like they did in 2018 and 2019. Surely K.C. will challenge the Pats’ defense more than the Dolphins, Raiders or even the Seahawks did — and in the process will provide a real stress test for the changes Belichick and his defensive coaches put in since last season.

Offensively, the Patriots have changed even more radically since 2019. They’ve called for designed runs on 49 percent of snaps, up from 40 percent last season; since 2001, only one Pats team — the 2004 version that saw RB Corey Dillon run for over 1,600 yards — rushed on a greater share of its offensive plays. Quarterback Cam Newton alone has accounted for 33 percent of New England’s runs and 28 percent of its rushing yards, which is by far the most for any Patriots QB since at least 2001.3 Although the Patriots are using pass-friendly sets like shotgun (roughly 51 percent of snaps) and 11 personnel4 (61 percent) at the same or higher rates than last year, they are clearly a run-based team now. Rushing accounts for 44 percent of all Patriots yardage, compared with just 30 percent last season.5

For now at least, it seems to be working. In terms of schedule-adjusted EPA per game, New England ranks 13th on offense, up from 18th with Brady a year ago. The Pats have the most effective rushing attack in the league, adding 6.97 more points on the ground per game than we would expect from an average team facing their schedule. (The Arizona Cardinals are No. 2 at +6.57 EPA per game.) Newton has contributed greatly to that ranking; only Kyler Murray and Lamar Jackson have more rushing yards among QBs this year. But Newton’s passing numbers have quietly been average at best — he ranks just 17th in adjusted net YPA, 23rd in completion percentage versus expected and 24th in passer rating. And as a team, the Pats are 27th in schedule-adjusted passing EPA per game, down from 19th in 2019.

Perhaps those numbers will improve as the season goes on. But if not, this is the big question for New England: In today’s pass-oriented NFL, is it possible to sustain an effective offense based primarily around rushing efficiency? Since 2006, only one team finished a season ranked in the top half of schedule-adjusted EPA offenses despite a passing ranking as low as New England has now: the Michael Vick-led 2006 Atlanta Falcons. Newton certainly can put up impressive rushing numbers in the Vick mold, but it would make the Patriots’ task much easier if his passing numbers were to get better from here.

Still, this season has been vintage Patriots so far, if simply in the sense of Belichick and staff doing an impressive job of adjusting around the team they have, not the one they would prefer in a perfect world. They’ve reworked the offense to fit the strengths of rushers like Newton and RB Sony Michel and schemed around the many defensive departures. They’re still sticking to long-standing Patriots fundamentals, like a good turnover differential and making the opponent work harder for points than they do. Although our model considers New England an underdog (!) to win the AFC East — trailing the Buffalo Bills in division probability, 57 percent to 38 percent — it has done about as good a job of weathering Brady’s departure (and many others) as could be expected. Now we’ll see if it can hold up against its toughest test yet, in the form of Mahomes and the Chiefs.

FiveThirtyEight’s NFL Elo ratings

How each team ranks through Week 3 of the 2020 season, according to our quarterback-adjusted predictions

Chance To …
Rk Team Starting QB QB Rk* Elo Rating Proj. Record Make Playoffs Win Division Win SB
1 Chiefs Mahomes 1 1729 13-3 98% 91% 25%
2 Ravens Jackson 5 1638 11-5 85 45 7
3 Packers Rodgers 6 1636 12-4 90 63 10
4 Seahawks Wilson 4 1629 12-4 87 50 9
5 Steelers Roethlisberger 8 1603 11-5 83 43 6
6 Bills Allen 7 1599 10-6 78 57 5
7 49ers Garoppolo ✚ 24 1592 10-6 64 21 6
8 Patriots Newton 17 1581 10-6 68 38 3
9 Rams Goff 10 1580 10-6 64 21 4
10 Titans Tannehill 11 1580 11-5 80 63 4
11 Saints Brees 3 1577 9-7 58 41 3
12 Bears Foles 21 1571 10-6 76 34 4
13 Buccaneers Brady 20 1558 9-7 62 45 3
14 Cowboys Prescott 2 1554 9-7 73 70 3
15 Colts Rivers 18 1549 9-7 54 29 2
16 Browns Mayfield 26 1507 9-7 46 11 <1
17 Raiders Carr 9 1506 8-8 42 7 <1
18 Cardinals Murray 14 1500 8-8 36 8 1
19 Vikings Cousins 27 1473 6-10 12 2 <1
20 Eagles Wentz 22 1468 5-10-1 18 15 <1
21 Falcons Ryan 12 1460 5-11 11 7 <1
22 Texans Watson 15 1456 6-10 12 6 <1
23 Dolphins Fitzpatrick 13 1451 6-10 16 4 <1
24 Chargers Taylor ✚ 30 1450 6-10 14 <1 <1
25 Lions Stafford 16 1437 6-10 17 2 <1
26 Panthers Bridgewater 19 1423 6-10 13 7 <1
27 Jets Darnold 29 1391 4-12 4 1 <1
28 Bengals Burrow 28 1379 4-11-1 5 <1 <1
29 Jaguars Minshew 25 1379 5-11 7 2 <1
30 Washington Haskins 31 1369 5-11 12 10 <1
31 Giants Jones 23 1366 4-12 6 5 <1
32 Broncos Rypien 32 1343 5-11 6 <1 <1

*Ranking among Week 4 starters, according to our QB Elo ratings.

✚ Starter is currently injured, and may not play.

Simulations as of Sept. 30, 2020.

Source: ESPN

Check out our latest NFL predictions.

Footnotes

  1. With more weight given to 2019 because we’re only three games into 2020.

  2. Playing at least 33 percent of team snaps.

  3. The previous mark was held by Matt Cassel, who had 12 percent of New England’s rushing yards in 2008.

  4. One running back, one tight end and three wide receivers.

  5. The league average this year is a shade under 33 percent.

Neil Paine is a senior writer for FiveThirtyEight.

Comments