Week 4 was a weird one that slowed with some low-quality of games in the late afternoon. We’re now nearly a quarter of the way through the season and we’re getting a clearer picture of what could be trends to keep watching. Let’s dive into Week 4.
All stats listed are provided by TruMedia unless noted otherwise.
1. The Bills and Ravens Traded Scores
Josh Allen and Lamar Jackson are two of the most singularly talented quarterbacks in the NFL. While both offenses can play to the quarterback’s strengths, Buffalo’s 23-20 win over Baltimore highlighted some of the differences between embracing those strengths and relying heavily on them.
Early in the game, the Ravens had an advantage moving the ball. Baltimore had Jackson in empty on 45% of his dropbacks in the first half, which allowed the Ravens to spread the ball around and hit some quick passes and they worked their way down the field.
Jackson averaged 0.20 EPA per dropback in the first half, despite a 5.19-yard average depth of target and 6.75 yards per attempt. After Jackson killed defenses that sent extra rushers over the first three weeks of the season, the Bills blitzed on just 10% of his dropbacks in the first half.
With a run game that finally looked explosive with J.K. Dobbins, Baltimore cruised to a 20-10 halftime lead with that Bills touchdown only coming in the final minute of the half.
But a switch was flipped in the second half. The Bills ramped up some of their two-high coverages, using either Quarters or Cover-6 on nearly half of Jackson’s dropbacks. With those coverages defending deep, Buffalo got a bit more aggressive up front and blitzed on 22% of Jackson’s dropbacks in the second half.
Even Baltimore’s attempted answers didn’t work.
By putting a roof on the coverage and sending extra pressure, Jackson averaged -0.63 EPA per dropback in the second half. A lot of that came from the late fourth quarter interception (we’ll get to that), but even with that excluded, Jackson averaged around -0.47 EPA per dropback.
The Ravens just didn’t have a lot of answers while the Bills controlled the game from the defensive side. Jackson was getting the ball out quickly, but had just a 3.85-yard aDOT in the second half with 2.77 yards per attempt. He did not attempt a pass more than 10 yards past the line of scrimmage in the second half.
It was the opposite for the Bills, who started slow and improved as the game went on. The Bills have been ok with not necessarily starting slow, but they’ve accepted the way defenses will play them in the passing game and Allen has worked the quick game early.
Josh Allen By Quarter, 2022
data per TruMedia
| Quarter | EPA/DB | Comp% | aDOT | At/Behind% | Short% | Intermediate% | Deep% |
|---|---|---|---|---|---|---|---|
| 1 | 0.12 | 75.6% | 4.13 | 33.3% | 53.3% | 8.9% | 4.4% |
| 2 | 0.29 | 70.6% | 7.24 | 19.6% | 54.9% | 17.6% | 7.8% |
| 3 | 0.35 | 56.8% | 6.89 | 22.7% | 47.7% | 20.5% | 9.1% |
| 4 | 0.29 | 64.3% | 10.5 | 17.9% | 39.3% | 32.1% | 10.7% |
That’s led to the start of games being a feeling-out period for the Bills. The short passes do take out an explosive element of the passing game, one Buffalo eventually figures out how to manufacture later in games.
Against Baltimore, the run game played a big part in the second half control. Buffalo has struggled with the run game so far this season. By ESPN’s run block win rate, the Bills ranked 32nd through three weeks. In the first half, Buffalo had eight carriers for 34 yards with a long of nine.
The Bills were 73% pass in the first half. But in the second half, they were 53% run, including Allen scrambles. Allen running more also opened up some holes for the designed run game and the Bills had 17 carries for 91 yards in the second half — good for 0.13 EPA per rush.
With both scrambling and passing, Allen made some plays that perhaps only the two quarterbacks playing in this game could make. And when Allen needed to make a throw, he did.
On the fourth down decision
Tied 20-20, the Ravens faced a fourth-and-goal from the Buffalo 2-yard line with 4:15 left to go in the fourth quarter. John Harbaugh kept the offense on the field and chose to go for it, instead of taking the easy field goal to take the lead.
Multiple win probability models agreed with going for it. After the game, Harbaugh explained the rationale:
Coach Harbaugh on the decision to not kick a FG: pic.twitter.com/Rkapy4Ve5H
— Baltimore Ravens (@Ravens) October 2, 2022
With four minutes to go, a field goal is nowhere near a guarantee to win the game. Buffalo could still drive down the field (something they did) and score a touchdown to take the lead. With a score, the Bills would need to drive down the field just to tie.
Coming up short, you’re banking on setting up the Bills at the 2-yard line and making them drive the entire length of the field. The worst possible scenario would be an interception, which erases the added benefit of field position with an unsuccessful attempt — and that’s what happened.
An interception is probably the least likely outcome if you redo that play, even on a straight dropback that doesn’t use Jackson as a running threat.
2. Seahawks & Lions Played The Best Game You Sort Of Paid Attention To On RedZone
Let’s get this out of the way now… Geno Smith led all quarterbacks in EPA per dropback in Week 4 (0.61) and is currently fourth through four weeks.
In a 48-45 shootout, the Seahawks and Lions put on a show that would have gotten much more attention had it been either in primetime or between teams we as a general football watching collective cared more about.
Seattle and Detroit combined for 1,075 yards of offense, which is the 14th-most in a game since 2010, per StatMuse. That’s more than the combined yards between the Rams and Chiefs in that 2018 Monday Night Football game (1,001).
These were the 29th and 31st ranked defenses by EPA per play entering the week (they are now 31st and 32nd) but that should not take away from some of the things they’re doing on offense.
If the nerd ideal of an NFL offense is deep shots off play-action, then Seattle has somehow become nerd heaven. Against the Lions, Smith used play-action on 56% of his dropbacks while his aDOT was 10.7. He went 13-of-17 on play-action with a 10.8-yard aDOT and averaged 0.56 EPA per dropback.
Smith has always been a good intermediate thrower and that area has been opened up in the offense with him behind center. On the season, Smith is 20-of-26 with a league-leading 1.36 EPA per play on throws between 11-19 air yards.
That was highlighted by Smith’s first touchdown of the game in the first quarter, a 17-yard pass to Will Dissly.
We should also stop to appreciate the Seahawks having three routes going into the end zone and one more to the first down marker on a first-and-15!
Smith had to balance patience and aggression against a Detroit defense that blitzes often and plays a lot of man coverage behind it. The Lions blitzed on 39% of Smith’s dropbacks in this game. The heavy play-action helped some to keep Smith in control inside the pocket. Only 28.1% of Smith’s attempts were within 2.5 seconds of the snap, but 59% were between 2.5 to four seconds, which Next Gen Stats considers “in rhythm.”
The Seahawks’ run game also took off and Rashaad Penny remains an explosive piece of the offense. He had 151 yards and two touchdowns on 17 carries. 23.5% of his carries went for 10 or more yards in the game. For the season, Penny ranks fifth with an 18.4% rate of runs of at least 10 yards among 28 backs with at least 40 carries.
Seattle’s defense has not been good but there have been young pieces that have flashed to give optimism toward the future for once there is more overall talent on the unit. The best example of this might be Tariq Woolen, the 6-foot-4 fifth-round pick cornerback who ran a 4.26 40-yard dash at the combine.
Woolen has made some splash plays and entered Week 4 29th among cornerbacks in Adjusted Yards allowed per coverage snap, which accounts for touchdowns and interceptions. He added a pick-six in this game.
Detroit’s offensive effort in this game should not go unnoticed, especially since they weren’t nearly at full strength. The Lions were missing D’Andre Swift, Amon-Ra St. Brown, D.J. Chark, and Jonah Jackson.
Jared Goff wasn’t running a training wheels offense, either. Goff had a 9.92-yard aDOT and had the fourth-best EPA per dropback of the week. Without the top receiving targets, T.J. Hockenson took over and showed what many had thought he was capable of as an early draft pick.
Hockenson’s 179 receiving yards led all players in Week 4 and 109 of those yards came after the catch. Even with those yards after the catch, Hockenson had an aDOT of 9.0 yards or more (9.25) for just the ninth time in his career.
The Lions’ run game was another factor and has been all season. Even without Swift, the Lions averaged 0.26 EPA per rush and they’re second behind the Browns for the season. Detroit’s 0.19 EPA per rush would be the fourth-best passing game by EPA.
This is another close loss for Detroit, a team that remains a little better in theory than in practice. Under Dan Campbell, the Lions are 2-8-1 in one-score games, that’s the third-worst winning percentage and tied for the most losses in that span.
3. The Eagles Have More Answers
The Eagles and Jaguars entered Week 4 looking like two of the best teams in the NFL. That could still be true, but in a 29-21 win, the Eagles showed they have a bit more depth and a few more countermoves, which gives them the advantage.
After Jalen Hurts and Trevor Lawrence dominated the first few weeks, neither quarterback played particularly well on Sunday.
For Lawrence, the secret had been getting the ball out quickly and controlling the pace of the game. Just behind Lawrence’s quick release was an offensive line that still struggled to pass block. Through Week 3, the Jaguars were 32nd as a team in ESPN’s pass block win rate. That became a problem against the Eagles.
Lawrence was only technically hit three times (though he was sacked four due to a strip sack) but he was under more pressure than he had been in the previous two weeks. On top of the four sacks, Lawrence was 1-for-5 throwing under pressure, including a red zone interception in the third quarter.
The Eagles brought a blitz and Lawrence thought he had a window to hit Christian Kirl down the sideline but James Bradberry charged in front of the pass for the interception.
Bradberry has been one of the most impressive cornerbacks through the early part of the season. Darius Slay has gotten more attention as the “No. 1” but no corner had more coverage snaps entering the week The pair have been among the league’s best, both in the top 10 of Adjusted Yards allowed per coverage snap.
Cornerback Metrics, though Week 3
data per Sports Info Solutions
On the other side of the ball, the Eagles were able to unleash their run game, especially without Foley Fatukasi on the field for the Jaguars. The Eagles averaged 2.0 yards per carry with Fatukasi on the field and 5.4 without him.
Hurts remained a threat in the run game and that opened up opportunities for Eagles backs. On Miles Sanders’s first touchdown run, the backside defender and backside safety were frozen waiting for Hurts to potentially keep the ball. That allowed Sanders to hit an open hole and follow downfield blocks from Jason Kelce and DeVonta Smith into the end zone.
As a passer, Hurts had some ups and downs in the game but the encouraging part is the Eagles had enough talent around the roster to win in different ways. Even Hurts was able to offer something in the run game with 16 carries and a touchdown. This is a complete team and maybe the most complete in the league with the ability to get better throughout the season.
4. The 3-1 New York Giants Are Well Coached And Oddly Fascinating
The 4-0 Eagles don’t even have a massive lead in their own division because the New York Giants are also 3-1 after a 20-12 win over the Chicago Bears.
At 3-1, the Giants might be one of the league’s most fascinating teams based on how they’ve gotten to that record. This is not a good football team but its a well-coached team and Brian Daboll has gotten players into good positions.
We can focus on the quarterback to start. Daniel Jones has not been impressive throwing the ball but he’s been one of the most dangerous runners and that’s been such a big piece of the offense. When Jones injured his ankle late in the game, he was pulled for what Jones believed was an inability to execute the rollouts in the game plan.
Daboll didn’t want the quarterback taking too many straight dropbacks. Again, this team is 3-1. When Jones had to return to the game after Tyrod Taylor was forced out with a concussion, the Giants ran snaps with Saquon Barkley as a wildcat quarterback.
That brings us to the run game, which between Jones and Barkley, has been quite good. The Giants have also relied on it. Including scrambles, the Giants had a 72% run rate against the Bears, easily the highest in the league this week. The last time the Giants had a run rate of 70% or more was last season’s game against the Bears with Mike Glennon at quarterback. These are very different circumstances.
On the season, the Giants have the sixth-highest run rate as one of six teams over 50%. Jones had two long runs into open space off bootlegs in the game. Jones was among the leaders in EPA per dropback at 0.22 in Week 4. But on just pass plays (without scrambles), that dropped to -0.10.
The Giants are pushing a lot of the right buttons. It might not consistently lead to more wins in 2022 — though the schedule is not hard — but it should give the organization a lot of faith in the direction it’s going for the first time in a while.
5. It’s time for the rookie QBs
Kenny Pickett saw his first game action in the Steelers’ 24-20 loss to the Jets. Pickett was put in after Mitchell Trubisky averaged -0.59 EPA per dropback, which included three sacks and an interception.
Pickett came in a did a bit of the Pickett experience. He threw the ball much deeper than he did in the preseason (and in college) but that wasn’t always for the best. Pickett’s first pass was a deep floater to a well-covered Chase Claypool that was deflected and intercepted.
The Steelers had 10 days to prepare for this game, which could have given Pickett more than a week to get starter reps. But the team wanted to stick with Trubisky, only to make a change mid-game.
Mike Tomlin did not declare Pickett the starter and still could go back to Trubisky with Buffalo, Miami, Tampa Bay, and Philadelphia next on the schedule before a Week 9 bye. But we know what Trubisky is. There is no guarantee Pickett is an upgrade, but that’s a leap the Steelers need to take sooner than later.
There should be a similar discussion happening with Atlanta. The Falcons pulled out a 23-20 win over the Browns but that was greatly aided by the run game. Atlanta continues to do some fun things on offense and in Week 4 that was creating some opening for Tyler Allgeier and Caleb Huntley, who combined for 20 carries, 140 yards, and a touchdown.
Marcus Mariota went just 7-of-19 passing in the game. Desmond Ridder is athletic enough to keep some of the quarterback run game and he should be accurate enough to feed the ball to Drake London and Kyle Pitts. Mariota is 14th in EPA per dropback and while his flashes have been fun, Atlanta should see what else they have at the position.
6. The Packers Still Have Some Things To Work Out
The Packers escaped with a 27-24 win over the Bailey Zappe-led Patriots. They did not look good while doing it. The good news is the Packers are 3-1 with a little margin for error to figure things out as they go. But they need to figure some things out soon.
By EPA per dropback, this is the worst first four games for Aaron Rodgers since the 2009 season when he also had -0.02 EPA per dropback.
Aaron Rodgers EPA per dropback through Week 4, 2011-2022
data per TruMedia
| Season | EPA/DB |
|---|---|
| 2020 | 0.45 |
| 2015 | 0.24 |
| 2014 | 0.21 |
| 2011 | 0.18 |
| 2013 | 0.16 |
| 2016 | 0.15 |
| 2017 | 0.10 |
| 2019 | 0.10 |
| 2008 | 0.10 |
| 2021 | 0.09 |
| 2018 | 0.05 |
| 2010 | 0.00 |
| 2012 | 0.00 |
| 2009 | -0.02 |
| 2022 | -0.02 |
There have been some plays that could be fixed just from error alone, such as drops that have plagued receivers early in the season. But there are also structural issues. Green Bay is playing Elgton Jenkins at tackle when they should keep him inside — potentially making two positions on the line worse by playing him there.
The Packers also haven’t figured out their Pony personnel package with Aaron Jones and A.J. Dillon on the field at the same time. On 14 plays against New England, Green Bay averaged 4.0 yards per play and -0.34 EPA per play. The problem is Jones has been the better runner but he’s also the better receiver, which puts him out wide when the two backs are on the field together. Green Bay also has a 52% run rate with Jones and Dillon together.
The defense also hasn’t lived up to its preseason hype. They’re 11th in EPA per play on defense but still struggle against the run (27th), which was the problem last season and the reason for many of the defensive moves the team made in the offseason.
7. Titans Opening Drives
Two teams have scored on their opening possession of the game in all four weeks, the Las Vegas Raiders and Tennessee Titans. The Titans are the only team to score a touchdown on all four opening drives.
In a 24-17 win over the Colts, the Titans’ opening drive started on the Indianapolis 32-yard line after a Denico Autry strip sack of Matt Ryan, It took Tennessee five plays to reach the end zone on a 7-yard pass from Ryan Tannehill to Robert Woods.
The Titans are the first team since the 2011 Patriots to score an opening drive in four straight games to start the season. New England punted on the first drive in Week 5.
8. Chart of the day
The Cowboys have figured out how to use CeeDee Lamb. Over the past two weeks, Lamb has the highest target share in the league at 35.7%. Lamb has yet to have a monster game, but the usage has improved. Against the Commanders, Lamb was used in motion and consistently put in favorable matchups in zone coverage. He had an open crosser with space against zone and on his touchdown, linebacker Cole Holcomb was forced to chase him. With Dak Prescott potentially returning soon, getting Lamb’s role in the offense figured out will be a huge boost.
9. Play of the day
The Tampa Bay Buccaneers entered the week as the best defense in the league and that just didn’t matter to Patrick Mahomes and the Kansas City Chiefs. Kansas City averaged 3.73 points per drive and 1.32 EPA per drive in Week 4. This insane Mahomes touchdown pass explains everything.
10. The Return of the Run Game
There’s been a lot of talk about the run game in this column — more than ever — but there is a reason. The run game is making a bit of a comeback. For the first time this season, the run game hit positive EPA (or at least not negative) overall.
EPA By Week, 2022
data per TruMedia
| Week | EPA/Play | EPA/Drive | EPA/DB | EPA/Rush |
|---|---|---|---|---|
| 1 | -0.01 | -0.07 | 0.01 | -0.07 |
| 2 | 0.00 | 0.00 | 0.03 | -0.03 |
| 3 | -0.03 | -0.15 | 0.01 | -0.05 |
| 4 | 0.00 | -0.02 | 0.02 | 0.00 |
This also blends with the overall trend of passing efficiency being down across the league from prior years as defenses are doing better at stopping the big play. Passing efficiency through four weeks is at its lowest over the past 10 years. Meanwhile, rushing efficiency is at its highest. Rushing still brings negative EPA, but we currently have the smallest gap between passing and rushing that we’ve seen in a while — and for this, scrambles are considered dropbacks.
League EPA Through Week 4, 2013-2022
data per TruMedia
| Season | EPA/Play | EPA/Drive | EPA/Dropback | EPA/Rush |
|---|---|---|---|---|
| 2013 | -0.02 | -0.12 | 0.04 | -0.11 |
| 2014 | 0.04 | 0.22 | 0.11 | -0.08 |
| 2015 | 0.00 | -0.01 | 0.06 | -0.09 |
| 2016 | -0.01 | -0.07 | 0.06 | -0.12 |
| 2017 | -0.01 | -0.07 | 0.03 | -0.08 |
| 2018 | 0.00 | -0.01 | 0.06 | -0.10 |
| 2019 | 0.00 | -0.01 | 0.06 | -0.08 |
| 2020 | 0.02 | 0.13 | 0.06 | -0.06 |
| 2021 | 0.02 | 0.14 | 0.08 | -0.06 |
| 2022 | -0.01 | -0.06 | 0.02 | -0.04 |
We have a number of teams excelling at the run game this season with six putting up at least a 50% rushing success rate. It’s a necessary change up for how defenses are trying to play and the nature of having that extra element without being one-dimensional could be a big factor for these teams throughout the rest of the season.
