Tom Gower

Tom Gower has written about the NFL and its numbers for Football Outsiders since 2009.

Saving the Packers

3 Comments

It seemed reasonable to expect the Green Bay Packers to have one of the best offenses in the NFL this year. Jordy Nelson‘s August ACL tear hurt their prospects, but they still had Aaron Rodgers, Randall Cobb, Eddie Lacy, and one of the league’s better offensive lines. Instead, they have been shockingly mediocre. What do the numbers suggest, and is there any reason to expect the Packers to be better in Week 17 against the Minnesota Vikings and in their playoff game the following week?

The Trend Is Discouraging

There is no doubt their Week 16 offensive performance against the Arizona Cardinals was miserable. By DVOA, Football Outsiders’ play-by-play metric that adjusts for game situation and opponent quality, it was the second-worst of Rodgers’ tenure as the Packers’ starter. Only his three-interception performance in Tampa back in 2008, in his first season as a starter, is worse, and just barely (minus-56.2 percent versus minus-55.2 percent).

If that was just a one-off, we could write it off. It was not. The Packers have posted a below-average DVOA on offense in four of their past five games. Coach Mike McCarthy taking the play-calling reins back from offensive coordinator Tom Clements worked against Dallas, but not against Oakland and certainly not last week.

Some week-to-week variation is to be expected, but Green Bay’s three best offensive performances of the season by DVOA came in Weeks 1, 2, and 3. Since then, they are 21st in DVOA, just ahead of the Browns (minus-6.3 percent compared to minus-7.2 percent). This is not the company the Packers expected to be keeping.

Offensive Line Issues Are Part of the Problem

In 2014, the Packers had an excellent, consistent offensive line. From left to right, David Bakhtiari, Josh Sitton, Corey LinsleyT.J. Lang, and Bryan Bulaga started 79 of 80 games. All five returned, but not for every game. Sitton is the only one with 15 starts, and he has been worse than his near-flawless performance last year (0.0 blown pass blocks per Football Outsiders’ charting). The others have all missed time, and Green Bay’s offensive line depth has not been up to the challenge.

The line issues have shown up in both the run and pass games. It seemed like Rodgers turned a corner in 2014, going from a quarterback sacked more than most to one adept at avoiding sacks. His adjusted sack rate, which accounts for down and distance and opponent was just 5.3 percent last year and is 7.3 percent this year. By offensive line yards, which attempt to account for the distribution of yards between the back and the line, they went from a top-eight unit the past two seasons to 22nd. That includes fifth-worst in stuffs, those runs for no gain or a loss.

Receiving Depth Was Tested and Flunked

The loss of Nelson was potentially devastating because Rodgers was extraordinarily reliant on just two receivers last year, both by volume and for big plays. Both Nelson and Cobb caught over 90 passes and did so with great efficiency. Cobb ranked first in DVOA last year, while Nelson was eighth. No other Packers receiver or tight end came out better than average.

The story is similar this year. James Jones, reunited with Rodgers after the Giants cut him, is the only Packers wide receiver or tight end with a DVOA better than 0.0 percent on 25 or more targets. Green Bay’s young players have not developed the same rapport with Rodgers as Cobb and Nelson.

Davante Adams, praised to the skies by Rodgers in the offseason, ranks next-to-last in DVOA among qualifying receivers. The other young player who saw a big uptick in his role, tight end Richard Rodgers, is merely below average.

The more puzzling case is Cobb, who is below average by DVOA after he topped the league last year. His yards per catch have fallen from 14.1 to 10.8. Some decline in his YAC was expected; he had a number of long touchdowns last year, more than was sustainable. But Rodgers has found him more than 20 yards downfield on just six targets this year, even on one of his customary extended plays.

Cobb has seen his usage change. In 2014, he was largely a short-area receiver, with 38 percent of his targets between one and five yards downfield and was rarely targeted at or behind the line of scrimmage. This year, his distribution is much more even, with between 23 and 32 targets at or behind the line of scrimmage, from 1-5, 6-10, and 11-15 yards downfield. That looks a lot more like how Nelson was targeted last year. But as good as Cobb is, he is not Nelson, and trying to force him to be him has not worked and might have taken Cobb away from what he does best.

What Can Be Done?

McCarthy took back play-calling duties because he wanted the offense to work differently. The Packers have tried to run the ball more since, but they haven’t had much success outside of the Dallas game. Yes, Lacy and James Starks have been part of the problem, or at least not the answer. Both have been efficient receivers this year, though, and a pass game oriented more towards the backs and short passes could be more successful than the receiver-driven offense McCarthy has preferred. It is too late to make major changes, though, so only incremental modifications are likely. Green Bay’s best chance to realize the promise of their 6-0 start is to rely on Rodgers’ improvisational ability and a defense that outside of the Broncos game has played at least reasonably well.

Is Sam Bradford suddenly … better?

Leave a comment

When we last visited Sam Bradford‘s performance as Philadelphia Eagles quarterback, it wasn’t at all clear that the decision to trade Nick Foles for him was correct. Chip Kelly recently doubled down on Bradford’s performance, declaring he saw improvement on a weekly basis that got Kelly excited. What might Kelly be seeing, and what context can we add to Bradford’s performance?

The Contrast with Sanchez

One thing we’ve seen lately that we didn’t see earlier in the season was the Eagles’ 2015 offense with another quarterback. Mark Sanchez started two games and finished a third as Bradford suffered a shoulder injury and a concussion. While playing half the season in 2014, Sanchez was, on the whole, roughly as good as Foles was. In his playing time this season, Sanchez has been a significant downgrade from Bradford. Table 1 shows the statistical details, using adjusted net yards per attempt (ANY/A) and Football Outsiders’ metric DVOA.

Table 1. Mark Sanchez vs. Main Starters, 2014-15

Year QB DVOA ANY/A
2014 Foles 1.8% 5.93
Sanchez -1.4% 6.18
2015 Bradford -11.3% 5.50
Sanchez -43.8% 4.62

Two specific areas stand out as major issues for Sanchez. First, he has continued the trend from most his career of throwing interceptions at a rate above the league-average while Bradford has taken better care of the football. Sanchez’s interception rate is 4.4 percent, compared to Bradford’s 2.8 percent.

Second, the Eagles’ offensive line has been much more problematic for Sanchez in pass protection. His adjusted sack rate, as calculated by Football Outsiders, is 9.0 percent, the seventh-highest rate of the 40 passers with at least 100 dropbacks this season. Bradford’s, by contrast, is just 4.9 percent, the tenth-best rate in the same group, even though he started both games the Eagles played without Jason Peters. Quarterbacks can do a lot to control this number, as Denver’s similarly contrasting numbers with Brock Osweiler and Peyton Manning and many other examples show, and Bradford is helping the line in a way Sanchez is not.

This contrast suggests that even though Bradford’s production is down from what Foles did in 2014, this could be the result of broader offensive issues which Bradford is doing well to mask.

Bradford Is Playing Better

Bradford’s numbers since returning from injuryare better than the ones he had earlier in the season. His DVOA earlier was -17.9 percent. Since then, it is 2.5 percent, better than the league average. His game against the Bills was his worst in the last four contests; on a per-play basis, it would have been his third-best performance in the first seven games.

One other thing Bradford is not doing is running. He has just one actual rushing attempt in the past couple games, while most of his nominal carries have been aborted snaps or kneeldowns. This is important, because he was not an effective runner when the Eagles tried to use him as one in the option game.

Finding the Range Downfield

The biggest development in Bradford’s game of late has been the deep pass. His performance on short throws has been roughly the same as it was earlier in the season, with a DVOA of -1.9 percent compared to -0.2 percent over the first seven games. If anything, he has been less consistently productive on them, posting a success rate of 39 percent compared to 47 percent earlier. Only throwing fewer interceptions has saved the numbers from being worse.

Where Bradford has improved the most has been on deep throws. This was where the great Foles-led offense of 2013 made most of its hay, and it was a big struggle for Bradford earlier in the season. He had the sixth-worst success rate of the 32 passers with the most deep attempts earlier in the season. Over the past six weeks, he has the second-best success rate, 60 percent, and the third-best DVOA among the 32 passers with the most deep attempts.

The easy response is to say that this is a fluke due to a small sample size, the product of a limited number of games and passes. What makes Bradford’s performance so encouraging, though, is he has been doing it at multiple levels. His numbers are better on throws 11-15 yards downfield, 16-20 yards downfield, and even that 21-25 yard area where he did not have a single completion on 18 attempts in the first seven games, while he has continued to enjoy success on very deep throws like his 53-yard touchdown pass to Nelson Agholor last week.

Going Forward

The numbers suggest Kelly was not just blowing smoke, that Sam Bradford has been playing better of late and may well be doing much better than Nick Foles would have done for the 2015 Eagles. Kelly still has a problem, though: the Eagles are still a below-average offense. Looking just at the last four games Bradford has played, they are 20th in DVOA and just 17th in passing DVOA. Even if Bradford is in fact the answer, Philadelphia still has offensive problems to address.

How great is J.J. Watt?

2 Comments

Despite playing with a broken hand, J.J. Watt, the Houston Texans’ All-World defendsive end, will be the biggest challenge the New England Patriots’ offense will face this Sunday. Since his breakout pick-six in Houston’s playoff win over the Cincinnati Bengals as a rookie, the fifth-year defensive end developed into one of the league’s best defensive players. Just how good is he, and how does he compare to some of the greatest defensive players of the recent past?

Sacks, Sacks and More Sacks

The most visible number for Watt’s dominance is his sack total. He had 20.5 his second season, 20.5 again in 2014, and has 13.5 in 12 games this year. His 36.5 sacks in his first three seasons are the seventh-most since the NFL started officially keeping track of stats in 1982. His 57 sacks in his first four seasons is the third-best total. With four games left to play in his fifth season, he is already at 70.5 career sacks. The only other player with at least 70 sacks in his first five NFL seasons is Reggie White, who had the benefit of playing two USFL seasons first. Just three other players have had at least 60 sacks in their first five seasons, though Von Miller will likely join that group. By any standard, Watt is off to the best start to his career of anybody in the history of the official sack metric.

Watt’s five-year streak has been impressive not just for a rookie, but for any player at all. He is just the sixth different player since 1982 to have at least 70 sacks in a five-year period at any point in his career. (Table 1.)

Table 1. 70-plus Sacks in Five-Year Span, 1982-2015

Player Years
Jared Allen 2007-2013
Bruce Smith 1986-1990
Lawrence Taylor 1984-1990
DeMarcus Ware 2006-2012
J.J. Watt 2011-2015
Reggie White 1985-1992

More Than Just Sacks

It would be one thing if Watt was just a pure sacker, or if he was converting pressures to sacks at a particularly high rate. He is not. Even in his modest, 5.5-sack rookie season, he was tied for sixth in the league in quarterback hits. In each of 2012, 2013, and 2014, he led the league in combined sacks and quarterback hits by at least 10 each season. In 2014, he had a combined 54 sacks and hits, a whopping 23 more than second-place Elvis Dumervil. Like his 13.5 sacks so far, his 42 quarterback hits also leads the NFL. He is also getting near the passer when he does not hit him. According to Football Outsiders data, he was fifth in the NFL in hurries in 2012, fourth in 2013, and first in 2014.

Watt also excels at winning when he does not get near the passer, by getting his hands into passing lanes and deflecting passes. His most impressive season in that category was 2012 when he tipped 18 passes at the line of scrimmage, per Football Outsiders, more than twice as many as any other defender. In 2013, he was one back of the league leaders in that area, and he again topped the league in 2014. This season, his five passes defensed lead all defensive linemen.

His Position Makes His Production More Impressive

What makes Watt’s productivity even more impressive is where he’s doing it from. The lists of the top sackers and hurriers, both present-day and in the past, is full of edge-rushers, 4-3 defensive ends and 3-4 outside linebackers. Watt’s versatility has increased in recent years, but he has spent much of his career as a 3-4 defensive end while playing defensive tackle in sub-package situations. The interior line is the shortest path to the quarterback, but it also tends to be the hardest one. The only other non-edge rushers among the historical lists of the top sackers are Bill Pickel, a 1980s Raiders nose tackle who excelled early, and Bruce Smith.

Great Run Defense As Well

The criticism, fair or not, against many interior linemen who do get pressure, from the modestly successful to Hall of Famers like John Randle, is they are not good run defenders. Watt is almost undeniably a superb one, ranking first in the league in run tackles for loss in both 2012 and 2014 and third in 2013. His particular specialty is making plays from the backside of runs, where he excels at getting inside and forcing the runner to bounce the run to avoid being taken down in the backfield.

The Difficulty of Putting Watt in Perspective

Unfortunately, some of the stats that show Watt’s greatness make it hard to compare him to other players. Sacks were not an official stat before 1982. Passes defensed are not widely available before 2001. Quarterback hits go back to 2006. Hurries and other unofficial stats are about as old. Tackle totals still vary widely between scorers, and they were worse before the mid-1990s. This makes it hard to more precisely compare Watt to greats like Bruce Smith, Lawrence Taylor, and Reggie White.

The most comprehensive measure we have may be Football Outsiders’ defeats, which goes back to 1996 and includes sacks, tackles for loss, forced fumbles, interceptions, and third-and-fourth-down tackles short of the sticks. Watt’s 56 defeats in 2012 is the most by any single player in that time span. His 43 in 2014 ranks fourth. His league-leading 34 through 12 games this year puts him on pace to match Ray Lewis’ third-highest total of 45. By almost any measure Watt is the most consistently impactful defender of the past 20 years. His only challenge now may be sustaining that greatness for the rest of his career.