Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Futures: Josh Rosen

UCLA's quarterback clearly has the talent to succeed as an NFL starter. The question is whether or not he can avoid enough mistakes to become a superstar.

19 Aug 2004

Packer Defense vs. Packer Offense

by Aaron Schatz

As I mentioned last week, a couple of our essays were cut from the book Pro Football Forecast 2004, and one of them was the essay on the Packer offense and defense which was written specifically to force the reader who signs his comments "A Packer Owner" to buy the book.  I promised him I would explain why I had the 2003 Packers defense rated higher than the 2003 Packers offense, and explain it I shall.

This is a big issue with Green Bay fans who can't seem to stop worrying about whether their defense is good enough to make the Super Bowl.  In fact, many think their defense is just plain pathetic.  People need to get over 4th-and-26.  It was a horrible, horrible play, it was the most important play of the year, but it was one play.  The high quality of the Packers defense from the second half of the season is not completely negated by 4th-and-26; even the best defenses occasionally blow it and, unfortunately for Packers fans, Green Bay chose the worst possible time to suck.  Packers fans should also feel optimistic because my research shows that late-season defensive improvement is much more likely to carry over to the following season than late-season offensive improvement.

I'm guessing (and hoping) that this article will get a lot of attention from Green Bay message boards so let me quickly introduce the stats.  DVOA, or Defense-adjusted Value Over Average, represents every single play during the NFL season compared to the league-average based on situation and opponent.  It is called DVOA even when we're adjusting defenses for the quality of the offense, and since positive numbers mean more scoring, defense is better when it is negativeRead more here.  DPAR is a related stat that takes into account both a player's quality and how much he is used, it stands for Defense-adjusted Points Above Replacement.  Read more here.  NFC North fans also want to check out our look at NFC Over-Unders, Part I is here and Part II is here.

I'm publishing the essay here exactly as it was submitted for the book, which means that the DVOA numbers are the "older version" from before the May revisions.  The revisions moved the 2003 Packers up a bit in offense, so that they are now ranked #9 for offense and #8 for defense.

Packer fans who are seeing DVOA numbers for the first time may be rubbing their eyes, not to mention questioning our sanity. The conventional wisdom in Green Bay puts on emphasis on the offensive duo of Brett Favre and Ahman Green, and the defense is a bit of an afterthought. The NFL's team rankings support that view, with Green Bay ranked fourth in offense for 2003 and seventeenth in defense. Why does DVOA's opinion of the Packers conflict so much with the standard statistics?

Many times, when DVOA diverges from the NFL team rankings, turnovers are the culprit. St. Louis is the best example, as they led the NFL in turnovers on both offense and defense. That moves their offensive DVOA down and their defensive DVOA up.  Green Bay's turnovers had a similar effect in 2003. The Packers were tenth in offensive turnovers, with 32, and seventh in defensive takeaways, with 32. How symmetrical! That explains part of it, but not all. Let's look closer at Green Bay's 2003 season.

We'll start with the offense, since that's where discussions of the Packers always begin. Packer fans will likely look at that #11 rating (note: May revisions moved it up to #9) and ask, "How can a team with such a great duo like Favre and Green be rated below the Jets and Bengals on offense?" Part of the problem, as relates to 2003, is that Packer fans have in their head a perfectly healthy Favre, and that's not the man who was playing quarterback for Green Bay during the second half of the year. For a few weeks, Favre's thumb injury seriously impacted his performance.

The table below shows Favre's DVOA weekly performance in 2003.  (+) notes a game where rating is adjusted significantly upwards due to opponent's good defense; (-) notes a game where rating is adjusted significantly downwards due to opponent's poor defense.

Week Opponent DPAR DVOA
1 MIN -4.4 -48.5%
2 DET (-) -2.6 -42.4%
3 ARI (-) 2.2 9.2%
4 CHI 1.9 6.2%
5 SEA 7.4 67.1%
6 KAN 5.2 30.1%
7 STL (+) 8.3 64.4%
9 MIN 4.1 33.8%
10 PHI -7.0 -94.2%
11 TAM (+) -2.0 -35.0%
12 SFO -5.2 -122.5%
13 DET (-) -6.4 -63.5%
14 CHI (-) 2.2 9.2%
15 SDG (-) 4.6 36.2%
16 OAK (-) 13.1 145.3%
17 "DEN" (Yeah, right) -0.7 -22.2%

Favre apparently injured the thumb at some point during the Week 7 loss to St. Louis, as it was reported in the press during the bye week. He had a reasonable game in Week 9 against Minnesota, and then suffered through five fairly dismal games where he turned the ball over 13 times. In three of these games he threw for fewer than 140 yards.

Fortunately for the Packer faithful, Green Bay managed to go 3-2 in these games solely because of their defense. They nearly beat Philadelphia despite six fumbles (two lost) because the defense kept the Eagles -- one of the league's top rushing teams -- to three yards a carry. Against Tampa, the Packers forced three turnovers and held the Bucs to a pitiful 1-of-11 on third down conversions. (This was also the one game where special teams made a difference, thanks to excellent punt and kick returns by the Packers.) They beat the 49ers 20-10 by derailing the Tim Rattay bandwagon, sacking him four times and keeping him to 119 net yards passing. They picked off Kordell Stewart three times in a 34-21 win over Chicago where the Bears offense had only one touchdown (the other two were on a kickoff return and an interception return).

The timing couldn't have been better for the Packers, because they could counteract the loss of one pound of healthy Brett Favre thumb with the addition of 325 pounds of Grady Jackson. When the Packers picked up the defensive tackle, waived by New Orleans at midseason, and stuck him in the lineup to replace an injured Gilbert Brown (who had been playing with a torn right biceps muscle) their defensive performance dramatically improved.

For the first eight games of the year, Green Bay's defense had a +6.3% DVOA, and they were better then average in only two games -- the first two weeks against Minnesota and Detroit (and the Minnesota game is only rated as better than average because of adjustments for the powerful Viking offense). From Week 10, when Jackson arrived in Green Bay, through Week 16, the Green Bay defense had a -18.6% DVOA, and they were worse than average in only one game, Week 13 against Detroit -- and then only because of adjustments for the pitiful Detroit offense. The Packer defensive rating with Jackson would be even better if I included Week 17, when the Denver Broncos seemingly picked random fans out of the stands and stuck them on offense. In seven of the eight regular season games after Jackson's arrival, Green Bay's defensive DVOA was better than its offensive DVOA (the exception was the Monday nighter against Oakland where Favre had the ridiculous first half).

Although Jackson is known primarily as a run-stopper, his arrival actually had a stronger impact on Green Bay's pass defense (see below). In 2002, the Packers had one of the league's top pass defenses, but they had been struggling through the first half of 2003. Once Jackson entered the lineup, ends Kabeer Gbaja-Biamila and Aaron Kampman were free to play the pass more, with less worry about running backs scampering up the middle.

Green Bay defense before and after Grady Jackson
  Weeks 1-9 Weeks 10-16
DVOA vs. run 0.4% 0.4%
DVOA vs. pass 10.8% -21.9%
DVOA total 6.3% -18.6%
Week 17 not included because Denver played your mom at quarterback

Stopping the run on first down means more second-and-long and third-and-long, and the Packers were nearly impenetrable on third-and-long after Jackson's arrival (below). Remember, no matter the situation, the average DVOA is 0%, so Green Bay's defense was extra dominant even after taking into account the fact that third-and-long is naturally harder to convert than third-and-short.

Green Bay defense before and after Grady Jackson
  Weeks 1-9 Weeks 10-16
1st Down 12.3% -6.8%
2nd Down 1-5 to go 22.5% -12.7%
2nd Down 6+ to go -17.0% -24.4%
3rd Down 1-5 to go -57.7% 8.8%
3rd Down 6+ to go 118.2% -127.6%
Week 17 not included because the part of Denver
was played by the cast of Queer Eye for the Straight Guy

The Saints were apparently unhappy with Jackson because he couldn't get his weight down, but, according to the Milwaukee Journal-Sentinel, the Packers had him in their sights for weeks before finally picking him up. It turned out the Packers didn't need to make the trade they had been considering, because the Saints waived Jackson and only one other team put in a claim -- Carolina, which was higher than Green Bay in the standings. Between Jackson and Ted Washington, you would think that 2003 proved that overweight defensive tackles are not a significant problem. Oddly enough, waiving Jackson didn't seem to hurt the Saints the way it helped the Packers. Their defensive DVOA was the same before and after Week 10.

By the playoffs, Green Bay was hitting on all cylinders, with Favre's thumb injury no longer causing him trouble and the defense playing as well as they had all season. Many people felt that they were headed to the Super Bowl, and it certainly looked that way until 1:12 was left. Remember, we said that the Packers were nearly impenetrable on third-and-long after Jackson's arrival, but we said nothing about fourth-and-long. In one of the more unexpected meltdowns of the last few years, the Packers let the Eagles convert on 4th-and-26, the game went to overtime, and Brett Favre tossed up a horribly ill-conceived pass to nobody that came down in the arms of Brian Dawkins. The next day, television repair shops in Wisconsin reported a 500% increase in broken cathode-ray tubes.

But the DVOA trends from 2003 give Green Bay fans hope for 2004. Favre decided not to get surgery on his thumb, and according to reports out of Green Bay it still causes him pain. But while Favre may be on the downside of his career, and may never again be the great quarterback he was when the Packers were winning Super Bowls, the Packers don't need that quarterback if they want improve on last year's record. They just need the quarterback from Weeks 10-14 to stay as far away as possible, and that seems likely after Favre righted the ship somewhat at the end of last season.

At the same time, the Packer defense will have Grady Jackson for the entire season. Nick Barnett will have another year of experience under his belt. It is reasonable to expect that the Packers will have a top five defense from the beginning of the year. There are depth issues on the Green Bay defense, but if most of the unit can stay healthy (note: and Mike McKenzie stops being a jackass), Favre may get that one last shot at another Super Bowl title.

Posted by: Aaron Schatz on 19 Aug 2004