49ers-Jets Statistical Preview: Near, Favre, Wherever You Are

The 49ers welcome their nemesis, Brett Favre, to San Francisco this week in a game that has far more in the way of playoff implications for the Jets than it does for the Niners (sigh). The fact that the 8-4 Jets are playing a meaningful game during December is quite a turn of events considering that they were 3-9 at the same point last season. The conventional wisdom says that this 5-game improvement is the direct, tangible result of replacing Chad Pennington (and Kellen Clemens) with Favre. The opposite can be said of Favre's former team: With Favre last year, the Packers were 10-2 through 12 games; without Favre this year, they're 5-7. If you haven't noticed, that's an amazingly symmetrical 5-win turnaround as well. So it's pretty obvious that the Jets are a better team with Favre, and the Packers are a worse team without Favre. That much we can agree on. One important question remains, however: Did the acquisition of Favre CAUSE the Jets' improvement and/or the Packers' decline?

That pesky five-letter word, "cause," is far more complicated than its length suggests. I'm not going to turn this preview into some kind of science or philosophy class, but I'll make one very important scientific point. Two conditions are necessary to say that variable X causes outcome Y: (1) X must occur at a point in time prior to Y, and (2) the effect of X must be isolated from the effects of other variables. In the case of Favre's impact, we can only assert causation if Favre's arrival came before the Jets' improvement (and/or the Packers' decline), and if we can isolate Favre's arrival from other factors that might have "caused" the Jets' improvement (and/or the Packers' decline).

Obviously, the time condition is easily satisfied here. In contrast, the "isolated effect" condition is very difficult to satisfy because football is played 11-on-11 (except when the 49ers play 11-on-17 when the officials suit up for the opposition). More than any other team sport, the successful execution of a play requires several players moving in unison, such that perfect individual execution may not translate directly into team success. And let's not forget about the "on-11" part of 11-on-11. There's a defense on the other side of the ball trying to disrupt what the offense is trying to execute. These within- and between-team dynamics make it very difficult for football statisticians to cross the threshold from correlation to causation. In other words, there are so many factors influencing a given football outcome, that it's nearly impossible to tell what is a cause and what is a coincidence.

Amazingly, assessing the impact of Favre is as pristine a cause-and-effect analysis as it gets in football statistics. Think about it. Going into this season, the sans-Favre Packer offense was essentially the same unit as the 2007 version: Same coach, same running back, same wide receivers, and same offensive line. Likewise, the sans-Favre Jet offense of 2007 was essentially the same unit as this year's version: Same coach, same running back, and same wide receivers. It turns out that, to assess Favre's causal impact, there are really only three factors that we have to account for. First, the Jets acquired LG Alan Faneca and RT Damien Woody to strengthen their offensive line. Second, the Jets and Packers have played different schedules this year than they did last year. Finally, the Jets and Packers have played in different situations than they did last year. Luckily, we can account for these factors thanks to Football Outsiders because, as we all know by now, they have developed a statistic (ALY) that isolates offensive line performance (as much as it can be isolated), and another statistic (DVOA) that adjusts performance for game situation and opponent.

So in this week's statistical preview, I'm going to attempt to answer a question that has permeated NFL debate all season: After adjusting for schedule strength, game situations, and the Jets' offensive line improvement, did the acquisition/loss of Brett Favre cause the Jets' offense to improve and the Packers' offense to decline?

RUN BABY, RUN!

Below is a table showing a comparison of the 2008 Jets offense to the 2007 Jets offense, as well as a comparison of the 2008 Packers offense to the 2007 Packers offense:

Team	OFF DVOA	RK	PASS DVOA	RK	RUSH DVOA	RK
NYJ 07	-11.10%	25	-8.30%	23	-14.00%	28
NYJ 08	4.60%	19	3.10%	19	6.40%	8
NYJ DIFF	15.70%	-6	11.40%	-4	20.40%	-20
Team	OFF DVOA	RK	PASS DVOA	RK	RUSH DVOA	RK
GBP 07	17.30%	5	26.10%	5	3.50%	9
GBP 08	12.00%	10	21.20%	9	1.40%	15
GBP DIFF	-5.30%	5	-4.90%	4	-2.10%	6

After the jump, I'll discuss this table, attempt to adjust these numbers for the Jets' offensive line improvement, compare Favre and non-Favre quarterback performance for the Jets and Packers, detail 5 crucial game situations, and make my prediction for this week's 49ers-Jets game...

Clearly, the Jets' offense has gotten better this season with Favre at quarterback, and the Packers' offense has gotten worse this season without Favre at quarterback. And remember...these changes have been adjusted for changes in schedule strength and game situations. What's interesting in this table is that the improvement/decline of each offense has not come via the passing game: The Jets' passing offense has only moved up 4 spots in the rankings, while the Packers' passing offense has only moved down 4 spots. By far, the most glaring Favre-related change has been to the Jets' rushing offense: They improved from 5^th-worst in the league to 8^th-best. This begs the question: Did the acquisition of Favre cause the Jets' running game to improve?

Below are two tables. One compares Thomas Jones' 2008 performance with his 2007 performance, and the other compares the Jets' offensive line performance in 2008 with their performance in 2007:

RB	DYAR	RK	DVOA	RK	RUNS	YDS	EYDS	TD
T. Jones 08*	279	3	12.5%	5	311	1,451	1,600	15
T. Jones 07	-36	43	-11.3%	40	310	1,126	1,074	1
Jones DIFF	315	-40	23.8%	-35	1	325	526	14

* Prorated to a 16-game season

Team	ALY	RK	MID ALY	RK	RT ALY	RK
NYJ 07	4.10	21	4.22	12	4.38	15
NYJ 08	4.71	4	4.66	4	4.92	2
NYJ DIFF	0.61	-17	0.44	-8	0.54	-13

As you can see by these two tables, the Jets' offensive improvement has been largely due to the concomitant resurgences of Thomas Jones and his offensive line. With nearly an identical number of projected carries, Jones will gain about 325 more yards (1.04 more per carry) and 526 more EYds (1.64 more per carry) than he did in 2007 because he has been almost 25% more situationally efficient than he was last season. Furthermore, Jones is over 300 yards more valuable to the Jets' offense this season as compared to last. Much of Jones's renaissance can be traced to the addition of Faneca and Woody, which has resulted in an increase of about half-a-yard per carry in terms of overall ALY, middle-of-the-line ALY, and right-tackle ALY.

OK. So I think I've established that the Jets have Faneca, Woody, and Jones to thank for their offensive improvement this season. Surely, however, Brett Favre's move from Green Bay to East Rutherford has had some impact on the Jets' and Packers' passing game, right? Well, below are two tables comparing Favre's performance on the 2008 Jets with that of Favre on the 2007 Packers, Aaron Rodgers on the 2008 Packers, and Pennington on the 2007 Jets.

Player	DYAR	RK	DVOA	RK	ATT	YDS	EYDS	TD	INT	COMP%
Favre 08*	580	16	4.7%	20	561	3,525	3,809	27	19	69.30%
Favre 07	1,437	3	28.0%	5	556	4,151	4,915	28	14	66.90%
Rodgers 08*	1,035	11	17.8%	10	573	3,639	4,308	27	13	64.70%
Pennington 07*	251	22	-3.6%	26	508	2,843	3,033	18	16	69.40%

*Prorated to a 16-game season

Comparison	DYAR	RK	DVOA	RK	ATT	YDS	EYDS	TD	INT	COMP%
Favre 08* vs. Favre 07	-857	13	-23.3%	15	5	-626	-1,106	-1	5	2.4%
Favre 08* vs. Rodgers 08*	-455	5	-13.1%	10	-12	-113	-499	0	5	4.6%
Favre 07 vs. Rodgers 08*	402	-8	10.2%	-5	-17	512	607	1	1	2.2%
Favre 08* vs. Pennington 07*	329	-6	8.3%	-6	53	683	776	9	3	-0.1%

*Prorated to a 16-game season

What we see in these two tables is the following:

• Favre is on pace to have a less-efficient 2008 season than Rodgers.
• Favre is on pace to have a less-efficient 2008 season than he did in 2007.
• Rodgers is on pace to have a less-efficient season for the Packers in 2008 than Favre did in 2007.
• Favre is on pace to have a more-efficient season for the Jets in 2008 than Pennington did in 2007.

So basically, Favre's not having a particularly valuable or efficient individual season, but he's still better than the guy he replaced in New York who had essentially the same supporting cast. Likewise, Favre's replacement in Green Bay is worse than Favre was last year with essentially the same supporting cast. Also, if you compare these QB DVOA differences to the team passing DVOA differences above, you find that the gain/loss of Favre accounts for a large percentage of the overall differences. Clearly then, acquiring Favre has improved the Jets' QB position, while losing Favre has deteriorated the Packers' QB position.

And finally, here's one last piece of evidence showing Favre's impact on the Jets' and Packers' passing offenses:

Team	ASR	RK
NYJ 07	9.3%	30
NYJ 08	6.6%	19
NYJ DIFF	-2.7%	-11
GBP 07	3.1%	1
GBP 08	5.9%	16
GBP DIFF	2.8%	15

Here we see that the Jets' offensive line has cut its ASR by nearly one-third this season, while the Packer line's ASR has nearly doubled. What's interesting though is that the magnitude of ASR change is practically identical for both teams. Remember, the only thing that's changed on the Green Bay offense is the QB, so we can allocate almost the entire ASR difference to losing Favre. Allocating the Jets' ASR difference is more difficult because the offensive line acquisitions have surely had as much (or more) of an impact than Favre.

Wow, that was a veritable cornucopia of statistical evidence about the Favre impact!!! What conclusions can we draw? For fear of retribution in the comments section, I'm going to be as conservative as possible here...no hyperbole:

Bottom line: The major statistical difference between the with- and without-Favre Jet/Packer offenses is a clear improvement in the Jets' running game, which has been caused more by the acquisition of Faneca and Woody than by the acquisition of Favre. In terms of passing offense, the Favre move has caused the Packers' passing game to get worse because Favre was more efficient and better able to avoid sacks last season than Rodgers is this season with the same supporting cast. Finally, although Favre has caused an improved Jets' passing game by being more efficient this season than Pennington was last season, it's unclear how much of the impact has been due to the offensive line acquisitions and their effect on pass protection.

49ERS-JETS DVOA MATCHUP

Here's how the 49ers stack up against the Jets in terms of DVOA:

Team	Total DVOA	Rank	Offense	Rank	Defense	Rank	Special Teams	Rank
SF	-20.0%	26	-18.1%	29	7.9%	19	6.0%	1
NYJ	6.1%	14	4.6%	19	3.6%	12	5.1%	4

For the second week in a row, we have a battle of special teams...just kidding. :-) This week's game should definitely be much more action-packed than last week, for better or for worse. Although the Niners have a 26.1% overall disadvantage, which is very similar to the disadvantage they had going into the Saints game (a 31-17 road loss), almost all of the disadvantage resides with one unit. When the Niners' D and special teams are on the field, this is a pretty even matchup. The same can't be said when the 29^th-ranked 49er offense lines up against the 12^th-ranked Jet defense. Luckily, the Niners' offense plays better at home (-11.3% DVOA, 27^th) and the Jets defense plays worse on the road (9.7% DVOA, 19^th), so a 14.5% disadvantage reduces to a mere 1.5% disadvantage when game location is taken into account. With that said, this sure looks like a game that will be closely contested deep into the 2^nd half, which means that we'll find out once again whether Coach Singletary has to show off the "FINISH" mantra tattooed across his butt cheeks.

THE DEVIL IS IN THE DETAILS

The five SVW's to focus on during this week's 49ers-Jets matchup are detailed below:

Situation	SF Rank	NYJ Rank
SF RED ZONE OFFENSE VS. NYJ RED ZONE DEFENSE	31	5
SF RUSH OFFENSE VS. NYJ RUSH DEFENSE	26	7
SF RUSH OFFENSE VS. NYJ RUSH DEFENSE ON 1ST DOWN	28	6
SF PASS OFFENSE VS. NYJ PASS DEFENSE ON 3RD DOWN	30	8
SF DEFENSE VS. NYJ OFFENSE ON 2ND AND SHORT	27	6

OK, so the Niners have no DVOA matchups that they can exploit this week. Rather, they have five unfavorable matchups that they have to overcome in order to win this game. To me, the red zone and rush offense matchups are particularly troublesome given Mike Martz's penchant for red zone suckage and Mike Singletary's affinity for smashmouth football. Hey, at least the defense has had 2-games worth of practice on these 2^nd-and-short SVWs!

"WITH HIGH HOPE FOR THE FUTURE, NO PREDICTION IS VENTURED." - ABRAHAM LINCOLN

I got last week's pick wrong, but, to tell you the truth, we're all happy that I did. My rather small ego certainly allows for an unexpected Niner victory. On the other hand, what about an unexpected Niner loss? Hmmm...we just might find out on Sunday because...given the statistical matchup between the 49ers and Jets, here's my prediction for this week's game:

New York Jets	21
San Francisco 49ers	24

**DVOA, DYAR, ALY, and ASR statistics used to produce this article were obtained from Football Outsiders.