Revisionist History

Spadaro makes a number of good points in this piece about the future of the Eagles' offense, but this part is a little hard to swallow:

It isn't all about the wide receiver, folks. Back in 2004, the Eagles added Terrell Owens and the offense was terrific. But for those who focus only on that season, I'll remind you that the Eagles had a great offense the year before and that, yeah, the offense went up and down the field in the NFC playoff victories over Minnesota and Atlanta while Owens was recovering from his ankle injury.

Quick comparison:

Little bit of a difference there.

And as for the part about the playoffs ... the offense wasn't that great in those two games, nor were the Eagles facing championship-caliber defenses.

Comments

[Andrew]

Oh for shame.

You mix up the injured McNabb offense of the beginning of the season with the healthy McNabb offense of the end of the season in 2003. And you probably include the 2 JV games at the end of 2004 in that year.

If you are going to play stats, at least give stats that are meaningful.

Look, just take a couple of simple measures using Weeks 4-17 of 2003 and Weeks 1-15 of 2004.

Win-Loss
2003 - 12-2
2004 - 13-1

Points
2003 - 364 points (26 per game)
2004 - 369 points (26.4 per game)

First Downs
2003 - 266 (19 per game)
2004 - 272 (19.4 per game)

Yards
2003 - 4522 (323 per game)
2004 - 5121 (365.8 per game)

Net Passing Yards
2003 - 2728 (194.9 per game)
2004 - 3643 (260.2 per game)

Rushing Yards
2003 - 1794 (128.1 per game)
2004 - 1478 (105.6 per game)

Pass Attempts
2003 - 396 (28.3 per game)
2004 - 466 (33.3 per game)

Rushing Attempts
2003 - 385 (27.5 per game)
2004 - 331 (23.6 per game)

Raw Passing Yards per Attempt
2003 - 7.3
2003 - 8.2

Raw Passing Yards per Completion
2003 - 12.1
2004 - 12.9

Offensive Turnovers
2003 - 14
2004 - 16

Summary? Owens meant more passing, more passing yards, more productive passing attempts and marginally better yards per completion. Owens also meant less rushing and less rushing yards.

Owens did not mean more scoring, more first downs, a "more explosive" passing game, or a change in turnovers. Owens also did not mean appreciably more wins.

Essentially the Eagles became more efficient at passing with Owens, but the efficiency just changed field position (meaning less scoring by opponents), they didn't change offensive point production.

One should also be careful in attributing to Owens the higher completion percentage in catches enjoyed by every receiver on the team (except Mitchell). 2004 was the year of the reemphasis of the pass interference and illegal contact rules due to what happened in the Patriots-Colts and Eagles-Panthers playoff games. This reemphasis lead to a general explosion in offense across the league. Leaguewide, passing yards went up from 109,467 yards to 115,345 yards. Completion percentage went from 58.78% to 59.79%.

[Me]

No way, Andrew. Not on this one.

The first clue that you're off is that you're cherry-picking stats. Seasons don't start in week four.

Secondly, if you want to control for schedule strength, all you have to do is check out the football outsiders numbers. They've got an offensive DVOA of 11.1 percent for 2003 and 14.8 percent in 2004 (for other readers, higher numbers are better).

Thirdly, you can't use a leaguewide completion percentage increase of 1 percent to explain why McNabb put up a completion percentage that was five+ percentage points better than his career average and 2.5 percentage points higher than his second-best season.

Finally -- and this point matters greatly when you're looking at the current team -- when Brian Westbrook got hurt in 2003, the Eagles had no offense. Owens gave them two guys who scared opposing defensive coordinators.

Pending what happens in free agency, we're right back at 2003 again right now.

[Andrew]

I don't think it it cherry-picking to ignore the first two games of 2003 when McNabb had an injured thumb and Westbrook was not really being used and just compare two 14 game seasons of healthy McNabb and active Westbrook, since then you have equivalent numbers.

Second, McNabb's increase in completion percentage in 2004 was at least in part due to leaguewide officiating trends and in part due to his maturity in the position, and not just Owens. While McNabb had just kissed 60% a couple of times in a 8 game run in his career up to 2003, since early 2003, he has been mostly over it, with the only exceptions being the time around the Denver game with Owens in 2005 and the bombs away offense of 2006 through the Green Bay opener.

Third, his completion percentage to Owens was under his completion percentage as a whole - it was 61% on 127 pass attempts. The main reason for his high completion percentage as a whole was the magic with Westbrook who caught 84% of the 87 balls thrown his way.

Looking closely at the numbers, I'd say Owens added about 2% to the completion percenatage through his play and influence on defenses vs. where McNabb was in 2003 or in 2007, while the leaguewide trends added 1%. McNabb himself with Westbrook, LJ Smith, and whoever played wideout was at about 61%.

Yes, the 2004 offense was better. They got more yards because they threw the ball more. You almost always get more yards by rushing less and passing more. But as I noted, it didn't cause more points or more wins. The main effect was to put the defense in a better position so that it gave up fewer points 202 vs. 239 in those same 14 games.

Kind of funny that the biggest effect Owens had on the teams was to reduce points against by over 2 points per game, no?