Possibly the closest Super Bowl matchup in history also poses the question: how much does it mean when certain aspects of an NFL team improve dramatically in the second half of the season?
04 Feb 2005
by Aaron Schatz, with additional analysis by Michael David Smith, Michael Tanier, and Will Carroll
The Philadelphia Eagles are a team filled with talented players and led by one of the NFL's best coaches. They have more wins than any NFL team this decade and dominated their conference this season from wire to wire. It is their misfortune that, when they finally reach the Super Bowl, their opponent presents them with the worst matchup possible.
This year's Super Bowl has taken on a whiff of inevitability, and very few fans expect the Eagles to triumph. The New England Patriots are coming off two games where they kept the NFL's highest-ranked offense to just a field goal and then laid waste to the NFL's highest-ranked defense with 41 points. The imbalance between the AFC and NFC this season makes the case for a Patriots victory that much stronger (and, as you'll see near the bottom of this preview, Philadelphia's struggles against the AFC go back further than just this season).
Football Outsiders has now been in existence for two seasons, and both of those seasons the Patriots have made it to the Super Bowl. I have always tried to put my biases out in the open, and I do not hide the fact that I am a New England fan. But this preview is a bit of a problem, because this is the first time in two seasons where I believe that there is a significant chance that the Patriots will win by a large margin. Last year, I predicted that the Patriots would play close games in all three rounds of the playoffs. This year, I (incorrectly) predicted close games against both Indianapolis and Pittsburgh. In fact, overestimating the importance of the Richard Seymour injury, I felt that the Colts had a slim advantage in the divisional round.
In Super Bowl XXXIX, however, I think that all the indicators point towards a New England victory, and not a close one. If you believe that this analysis represents my Patriots fan bias, well, I can't really do anything to dissuade you. I'm not one of those Patriots fans who believes that the Patriots are unbeatable, and to suggest that they might lose is "disrespect." But in this matchup, they are heavy favorites. (By the way, the views expressed here are solely my own -- I'm guessing that Mike Tanier, who helped me with some analysis of the Eagles, would disagree with my conclusions.)
My belief that the Patriots will win big does not mean that a win is guaranteed. I'm guessing there's roughly a 50 percent chance that the Patriots win this game convincingly, a 30 percent chance that they win it close, and about a 20 percent chance of a Philadelphia upset. To stop the coronation of an NFL dynasty, the Eagles will need a stalwart performance from their defense, strong field position thanks to turnovers and special teams, and an abnormally strong performance from one of their secondary receivers. Either that, or a near-impossible complete recovery from the most famous joint in sports.
* * * * *
For those who may be visiting this site for the first time to read this preview, some explanations for our statistics. DVOA (Defense-adjusted Value Over Average) breaks down each play of the season and compares it to the NFL average based on situation and opponent. You'll find it explained further here. Since DVOA measures ability to score, a negative DVOA indicates a better defense and worse offense, and a positive DVOA indicates a better offense and worse defense.
Each team is listed with DVOA for offense and defense, split into rush and pass, along with rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) I've also listed each team's red zone performance on offense and defense.
SPECIAL TEAMS numbers are different; they represent value in points of extra field position gained compared to NFL average. Field goal rating represents points scored compared to average kicker at same distances. All special teams numbers are adjusted by weather and altitude; the total is then translated into DVOA so it can be compared to offense and defense.
TREND is the WEIGHTED DVOA trend, based on a formula which drops the value of games early in the season to get a better idea of how teams are playing now (explained here). This includes playoff performance.. Except for TREND numbers, all DVOA statistics on this page are regular-season only unless otherwise noted.
Each team also gets a chart showing their performance this year, game-by-game. For the Super Bowl there are two charts for each team, one for offense and one for defense. In addition to a line showing each game, another line shows the team's trend for the season, using a third-power polynomial trendline. That's fancy talk for "the curve shifts direction once or twice." Once again, because defense is negative, a downward curving trend reflects an improving defense. Because Philadelphia played primarily second-stringers in the final two games of the regular season, those games were removed when computing the Eagles' trendlines. However, those games are included in all other stats unless otherwise noted.
|PHI OFF||NE DEF|
|DVOA||17.7% (6)||-9.1% (6)|
|TREND*||5.7% (12)||-13.5% (7)|
|PASS||21.8% (9)||-6.5% (11)|
|RUSH||11.4% (6)||-12.3% (4)|
|RED ZONE||31.1% (3)||-39.7% (2)|
Any discussion of the Eagles offense has to begin with Terrell Owens. There might be a handful of sherpas living in remote Himalayan villages who are not aware of the fact that the Eagles' game-breaking receiver broke his right ankle six weeks ago but accelerated the timetable of his rehabilitation in an effort to be ready for the Super Bowl. He plans to play in the game, but at what strength?
A healthy Terrell Owens, more than almost any wide receiver in the league, is able to take an eight-yard slant, shake defenders, and turn it into a long gain. Michael Clayton and Andre Johnson were the only receivers in the NFL to ring up more yards after catch per game than Owens.
But barring a medical miracle, that ability will be significantly reduced on Sunday. While the only impediment to Owens' ability to run should be his own pain tolerance, the ankle will be a much bigger issue when Owens needs to change direction. If he attempts to make a hard cut, he won't be able to plant on the right ankle and go left. That gives the Patriot defender an opportunity to shade him, knowing any sudden move is likely to be to the right.
If the Eagles are going to use Owens, it will be in the red zone. During the first 15 weeks of the season, the Eagles threw to Owens 17 times in the red zone. They threw to other wide receivers a grand total of once: an incomplete pass to Freddie Mitchell on 3rd-and-goal from the 5-yard line against Green Bay. With tight end Chad Lewis out with an injury, Philadelphia needs Owens in the red zone even more. Throwing to Todd Pinkston in the red zone would be a shock roughly equivalent to throwing to Hugh Douglas lined up as a tight end.
With Owens on the sidelines, the Eagles won their first two playoff games primarily thanks to their defense. The offense, though not as poor as it had been in Weeks 14-17, was not as powerful as it had been for most of the season. Remember, that steady downward trend on the chart to the right does not include the final two games (the dark circles) where Jeff Blake and Koy Detmer were gift-wrapping the ball and presenting it to the Rams and Bengals with a pretty ribbon on top.
The safest assumption is that Owens will be useful, but not the game-changing force he was during the regular season, and so the Philadelphia offense will play at a level similar to the last two games. Without Owens at full strength, Philadelphia has an offense whose weapons, with the exception of the rare deep throw to Greg Lewis, are primarily short- and medium-range: Brian Westbrook on screens or in the slot, Freddie Mitchell on short routes, tight end L.J. Smith, and the running game with Westbrook and Dorsey Levens. Unfortunately for the Eagles, these weapons directly match the strengths of the New England defense.
Westbrook, for example, gains value from his ability to break long runs or gain yards after the catch when he's used in the passing game. But the New England linebackers are excellent at pursuit and tackling, and safety Rodney Harrison's physical presence stops a lot of long runs. No defense gave up fewer double-digit runs than New England.
Of course, while the Patriots don't give up long runs, they also don't stuff opposing runners at the line very often. That's a good reason the the Eagles to give a few carries to Dorsey Levens. Last year with the Giants, the 34-year-old Levens looked like his career was done; back in Philadelphia, in a familiar offensive system, he has had great success. Levens actually had a higher DVOA than Westbrook, and according to our Running Back Success Rate metric, he was successful on 57% of his runs while Westbrook was only successful on 43% of his.
After a hot start, Westbrook developed a habit this year of busting everything to the outside in search of a 20-yard gain. Around the middle of the season, Reid began using Levens more to run between the tackles, while Westbrook got a little smarter about taking what the defense gave him. That's not a bad plan against the Patriots. Try to break it outside for that big highlight run, and you're likely to get smushed by a New England linebacker, or perhaps Rodney Harrison, for whom I have the utmost respect.
Westbrook is also a big part of the Philadelphia passing game, of course, and passes to Westbrook might find moderate success. New England did give up more yardage on passes to running backs than the average defense. But many of those long passes came in the fourth quarter when the Patriots had a large lead and were playing a loose defense. The Patriots did not give up a single touchdown on a pass to a running back.
McNabb's next target after Owens and Westbrook is usually one of his tight ends, Chad Lewis or L.J. Smith. The Eagles like to move Smith around to create matchup problems for opposing defenses, but that strategy will be less effective with Lewis sidelined by a foot injury, and more difficult against the Patriots' flexible 3-4 defensive alignment. Thanks to the coverage ability of their linebackers and safeties, the Patriots' defense was the league's best against tight ends this year.
McNabb's ability to read defenses will be severely tested by New England's multiple-look 3-4 alignment. McNabb's two worst performances of the season may have come in his two games against AFC teams playing a 3-4 defensive scheme. In a 15-10 win over Baltimore, McNabb passed for 219 yards on 33 attempts, only 6.6 yards per attempt -- and nearly half those yards were on passes to Owens (eight catches, 101 yards). The following week, Pittsburgh handed the Eagles their only loss prior to clinching the NFC's top playoff seed, and McNabb passed for only 109 yards on 24 attempts, a miserable 4.5 yards per attempt.
Because no teams in the NFC play the 3-4, the Eagles do not have much experience against teams where the fourth pass rusher varies from down to down. Right tackle Jon Runyan, currently the weak link on the offensive line because of nagging injuries, will have to face pass-rushing linebacker Willie McGinest, who has a history of coming up big in big games. This is where the Eagles will miss Lewis the most, as he is an experienced pass blocker who could pick up a linebacker or double a defensive end. Smith, on the other hand, rarely stays in for pass protection, and Jeff Thomasson, Lewis's replacement, was more of a pass-catching H-back than a blocking tight end in his last go-around with the Eagles.
(Note added Fri. afternoon: As an Eagles fan pointed out in the comments below, I neglected to mention the overall strength of the Philadelphia offfensive line here. As discussed in the first two Philly playoff previews, the left side of the offensive line is very good. Tra Thomas and C Hank Fraley are outstanding veterans and second-year LG Artis Hicks is a future Pro Bowler. I think any trouble blocking the Patriots will likely come from the right side of the line.)
Making things even more confusing, the Patriots will diverge from the 3-4 when circumstances warrant. Sometimes they'll look like a 4-3, or they'll use five or even six linemen to stop the run. Sometimes they'll put out two linemen, or no linemen and six linebackers. Depth allows New England to run unique defensive schemes and shift them constantly.
The Patriots shocked many observers by often rushing only three linemen both against Indianapolis and against Pittsburgh, and rarely blitzing. The Patriots' defense is more effective when it sits back and gives its intelligent linebackers and defensive backs time to react to the opposing quarterbacks. Expect similar strategy against Philadelphia, especially with the return of Pro Bowl defensive lineman Richard Seymour from a partial tear of the MCL in his left knee. MCL strains hamper lateral movement, but this shouldn't prevent Seymour from taking on his usual double teams. Jarvis Green was impressive in Seymour's absence and will be even stronger now that he and Seymour can rotate every few downs and stay fresh.
If the Patriots have one weakness on defense, it's their inexperienced secondary, which made them susceptible to big games by opposing second and third receivers in 2004. But the Eagles obtained Owens in the offseason precisely because their other receivers were so poor. Todd Pinkston in particular is known for shying away from contact, and in case you have not heard, Rodney Harrison is feeling a little disrespected. This is not a good combination for the Eagles.
What about McNabb's ability to pick up first downs with his feet? Well, the Patriots did show in the AFC Championship Game that they will sometimes give up big gains to a scrambling quarterback, as Ben Roethlisberger had four runs of nine yards or more. Of course, he had fallen behind by two touchdowns before he had the opportunity.
|PHI OFF vs. NE DEF DVOA by Down|
|PHI OFF||NE DEF|
|(PHI does not include Weeks 16-17)|
The Eagles will need to gain big chunks of yardage on first and second down, or they will be in trouble. Philadelphia's offense is much stronger on first and second down than on third down when compared to the average NFL team, and New England's defense is one of the league's best on third down.
If the Eagles offense has a game plan that can penetrate the New England defense, it will show immediately. The Eagles averaged 7.4 yards per play in the first quarter; no other offense in the league, including Indianapolis, averaged more than seven. The average NFL team had five turnovers in the first quarter this season; the Eagles had only one. But the Patriots had the league's best first quarter defense, allowing only 5.0 yards per play with a league-high 11 takeaways.
So by the end of the first quarter, it should become clear whether we're headed for a close contest or a New England rout. The Eagles have used an early lead to dictate strategy all season, so it is difficult to imagine them winning if their first two or three drives are unsuccessful.
|NE OFF||PHI DEF|
|DVOA||26.3% (4)||-3.5% (13)|
|TREND*||26.0% (2)||-16.1% (3)|
|PASS||42.1% (2)||-7.2% (10)|
|RUSH||12.2% (5)||0.7% (20)|
|RED ZONE||16.7% (7)||-8.2% (14)|
It is often said that the best game plan is to take away whatever your opponent does best, forcing them to play a style that emphasizes their weaknesses. But while the public still believes that New England's offense rides the coattails of its defense, the Patriots have actually developed into a remarkably balanced top five offense with no weaknesses. That's why this Patriots team is better than its two Super Bowl-winning predecessors.
According to our DVOA ratings, only Indianapolis had a more successful passing game than the Patriots this season. Play a zone on their multiple receiver packages, and quarterback Tom Brady will find the holes. Play man coverage, and the Patriots receivers will beat your corners. Blitz, and the Patriots will run short timing routes, screens, and draw plays. Concentrate too much on pass coverage, and they will gash you with Corey Dillon and a running game that ranked fifth in DVOA this season.
The hallmark of the New England offense is Tom Brady's grace under pressure and ability to make smart decisions with opposing defenders bearing down on him in the pocket. The hallmark of the Philadelphia defense is an intense pass rush, which can pressure a quarterback with just the defensive line and then turn up the heat with defensive coordinator Jim Johnson's complicated blitz schemes. The Eagles have to win this battle if they are to win this game.
The Eagles keep their pass rush fresh with a rotation of interior linemen that includes Corey Simon, Darwin Walker, Hollis Thomas, and Sam Rayburn. Though he does not start, Rayburn in his sophomore season has emerged as the most consistent of these players, the best against the run and second on the team in sacks.
But the most important player on the Philadelphia defense is the fourth Pro Bowler, veteran middle linebacker Jeremiah Trotter. After their loss to Pittsburgh at midseason, the Eagles re-inserted Trotter into the starting lineup, and his intensity has turned Philadelphia into a more complete and physical defense. The Eagles will often swarm against an opposing runner or receiver, taking him down with several tacklers instead of just one.
Philadelphia's five best defensive performances have come during the second half of the season, and Trotter's presence has helped the Eagles finally solve their run defense problems after a season-and-a-half of mediocrity. Philadelphia allowed 4.9 yards per carry to opposing runners for the first nine weeks of the season. Since Trotter joined the starting lineup, and excluding the final two games of the season where the Eagles rested most of their players, they have allowed only 3.7 yards per carry.
Stopping Corey Dillon won't be quite the same as stopping Atlanta and Minnesota, however. Those teams were strongest running up the middle, right into Trotter; the Patriots are strongest when running to the sides. While Trotter is surprisingly nimble when pursuing runners on the outside, the Eagles are still susceptible to outside runs, particularly to the right where Jevon Kearse is an undersized pass rush specialist. And even when running up the middle, Dillon doesn't get stuffed behind the line of scrimmage often. When there is no hole, he generally manages to find a yard or two anyway, and the Patriots are one of the top offenses at converting runs on third-and-short.
If Dillon can find a few holes early and stay away from Trotter, that will set up the play-action pass, leading to hesitation by the defense and more time for Brady to do what he does best: find an open receiver. The Patriots will change offensive formations and personnel constantly, and run every kind of pass route imaginable, but they've had great success this season throwing long. Brady might not have the strongest arm in the league, but he's incredibly accurate when throwing it deep. New England was sixth in the league in the percentage of pass receptions that went for 20 yards or more.
Neither David Givens nor David Patten is extremely fast, but both have fairly high per-catch averages because Brady can hit them in stride even when they're 20 or more yards downfield. And the Patriots have thrown deep even more often since Brady's favorite receiver, Deion Branch, returned from an early-season injury, and Brady hooked up with Branch for a 60-yard touchdown and a 45-yard completion in the AFC Championship Game.
The success of the Brady-Branch connection plays right into the two weaknesses of the Philadelphia pass defense. Though above average against all other types of receivers, the Eagles were one of the worst pass defenses in the league against number one wideouts. That's not just Randy Moss: the Eagles gave up 100 yards in one game to Laveranues Coles, 116 yards to David Terrell, and a season-high 80 yards to Travis Taylor.
And despite the presence of three Pro Bowlers in the secondary -- cornerback Lito Sheppard, strong safety Michael Lewis, and free safety Brian Dawkins -- Philadelphia can be beaten deep, as Marcus Robinson showed in the divisional round when he beat Lito Shepard for a 40-yard gain. Back in Week 12, they were burned for two 50-yard gains in the same game by a Giants receiver, Jamaal Taylor, whose previous NFL career consisted of four receptions.
This vulnerability results in part from Philadelphia's complicated blitz schemes. Sheppard and Lewis may be a tad overrated, but Dawkins could be the best defensive player in this year's Super Bowl (taking nothing away from Rodney Harrison, for whom I have the utmost respect.) He is near-flawless in pass coverage, strong against the run, and dangerous when he comes as a pass rusher. With Dawkins or Lewis so often coming after the quarterback, however, the Eagles will sometimes leave a big hole in the middle of the field -- if the opposing quarterback who can find it in time.
The Eagles blitzed less in the first two rounds of the postseason than they did during the regular season, preferring to pressure with just the front four. A favored play when they do blitz is the overload blitz, trying to send in one more pass rusher than the Patriots have blockers on one side while a defensive end on the other side drops into zone coverage to help prevent a quick pass. This tactic will certainly test both New England running backs, Dillon and Kevin Faulk, who are excellent at blitz pickup.
During the regular season, only four teams allowed fewer quarterback sacks than the Patriots, whose strong offensive line is led by the twin Purdue alumni, left tackle Matt Light and right tackle Brandon Gorin. But the Patriots have trouble when tackles are left one-on-one against elusive pass rushers, and Philadelphia's Jevon Kearse resembles another pass rusher who causes the Patriots problems, Indianapolis's Dwight Freeney. The Patriots deal with Freeney by "chipping" him with a receiver in motion or a tight end going out on a pass route, or will simply keep tight end Daniel Graham in to block. They will likely deal with Kearse the same way, whether he is lining up in his usual position opposite Gorin or moving around like he did against Atlanta.
The Philadelphia pass rush will not have the same effect on Brady that it has on most quarterbacks, but the Eagles should still be able to sack him a couple of times and hurry him on a few throws. Picking off one of those hurried throws or forcing a fumble on one of those sacks gives them their best chance to win the game. Bringing so many pass rushers that the New England receivers are open for big gains is the surest way to lose it.
|DVOA||0.2% (16)||7.4% (3)|
|NE kickoff||-4.8 (24)||-1.6 (20)|
|PHI kickoff||7.2 (8)||15.0 (1)|
|NE punts||-0.5 (23)||-5.2 (16)|
|PHI punts||-16.2 (29)||19.3 (2)|
|FG/XP||15.2 (1)||13.7 (2)|
This is the one area where Philadelphia enjoys a significant advantage over New England -- in part because, contrary to conventional wisdom, the best kicker in the NFL is not New England's Adam Vinatieri but Philadelphia's David Akers. Yes, Vinatieri has a history of clutch kicks in the playoffs, but the only reason Akers does not have a similar history is that he has never had the opportunity.
Vinatieri and Akers were the top two field goal kickers in football this season, worth 15.2 points and 13.7 points, respectively, above an average kicker in the same situations. But Akers is consistently one of the league's top kickoff men while Vinatieri is only mediocre. This will help the Eagles to neutralize the threat of Bethel Johnson, New England's nimble kick return man.
Much has been made of the fact that Brian Westbrook will return punts in the Super Bowl, returning to the role he played in the 2003 season. But while it is true that Westbrook's average punt return last year was six yards longer than the average Philadelphia punt return this year, he also fumbled three of the 36 punts he fielded, and two of those were recovered by the Eagles' opponents. The Eagles use two players to return kickoffs, and J.R. Reed is more dangerous than Roderick Hood (although Hood is excellent in his other special teams role, tackling opposing returners).
Philadelphia's biggest special teams advantage comes when they have to punt to New England. Dirk Johnson is a fine punter, but the main reason why the Eagles ranked second in the NFL in the value of field position gained when punting is their ferocious punt coverage. The Patriots' punt returns, meanwhile, have been atrocious all season.
A typical Philadelphia punt followed by an typical New England return will lead to the Patriots starting their next drive roughly six to eight yards behind the NFL average. All it takes is a few yards of starting field position to turn a touchdown drive into one that stalls at a field goal. If the Patriots throw long, however, that helps counteract any lost field position from poor punt returns. And since special teams are the most variable part of football, even the least likely team can surprise with a big return.
Even if it did look like the Eagles and Patriots presented an equal matchup on paper, New England would have to be favored in this game because of the general imbalance between the AFC and the NFC. The AFC had a 44-20 record against NFC teams this season, the widest disparity in interconference games since 1979. According to our DVOA ratings, 12 of the top 15 teams in football this season were AFC clubs, and Philadelphia was the only NFC team in the top ten.
Philadelphia went 2-2 this year in interconference games. They were demolished by Pittsburgh, narrowly beat Baltimore, and got taken to overtime by Cleveland, arguably the worst team in the AFC. Even in the two games where they sat their starters, the second-stringers played better against St. Louis than against Cincinnati. As noted above, two of Donovan McNabb's worst performances of the year came against AFC opponents. So did Westbrook's two worst games of the year. Against Cleveland, he ran for only 43 yards on 13 carries and had just 17 yards on three catches. After missing the Baltimore game with an injury, he returned against Pittsburgh, with only 17 yards on six carries and just three catches for four yards.
Philadelphia's struggles against the AFC are not new. Last season the Patriots walloped the Eagles 31-10. Philadelphia won its other three interconference games, but they beat the Jets and Dolphins by only a touchdown and Buffalo by just ten points. Back in 2002, the Eagles got blown out by the Colts 35-13 at home and narrowly lost to Tennessee and Jacksonville. They did manage to beat one AFC team that year, the expansion Houston Texans. Yippee. All four of these games, by the way, featured Donovan McNabb at quarterback, before the injury that cost him the final six games of the 2002 season. Heck, not like it even matters, but the Eagles have lost five of seven games to AFC teams in the last two preseasons.
Here's a table to show just how large the gap has been between Philadelphia's games against NFC opponents, and their games against AFC opponents. To be nice, we won't even count the last two games of this season.
|Philadelphia Eagles vs. NFC and AFC, 2002-2004|
|Avg. Score Margin||Offense DVOA||Defense DVOA|
|Year||W-L||vs. NFC||vs. AFC||vs. NFC||vs. AFC||vs. NFC||vs. AFC|
|Regular season only. 2004 excludes Weeks 16-17.|
In a sport where games can be decided by the random bounce of a fumble, even the greatest underdog has a chance of winning the Super Bowl. And the Eagles are not the greatest underdog. Despite their subpar performances against the AFC, the Eagles were statistically the equal of the Patriots and Steelers until the final two regular season games when they played primarily second-stringers. A Philadelphia win would be far less shocking than New England's 2001 victory over St. Louis, or Carolina's near-victory last season.
But the more likely outcome is a Philadelphia fan's worst nightmare, with Donovan McNabb stuck trying to be a one-man team, grabbing an occasional first down on a scramble or a pass to Westbrook and, if he's lucky, hitting Greg Lewis deep for one solitary big gain. The Philadelphia defense has been outstanding over the season's second half, but eventually the balanced offense of the Patriots will find something that works against them.
Unless Terrell Owens comes out with a red spot on his sock and jersey number 38, Super Bowl XXXIX will probably be the third won by the New England Patriots, and their first championship decided well before the final seconds.
If you arrived at this preview from an outside link, and are interested in more Football Outsiders analysis of Super Bowl XXXIX, please check out the following links:
A third Super Bowl Debate Thread will go up on Sunday afternoon for discussion during and after the game.
*TREND numbers are updated through the playoffs, all other DVOA statistics are regular-season only unless otherwise noted.