TCU has played much better in the second half of games this year. What other schools have seen dramatic shifts of play after halftime?
by Ned Macey
|It's all-Philly day at Football Outsiders! The Pro Football Prospectus 2005 book tour hits the Philadelphia suburbs tonight at Chester County Book & Music Company in West Chester, PA. Book signing and Q+A with Aaron Schatz and Mike Tanier at 7:30pm. This article is being edited thanks to wi-fi access at Cosi's at 12th and Walnut.|
Most of the contributors to Pro Football Prospectus did not get to see the completed book until it arrived in our mailboxes a few weeks ago. Once I had read everybody else's chapters, my single strongest impression about the upcoming season was that the Eagles were far and away favorites in the NFC. Based on last year's results, I guess this should be obvious. The gap in DVOA (explained here) between the Eagles and the second highest NFC team, Carolina, was 24.5%, roughly the difference between DVOA leader New England and the tenth best team, Kansas City.
The Eagles also have underlying statistical indicators that would indicate improvement. Fumble recoveries are random, attributable primarily to luck, and the Eagles were in the bottom five of the NFL both recovering their own fumbles and their opponents' fumbles. They also had an offense worse on third down than on the first two downs, usually an indicator for future improvement. While Todd Pinkston is out with an injury, the Eagles had almost no off-season personnel subtractions, losing only two starters, Jermane Mayberry and Derrick Burgess.
They seem so strong that our projections have them with an 80% chance of winning at least 11 games and only a 1% chance of winning fewer than seven. The Patriots, on the other hand, have only a 36% chance of winning 11 games and a 17% chance of winning fewer than seven. No other team has even a 55% chance of winning 11 games or under a 10% chance of winning fewer than seven.
With the Eagles seemingly invincible, the most significant obstacle to overcome may be the supposed Super Bowl Loser's Curse. Since the 1998 Falcons made it to the Super Bowl and then followed it up with a 5-11 campaign, only the 2000 Titans have followed up a Super Bowl loss with a playoff appearance. The six teams who have tried to defend their Conference Championship since that date have averaged seven wins in their next season.
Some of this is certainly attributable to basic regression to the mean, but here is a chart with Super Bowl winners, Super Bowl Losers, and Conference Championship game losers with their average DVOA in the season they made it to that point and in the following season.
|Final Result||DVOA||Next Season|
|Super Bowl Champ||28.8%||15.3%|
|Super Bowl Loser||21.1%||0.4%|
|Championship Game Loser||20.0%||13.6%|
Obviously such a small sample size means that any wild data points completely skew the results. In an effort to see what may lie ahead for the Eagles, let us look at the fate that befell the previous seven Super Bowl losers. The reasons behind the declines may hint at why Super Bowl losers tend to struggle the following season.
The 1998 Falcons must be considered one of the biggest flukes in recent football history. In 1996 and 1997, they combined for 10 wins. In 1998, they won 14 games and had the best team in football according to DVOA. In 1999 and 2000, they combined to win nine games. Much of the decline is usually blamed on the ACL injury to Jamal Anderson. After rushing for 1,846 yards in 1998, he was limited to 19 carries the next season. The team's leading rusher was Ken Oxendine, who split carries with Byron Hanspard. They averaged about 3.0 yards per carry combined.
The lack of running game crippled the passing game, where Chris Chandler went from averaging 9.6 yards per attempt to 7.6 yards per attempt. The departure of Tony Martin left the team without a legitimate second receiver. The overall offense tanked from a DVOA of 8.8% to -12.5%.
To make matters worse for the Falcons, the defense declined almost as much as the offense. The defense went from a DVOA of -19.2% to 0.8%. The defense faced no major changes, other than the departure of the aging Cornelius Bennett. Really, the 1998 Falcons are a major data point in the truism that defense is much more variable than offense. In 1997 (before we have play-by-play data to create DVOA ratings), the Falcons ranked 21st in the league in scoring defense. After ranking 4th in 1998, they fell to 25th in 1999.
The Titans are the one team in recent history that has avoided the Super Bowl Loser's Curse. Instead, the 1999 Titans were building a great team and reached the Super Bowl before their time, much like the 2001 Patriots. The 1999 Titans were a growing team with a still-developing Steve McNair and young defensive players. The "Music City Miracle" allowed them to reach the Super Bowl the year before they actually peaked.
The emergence of young talented players such as Jevon Kearse and Samari Rolle made the Titans the second best defense in football in 2000, and they had the highest overall DVOA in the league. Ironically, injuries to McNair limited his development, and the offense declined slightly in 2000. The Titans, of course, fell to the Ravens in the divisional round in a game in which Al Del Greco missed three field goals, including one blocked that was returned for a touchdown. The Titans proceeded to cut Del Greco and spend their second round pick on a kicker ... oh wait, they picked up Joe Nedney for peanuts.
The 2000 season was much like the 2004 season in that the AFC was much stronger than the NFC. Last year, the Eagles were the only NFC team in the top 11 in DVOA. In 2000, the top 5 were all AFC teams, and only two NFC teams were in the top nine. The Giants were neither of these teams, ranking 13th overall and fourth in the NFC. Fortunately for the Giants, the two highest ranked NFC teams, Tampa Bay and St. Louis, got bounced in the first round by inferior teams.
The 2000 Giants were a very balanced team with an offensive DVOA of 7.7% and a defensive DVOA of -7.2%. The next year, the defense remained the same, while the offense fell off to -3.8% DVOA. Looking at personnel, you wonder how they achieved the 7.7% in 2000, as Ron Dayne led the team in carries. In 2001, they saw the emergence of Joe Jurevicius as a legitimate third receiver, and their running game remained roughly the same below-average unit it was in 2000.
The big difference was a pass offense that went from a DVOA of 24.2% (4th in the league) to -2.8% (14th). Most of this drop is attributed to Kerry Collins's inevitable decline after a career year. In 2000, he ranked sixth in the NFL in DVOA, while in 2001 he was 19th. In 2000, he had 22 touchdowns and 18 turnovers. In 2001, he had 18 touchdowns and 21 turnovers. The other major factor was the decline of Amani Toomer, who dropped from third in the league in DPAR to 28th. A year after catching 62.9% of the passes intended for him, he snagged only 50.7% in 2001. This decline in passing was the difference between a good team that made the Super Bowl by taking advantage of weak opponents and a mediocre 7-9 team.
The rise to greatness of the Patriots over the past couple seasons will lead history to forget that the Pats' victory over the Rams in the 2001 Super Bowl is one of the greatest upsets in NFL history. The Rams had won the Super Bowl in 1999. In 2000, they won 10 games and scored 540 points. In 2001, they were 14-2. Their 39.6% DVOA was comfortably the best in football. The Patriots were 11-5 coming off a 5-11 season. Their DVOA was -1.4%, 16th best in the league.
The 2001 Rams were a truly great team, so one wonders if there was any way to predict the Rams' fall to 7-9 the next season? Faulty memories may blame it on the injuries that drove Kurt Warner out of the lineup, but the Rams were actually 0-5 in the games where Warner got the majority of the snaps, leaving them 7-4 without him. When Marc Bulger got the majority of the snaps, the team was 6-0. Their defense did regress from a DVOA of -20.8% to -5.1%, but that was still the ninth best defense in the league.
The biggest problem was that their offense dropped from 25.6%, first in the league, to -5.7%, 25th in the league. Almost all of this decline was Kurt Warner's mysterious fall from grace. And that mysterious fall, well, remains rather mysterious.
Of the six teams in this study, the Raiders were the best team the year before they made the Super Bowl and the worst team the year after they made the Super Bowl. The Raiders' Super Bowl berth proved to be the last hurrah of an aging group of players. They had lost the AFC Championship Game in 2000 when Rich Gannon got hurt. In 2001, they lost the infamous "Tuck Bowl." In 2002, without Jon Gruden, they had their best season.
Much of this was credited to a short-passing attack expertly run by Gannon. The next season, Gannon completely fell off, watching his DVOA drop from 25.5% (3rd) to -7.5% (26th). When he went out for the season in Week 7, the Raiders were a dismal 2-5. Without an unknown gem like Marc Bulger on the bench, things only got worse from there. After backup Marques Tuiasosopo went down, the Raiders were forced to rely on Rick Mirer and his predictably substandard -11.4% DVOA.
Unlike with the Rams, the problems were not all with the quarterback. A Week 1 injury to Jerry Porter left the receiving corps without any deep threat. This problem was exacerbated by Father Time finally catching up with Tim Brown and Jerry Rice. Brown failed to catch 75 passes for the first time since 1992, hauling in only 52. Rice caught only 63 balls, his lowest total in a full season since his rookie year in 1985. Both Brown and Rice's careers as productive players were effectively over.
While the team's offense was the bulk of the decline, the Raiders defense also declined severely, from a DVOA of -8.9% (5th) to 8.7% (25th). Again, much of this was due to age. Bill Romanowski's career was ended. Rod Woodson missed six games and retired after the season. Trace Armstrong was injured, ending his career. Sam Adams and Tory James had moved on to finish their careers elsewhere, already hurting their defensive depth.
Although the Panthers went from 11-5 and three points away from Super Bowl champions to just 7-9 the next year, according to DVOA they were actually better in 2004 than in 2003. The only team they beat in 2003 with a winning record was the Indianapolis Colts. Last season, the Panthers were decimated by injuries: Stephen Davis, DeShaun Foster, Steve Smith, and Kris Jenkins. Still, the defense was actually better according to DVOA, and the offense was above average for the first time this decade (albeit at 0.1%).
The Panthers in fact are a good representation of the common myth surrounding Super Bowl runners-up. They took advantage of a weak schedule to propel them into the playoffs and used three hot weeks to get to the Super Bowl. Injuries may have prevented them from taking a giant leap forward, but it is hard to call the Panthers' missing the playoffs the result of the curse given that they actually became a better team.
Adding it all together, the reasons for the Super Bowl curse do not seem to be so much a mysterious hex as predictable flaws. The 1998 Falcons turned out to be a fluke, and a massive injury to Jamal Anderson exacerbated the problems. The Rams and Raiders suffered from the decline of their quarterbacks, with the Raiders adding in a team-wide breakdown caused by an aging nucleus. The Giants and the Panthers were just not great teams: mild decline by the Giants left them out of the playoffs, and mild improvement by the Panthers was not enough when their luck evened out.
The Eagles do not seem to have any of the issues named above. They have been a consistently excellent team for the past five seasons and are unlikely to be one-year wonders like the Falcons. They were much better than the 2000 Giants or 2003 Panthers, so a mild decline would still be enough to get them into the playoffs.
The two more comparable cases are the Raiders and Rams, teams who had been very good for several seasons and excellent the year they made the Super Bowl. The Raiders fell apart because of an aging roster. The Eagles, thanks to shrewd management, have not held the same aging core together. Instead, they have constantly been reshuffling, bringing in and developing young players and letting older players leave via free agency. Their oldest projected starters this year are Brian Dawkins and Terrell Owens, both of whom turn 32 this season.
We cannot yet project decline for individual defenders, but Dawkins certainly still appears to be very good. For offensive players, we can run similarity scores to get an idea of their upcoming contributions. Fears about Owens aging seem very misplaced. If you look at Owens' three-year comparables, the first name on the list is Jerry Rice after the 1992 season. The next year, all Rice did was catch 98 balls for 1503 yards and 15 touchdowns. Of Owens's top 10 most similar players, all but two had at least 900 receiving yards the next season, and four had 1399 or more yards. Considering everyone on this list was at least 29 in his most similar season, Owens appears to be in good shape.
As for the major stumbling block for the Rams, the sudden decline of Kurt Warner, the Eagles seem to have less to worry about with Donovan McNabb. Back to similarity scores, McNabb's top comp is also one of the game's greats: Joe Montana after his 1984 season. Among McNabb's top 10 comps are Tom Brady and Jake Delhomme from last year, and we don't know what they will do this coming season, but out of the other eight, only Steve McNair in 2004 failed to throw for at least 3000 yards, and he is only the 10th best comp. Given that McNabb sat out two games last season, his overall stats likely would have been even stronger, pulling in an even more impressive list of comparable players.
Still, the fact is that since 1999 only the 2000 Titans have made the playoffs after losing the Super Bowl the previous year, and then they suffered one of the most painful losses in recent playoff memory. No easily predictable reason may exist for the Eagles to miss the playoffs, but here are three possibilities:
Injuries: The Falcons, Rams, Raiders, and Panthers all suffered a rash of injuries that contributed to them missing the playoffs. The Eagles have already lost Todd Pinkston. Another wide receiver injury, particularly to Owens, would leave a team that made the NFC Championship with James Thrash as their # 1 on the outside a couple years ago. Also, a couple injuries to the secondary would severely limit the ability of defensive coordinator Jim Johnson to be creative.
Internal Dissension: Cory Simon, Brian Westbrook, and Terrell Owens are all unhappy with their contracts. I think once the games start this will have no impact, but it is certainly not ideal.
Madden Jinx: Not only do the Eagles have to overcome the jinx of Super Bowl losers, but Donovan McNabb is on the cover of the premiere NFL video game. The players who have previously been on the cover have struggled individually or battled injuries. Also, not one team which had a player on the cover has made the playoffs. If McNabb and the Eagles go down, it may prove that all the statistical indicators in the world cannot overcome two powerful curses.