TCU has played much better in the second half of games this year. What other schools have seen dramatic shifts of play after halftime?
20 Sep 2005
by Ned Macey
One week after making this column as an upset loser at the hands of New Orleans, Carolina is back with a huge win over the two-time defending NFL champion New England Patriots. The Patriots had gone 32-2 over their past 34 games, a run that included two Super Bowl championships. For any other team, losing a road game against a supposed Super Bowl contender would not be a major concern, but for the Patriots it is seemingly cause for alarm. The loss could be easily written off as a game of mental mistakes by the Patriots as they committed 12 penalties and turned the ball over three times. Still, as they head into a Week 3 clash with the Steelers (in Pittsburgh no less), the Patriots have to be worried about their running game. As for the Panthers, the win highlights a reason -- besides the commonly cited improved luck with injuries -- that they could be a force in the NFC.
Before we get into the specifics, the most salient observation I took from the game was how amazing New England's run has been. Luck is a factor in all NFL games, and in any game the inferior team can win due to lucky breaks or exceptional big plays that are unlikely to be repeated. Consider this game, where the Panthers used a fumble recovery and a punt return to start two drives inside the 15-yard-line. John Kasay was two for two on 50+ yard field goals. The Patriots had the second through fifth longest drives of the game but finished two of those drives with turnovers. The biggest referee call of the game awarded an apparently undeserved touchdown to the Panthers.
Almost every week, losing teams can point to facts like these as a reason that they lost, but for the past 35 games, this is only the third time the Patriots have lost. In that time, the Patriots are now 17-2 in games decided by ten points or less. In all games that Tom Brady has started (including playoffs), the Patriots are 32-7 in games decided by ten points or less and "only" 25-7 in games decided by more than ten points.
That's enough Pats admiration given that they actually lost this week, so let us turn our attention to the principal cause of the defeat, the lack of a running game. That the Patriots' running game improved after switching from Antowain Smith to Corey Dillon is self-evident, but look at the impact on the passing game:
Which one of these years is not like the other? While the raise in actual DVOA can be partially attributable to the increased offensive environment of a year ago, the jump in rank from middle of the pack to second overall is clearly an indication that last year was aberrant. As we state in our methods: â€œWhen we say, â€˜Priest Holmes has a DVOA of 17.6%,' what we are really saying is â€˜Priest Holmes, playing in the Kansas City offensive system with the Kansas City offensive line blocking for him and Trent Green selling the fake when necessary, has a DVOA of 17.6%.â€? While Brady likely improved as a quarterback a year ago, the increased attention defenses played to the running game allowed a slightly improved Brady to have a much more productive season.
The problem so far is that through two games Corey Dillon has looked a lot more like Antowain Smith than Corey Dillon. While two bad games is not the end of the world (and that goes for all you who followed our advice on Kevin Jones), Dillon was at risk even coming into this season. Aaron wrote frequently about the impending decline of Curtis Martin, and what he says for Martin is almost equally true of Dillon. A year ago, Dillon set a career high with 345 carries in the regular season. Add in 65 in the postseason for 410 total, and you have far too many carries for any back, let alone one turning 31 next month. Through two games against teams that ranked 13th and 19th in run defense DVOA a year ago, he has rushed 37 times for 99 yards. This is a troubling trend for a player who had a fairly negative KUBIAK projection of 1226 yards and 4.2 yards per carry.
Another explanation for the offensive struggles is the departure of Charlie Weis. Since I wrote the article in Pro Football Prospectus arguing that such a loss is unlikely to impact them, I am not going to change my opinion after two games. The central point of that article is that units that lost their coordinators to head coaching positions declined no less the next season than top units that retained their coordinators. (For fans in Minnesota, I did not study coordinators that made a lateral move). While the Patriots' offense is clearly off to a slow start the season, I would argue it looks no worse compared to last year than their fellow residents in the top five of 2004 DVOA: Indianapolis, Kansas City, Minnesota, and the New York Jets.
While Patriots' fans can take solace in the fact that they won two Super Bowls without Dillon, it is impossible to deny that last year's team was the best team yet, despite a defense that had some holes. Their overall DVOA of 35.6% was 13.2% better than the year before, or roughly the difference between the Jets and the Bengals a year ago. Had the Patriots maintained their 22.4% DVOA of 2003, which was second in the league that season, they would have ranked seventh last season and third in their own division. With the Colts and Steelers looking to be as good as a year ago, the 2003 version of the Patriots may not be good enough to deliver another title.
As for the Panthers, this victory was a serious statement by a defense that looked very ordinary the week before. Crucial to this victory was the Panthers defense's ability to get the Patriots off the field on third down. They held the Patriots to 4-14 on third down, quite a serious departure from a year ago. Patriots' fans can point to stupid penalties (six false starts) and a weak running game leaving them in many third and longs. This is all true, but for the Panthers a year ago, third and long was exactly where they did not want to be.
As we have mentioned on numerous occasions, a third down performance that is out of line with a unit's performance on other downs is unlikely to be repeated. A year ago, the Panthers' defense on first down had a DVOA of -16.0%. On second down, it was -19.6%. If they matched this performance on third down, they would have had a DVOA of -17.6%, second best in the league. Instead, thanks to a third down DVOA of 33.4%, they finished 10th in the league with a DVOA of -7.1%.
While the Panthers were below average on third downs of all sorts, they were particularly bad on long third downs, with a DVOA of 116.1% on third downs of seven to ten yards and 53.2% on third downs greater than ten yards. Given that the Patriots were exceptional on third down a year ago, this is a very encouraging sign for the Panthers, albeit one that was easily predicted.
The other key to the Panthers' victory was their ability to cash in on all three scoring opportunities in the red zone, something they have been unable to do in recent years. Fail to convert any of those, and the Patriots final drive that ended in a Ben Watson fumble would have been a chance to win the game.
What is notable about it from the Panthers' perspective is how much they have struggled to run the ball inside the ten yard line the last few seasons. Last year without Stephen Davis, they had a DVOA on runs inside the ten of -10.4%. The year before, with Stephen Davis, they were infinitely worse with a DVOA of -117.2%. Of course, had Davis's fumble been properly called, they would have been headed in that direction again. Given the Panthers' predilection for playing close games, improved efficiency at the goal line will lead to more wins.
What is also impressive is the way that they turned the Patriots' weaknesses against them. The Patriots have been an excellent red zone team in recent seasons with a DVOA of -20% in the red zone a year ago and -30.5% the year before. While they excelled in defending all red zone pass plays as well as runs between the 10 and 20 yard line, they have been below average defending runs inside the 10 yard line both of the last two seasons. Switching out Tedy Bruschi for Monty Beisel should only exacerbate this problem.
With the win in hand, the Panthers are in a much better position than their Super Bowl XXXVIII opponent. The Panthers play the Dolphins, Packers, Cardinals, Lions, and Vikings over the next six weeks. Provided Jake Delhomme starts playing a game resembling football where the goal is to throw to your own team, it is hard to imagine them losing more than one of those games. At that point, they would have a very interesting Week 9 confrontation with the resurgent Buccaneers.
The Patriots just lost the first game of what was certain to be a brutal stretch of games. Their next six games are against teams with winning records a year ago, and after hosting San Diego, they have to travel on back-to-back Sundays first to Atlanta and then to Denver. Four of their next five opponents ranked in the top eight in rushing defense DVOA a year ago, so getting Corey Dillon back to his 2004 level will be a very difficult task. I would not advise betting against the Pats yet, but chinks in the armor have definitely been exposed.
43 comments, Last at 23 Sep 2005, 8:23pm by Wilbur