And This Weekend's Upset Victims Will Be …

The Geek's Guide to the NFL Playoffs

And This Weekend's Upset Victims Will Be …

The Geek's Guide to the NFL Playoffs

And This Weekend's Upset Victims Will Be …
The stadium scene.
Jan. 11 2008 11:50 AM

The Geek's Guide to the NFL Playoffs


Philip Rivers. Click image to expand.
Philip Rivers 

Ah, Philip Rivers. It has been hard to watch him struggle this year, because he is the poster boy for the Lewin Career Forecast, our system for projecting highly drafted quarterbacks based on college numbers. The forecaster projected him to be one of the top five or six quarterbacks in the league, and he certainly appeared to be on that path after he played so well in 2006. But you are absolutely right that he's looked jittery for most of the year, though this may turn out to have been a health issue. He's had a knee problem all year, but according to injury expert (and auxiliary FO writer) Will Carroll, the Chargers finally coached him to deal with it properly just a few weeks ago. Rivers is concentrating on setting his feet correctly, which has led to more consistency, which has led to four straight strong games. As for the strength of the opposition, the streak started with a good game against a terrible Detroit secondary, but for the year Rivers has faced a harder-than-average set of pass defenses. Despite the ups and downs, our ratings actually have Rivers ranked 13th in value among quarterbacks this year.

What about the San Diego defense? It's true that they've built their recent success against some atrocious offenses. (JaMarcus Russell's uniform actually said "TRAINEE" on the back.) Still, that's what a strong defense is supposed to do: shut down bad offenses and slow down good ones. And the Chargers did play some good ones earlier in the year: New England, Indianapolis, Jacksonville, and Minnesota (OK, Minnesota is actually half a good offense). That made their defense look worse than it really was. The other change, of course, was pulling cornerback Drayton Florence from the starting lineup in Week 10 and replacing him with Antonio Cromartie, who made the Pro Bowl despite starting for only half the year.


You seem surprised that the Jaguars don't move Rashean Mathis around to cover the opponent's best receiver, which I think actually gets at one of the disagreements between our analysis and yours. You write a lot about matchups, and rightly so—fans often put too much stock in simple 1-to-32 power rankings and miss the details that determine how teams will play against one another. However, we've discovered that the specific matchup between one cornerback and one wide receiver isn't that big a deal, precisely because most offenses will move their wide receivers around on both sides of the field, while most defenses rarely switch sides with their cornerbacks.

In Pro Football Prospectus 2007, we looked at how often the offense's top receiver was the intended target when the defense's top cornerback was the defender in coverage. For the average team, the top cornerback covered the No. 1 receiver only 44 percent of the time! Even the cornerback most often used against the other team's top receiver—Green Bay's Al Harris—was covering that guy on fewer than 70 percent of charted passes.

Of course, while I disagree with you on the importance of the specific-corner-on-specific-receiver matchup, I don't disagree with you on how important it should be. I don't understand why defensive coordinators don't switch their cornerbacks more often when one is clearly better. That's clearly the case with four of this weekend's teams: Jacksonville (Mathis), New England (Asante Samuel), San Diego (Cromartie), and Dallas (Terrance Newman). Oddly enough, I don't think it is still the case in Green Bay. I love Al Harris, who was underrated for years, but this year's Pro Bowl selection was like a lifetime-achievement award. Even though his team was winning, our numbers show he's clearly having an off year. Harris and Woodson are definitely a top cornerback tandem, as you noted before, but Woodson was the better of the two in 2007, and I would be very surprised if your metrics did not say the same.

With all this intricate matchup analysis, readers are probably wondering where to find the shortcut to our picks. I generally fight against the pressure to make binary picks, to say straight out that one team will beat another. Every team has a shot to win every game, which is why the phrase "any given Sunday" exists. I think the home teams are properly favored, so I'll pick the games by ranking the upsets: from least likely, to most likely.

The least likely upset is San Diego over Indianapolis. Remember that when Manning threw six interceptions against the Chargers back in November, he was throwing to guys like Craphonso Thorpe and Devin Aromashodu. The Colts are now healthy, but Antonio Gates is not, and Peyton Manning should shred the secondary—the Chargers have only one good cornerback (Cromartie), and the Chargers safeties are a huge weakness.

Jacksonville is our next upset candidate. Yes, the Jaguars have been ridiculously hot over the last few weeks, and their running game is swell, but a lot of people are talking them up without really thinking through the matchup of the Patriots offense and the Jaguars defense. (As FO writer Mike Tanier said to me today, "It's a lot easier to sound witty and interesting when you are being contrary than when you are picking chalk.") The Patriots offensive line is one of the best in the league, especially run-blocking up the middle, while the Jaguars defense gives up a lot of yardage on runs up the middle. You and I both agree that the Patriots can pick on cornerback Brian Williams, and the Jaguars struggled against slot receivers this year, which is not a good weakness to have when you're playing against Wes Welker. If they blitz Brady and end up with a safety or linebacker manned-up on Welker or Jabar Gaffney, it will not be pretty.

I would put Seattle third. The Seahawks are a good team with an underrated front seven, but the Packers are just a little better in every other area. And that means that I'm picking the Giants as the team most likely to pull the upset this week. Two weeks ago, I never would have believed I would be writing those words, but the late-season decline of the Cowboys just scares me, especially because it is tied to a couple of injuries that may not be healed at this point: Tony Romo's thumb and Terrell Owens' ankle. If those guys are both healthy, the Cowboys should win easily. If Owens is out and Romo is still ineffective, the Cowboys are in serious trouble—and the winner of the Seattle-Green Bay game can start reserving hotel rooms in Arizona.

Who do you like, KC?