In this week's Varsity Numbers, Bill Connelly revisits some measures and concepts: Adjusted Scores, Covariance, and momentum (or whatever you choose to call it).
05 Jan 2007
by Aaron Schatz
You know that conventional wisdom about how important it is to stop the run in the NFL? Don't tell the teams that made the AFC playoffs. The two worst run defenses in the league will both be on display this weekend. One team has an offense that can overcome this weakness, the other does not.
For those who may be visiting this site for the first time to read this preview, we explain our stats at the bottom of the page, or click this link. Each preview also includes a link to the game discussion thread for that game. We're doing separate game discussion threads for each game this year, rather than combining both games on the same day like we did last year.
NFL broadcasts are promoted with simple, easy-to-hype storylines. Most of the time, these storylines focus on meaningless personality issues and dumb down the complicated strategic battle on the field.
This is not one of those times.
The Colts have the best offense in the league, but the Chiefs love to run, and the Colts defense couldn't stop the run if you let them put 20 guys on the field. That's it. That's the whole game right there.
Kansas City's Larry Johnson set a new NFL record by carrying the ball 416 times this year. He started slow, in part because Kansas City made a number of changes on the offensive line, but he averaged 4.5 yards per carry over the final 11 games. He's both large and agile, able to pound into defenders to gain extra yardage or nimbly cut back into an open lane.
Johnson presents the worst possible matchup for a Colts defense that allowed 5.3 yards per carry this season. Since the AFL-NFL merger in 1970, only the 1996 New Orleans Saints allowed more yards per carry. In fact, the Colts are only the fifth team since the merger to allow more than five yards per carry over an entire season.
Everything people say about the Colts defense is accurate. They are undersized. They cannot wrap-up on tackles. Defensive end Dwight Freeney is a great pass-rusher whose spin moves take him out of every running play, and teams run into the hole Freeney leaves behind. True, true, all true.
(Yes, despite all those yards allowed, the Colts did not actually have the worst DVOA run defense in the league. They don't even have the worst DVOA run defense on this page, because of opponent adjustments. They played the NFC East this year, and all four NFC East teams ranked among the top dozen run offenses.)
The Colts are counting on the return of run-stuffing safety Bob Sanders to help shore up their defense, but Sanders may not have the effect that the Colts are hoping for. Sanders missed most of the season to injury, but he did play in four games. In those four games, the Colts allowed 5.6 yards per carry -- worse than the 12 games with Sanders on the sidelines.
The Chiefs won't be using any trickery to try to fool the Colts defense. They aren't going to come in and shock everyone by going empty-backfield on their first two drives. They are going to run and run and run some more.
At some point, the Colts will manage to actually tackle Johnson after just a yard, and the Chiefs will have to throw the ball. But that's okay, the Colts' pass defense isn't very good either. The Chiefs mostly throw to their number one receiver, Eddie Kennison, and to one of the greatest tight ends to ever play the game, Tony Gonzalez. The Colts were an average defense against tight ends but were awful against number one receivers. The Tampa-2 thing is also huge problem against a team whose two main weapons are the running back and the tight end. If you put that middle linebacker back to defend Gonzalez in the seam, you're just giving away yardage to Johnson. And if you bring the safety and middle linebacker in to stop Johnson, who is going to cover Gonzalez on a play-action pass?
Also, note this FO FOX blog post on our early game charting stats. Although the data compilation is still incomplete, so far the Colts lead the league in passes marked with "Hole in Zone" rather than with a specific defender. Most of those "Hole in Zone" passes end up as first downs.
And yet... as virtually every Football Outsiders writer has pointed out this week, just as Johnson is a terrible matchup for the Colts defense, so too is Peyton Manning a terrible matchup for the Chiefs defense. No matter how many points the Chiefs can score, the Colts can probably score more.
The Colts had the best offense in the league by any measure. We have DVOA for every team since 1997, and the only team with a higher offensive DVOA than the 2006 Colts was the 2004 Colts.
Kansas City's defense played reasonably well over the first few weeks, but they've been one of the worst in the league since the middle of the season. In Weeks 10-17, Kansas City ranks 30th in defensive DVOA.
The Chiefs primarily play a man coverage scheme, which is a problem because they only have one good cornerback, Patrick Surtain. Veteran Ty Law is a shadow of his former self, and nickel back Lenny Walls is subpar as well. The Chiefs rank fourth in DVOA against number one receivers, but 25th against number two receivers and last against slot receivers. No quarterback is better than Peyton Manning when it comes to patiently going through his reads and finding the open man among multiple targets.
The Chiefs also don't get much pass pressure. Ends Jared Allen and Tamba Hali are pretty good, but the defensive tackles are nothing special and the Chiefs ranked 28th in Adjusted Sack Rate (sacks per pass play, adjusted for situation and opponent). The Colts' offense ranked second in Adjusted Sack Rate, and first in fewest sacks allowed.
But wait, there's more! Do you know what part of the Kansas City defense really fell apart in the second half of the year? The run defense. In Weeks 1-9, the Kansas City run defense had a DVOA of -15.6%, 10th in the league. In Weeks 10-17, the Kansas City run defense had a DVOA of 15.2%, 30th in the league. Dominic Rhodes and Joseph Addai aren't Larry Johnson, but they'll be getting some yards out there too.
The Colts have a good kicker in Adam Vinatieri and a good punter in Hunter Smith, but the Indianapolis coverage teams are terrible, which is why the values above for net kickoffs and net punts are so awful. On the Kansas City side, Dante Hall really isn't anything special anymore and Lawrence Tynes has never been anything special. Dustin Colquitt is an excellent punter, but that won't matter if this game is anything like the last time these two teams met in the playoffs -- and neither team had to punt even once.
If the Texans can beat the Colts by running the ball and limiting Manning's possessions, the Chiefs can do it too. But "can" does not mean "will," and in general, a dominant passing game will outscore a dominant running game.
Full disclosure: I'm a Patriots fan. I've announced that before every Patriots playoff preview I've written during the four years of Football Outsiders. It didn't stop me from picking against the Patriots at least once in each of the past three postseasons, but I know some people still might feel there's a bias here. C'est la vie.
Nobody expected the Jets to be 7-9 this year, let alone 10-6 and in the playoffs, but here they are anyway. Their most emotional win of the year came when they beat the Patriots in Foxboro back in Week 10. It was one of just three Jets games with a DVOA rating above 20%, and one of just three Patriots games with a DVOA rating below -20%.
Overall, the Jets rank 19th in DVOA this year, behind six different AFC teams that didn't make the playoffs, including Buffalo and Miami. Some readers have asked how much the 41-0 blowout loss to Jacksonville in Week 5 skews this rating. The answer is "not much." Remove that game entirely, and the Jets would go from 15th on offense and 26th on defense to... 11th on offense and 26th on defense.
The offense, not the defense, is the unit that bears responsibility for the Jets' magical season. We were all so busy before the season worrying about Chad Pennington's shoulder that we all forgot what a good quarterback he is when actually healthy. Jerricho Cotchery had a breakout year as the second receiver opposite Laveranues Coles, Nick Mangold was a Rookie of the Year candidate at center, and late in the season, the Jets even found a running game -- or, at least, one-third of a running game.
The Jets have used three running backs this year. Leon Washington ranked ninth in DVOA among all running backs with at least 75 carries. Cedric Houston ranked 35th. Kevan Barlow ranked 48th. Barlow averaged less than three yards per carry. He looked so good in San Francisco back when he was paired with Garrison Hearst -- what the hell happened to this guy?
Barlow had one good game this year: against the Patriots in Week 10, when he gained 75 yards on 17 carries. It was his only game over 50 yards all year. It was a complete fluke, and the Patriots are praying that the Jets will use Barlow against them again this week. They probably won't.
Washington will have to face a defensive line that is fully healthy for the first time in weeks. Ty Warren was recently awarded AFC Defensive Player of the Month for December; he missed just one game all year, but it was the loss to the Jets. Nose tackle Vince Wilfork will play for the first time since he was injured in the loss to Miami Week 14. With the line healthy, the Patriots may show a 4-3 look more often than their usual 3-4 -- something they did with success in the first game between these two teams.
As for Pennington, he'll go to Cotchery early and often. Cornerback Asante Samuel has enjoyed a standout year, and the Patriots are an excellent defense against number one receivers, but they have a habit of giving up huge games to number two receivers (29th in DVOA). Chad Scott just isn't very good anymore, and Ellis Hobbs has struggled with injuries this season. In the two games against the Patriots this year, Cotchery combined for 191 yards and two touchdowns, catching 80 percent of intended passes. Coles had just 129 yards and one touchdown, catching 48 percent of intended passes.
The Jets were a top five offense in "close and late" situations this year, but the Patriots were a top five defense in those situations.
One more note: The Jets will do something funky with wide receiver/running back/option quarterback Brad Smith, and I have no idea what it will be. Bill Belichick probably has about 25 ideas about what it will be and there's a good possibility none of those ideas are actually correct.
This is where the Patriots win this game -- not because Tom Brady has magic beans that make him indestructible in the playoffs, but because they have a balanced attack with a slew of weapons, and the Jets are not as good as they look on defense.
Yes, the Jets ranked sixth in the NFL in points allowed, but that doesn't mean they had a good defense. They ranked 20th in yards allowed, which is a little closer to reality. In this week's DVOA commentary, I ran through some of the reasons why the Jets did not allow many points despite a subpar defense: easy schedule, good luck recovering fumbles, and strength in areas that have nothing to do with the defense (like points allowed on turnover returns). They also allowed fewer points because they faced fewer drives, with opponents running the ball and slowing down the clock.
And why did opponents run the ball? Because it was the Jets, not the Colts, who had the worst run defense in the league according to DVOA. Certainly that defense has improved, but not much -- the Jets rank 26th in run defense DVOA since their Week 9 bye. The Patriots will run the ball and have success with both Corey Dillon and Laurence Maroney.
Weakness against the run is a big reason why the Jets had the NFL's worst defensive DVOA on first downs. But they have the worst DVOA on first downs against both rushing (18.8%) and passing (19.0%). That number has improved since midseason against the run, but not the pass. The Jets were much better on third down, ranking ninth in DVOA.
The Patriots, however, were strong on both first down (19.6%, fifth) and third down (31.4%, third). Weirdly, they're one of the worst teams in the league on second down (-9.5%, 25th). I have no idea why.
The Patriots get tight end Ben Watson back from injury this week, so there will be a lot of two-tight end sets rotating Watson with Daniel Graham and rookie David Thomas, who had a big breakout game against Jacksonville two weeks ago. However, New York's strength on defense is covering tight ends (sixth in DVOA), which makes sense since the two best players on the defense are middle linebacker Jonathan Vilma and free safety Kerry Rhodes.
One final note: The Patriots' offense gets it done in the most important situations. They were the best team in the league running the ball in "power" situations (third/fourth down with 1-2 yards to go) and rank second in red zone DVOA. The Jets were below average in both areas.
If you enjoy kickoff returns, boy, this is the game for you. The Patriots (Laurence Maroney and Ellis Hobbs) ranked number one in kickoff return value by our metrics, and the Jets (primarily Justin Miller) ranked number two. Stephen Gostkowski was the best kickoff man in the league, but the Patriots don't have a good coverage team, so Miller could bring one back. Then again, Jets don't have a good kickoff man or a good coverage team to stop Maroney from bringing one back himself. When the Jets punt, strengths will even out, and when the Patriots punt, weaknesses will even out.
That leaves the whole Stephen Gostkowski rookie field goal thing. Does anybody think Gostkowski is somehow less "clutch" than Mike Nugent? This morning on Fox Sports New England I saw perhaps the dumbest statistic of the year: Gostkowski is 1-for-5 in the third quarter. You know, since third-quarter field goals are so unique and five is such a gigantic sample size.
The two worst teams to make the playoffs during the DVOA era were the 1998 Cardinals and the 2004 Rams. Both of those teams actually won their first playoff game on the road against a familiar division rival. The Jets are much better than those teams, so a New England victory is by no means guaranteed. But when you look at DVOA combined with home-field advantage, this is easily the biggest mismatch of the first round -- unless you really believe that Eric Mangini has some sort of psychic hold on Bill Belichick.
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, total along with rush and pass, and rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) Red zone DVOA is also listed. These numbers are all regular season only.
WEI DVOA is WEIGHTED DVOA, which is 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 is the same formula used in this week's FOXSports.com power rankings.
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.
Each team also gets a chart showing their performance this year, game-by-game, according to total DVOA. 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." Note that even though the chart appears in the section for when each team has the ball, it represents total performance, not just offense.
56 comments, Last at 08 Jan 2007, 7:48pm by B