Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Defense and Rest Time

Do defenses really wear out over the course of a game? Do defenses benefit from long drives that give them more time to rest on the sideline? Guest columnist Ben Baldwin investigates.

17 Dec 2005

Game Previews: DAL-WAS, ATL-CHI

by Aaron Schatz

There are nine teams currently in the NFC playoff picture, and seven of them are separated by a single game. Four of these NFC playoff contenders face each other in two games of critical importance on Sunday.


(Sunday, 4:15pm)

It was the second Monday night of the season, four minutes remained in the game, and the Dallas Cowboys held a commanding 13-0 lead over archrival Washington. As the clock ticked under 4:00, Redskins quarterback Mark Brunell hit Santana Moss with a 39-yard touchdown strike. Sixty-eight ticks of the clock later, Brunell hit Moss in stride on a 70-yard bomb and the Skins took home a shocking 14-13 victory.

Thirteen weeks later, Moss's 1167 receiving yards rank him second in the NFL, but Washington's other wideouts have been dismal: David Patten, now out for the year, averaged just 24 yards per game. His replacement, Taylor Jacobs, has averaged just 17 yards in four starts. Once opponents learned that Moss was Washington's only reliable wide receiver, his output dropped in half: Moss averaged 124 yards per game over the first seven weeks, 61 yards per game over the last seven.

The Redskins should have more success with running back Clinton Portis and tight end Chris Cooley. According to Football Outsiders' DVOA (Defense-adjusted Value Over Average) system -- which breaks down each play of the season and compares it to the NFL average based on situation and opponent, explained here -- Dallas ranks 22nd in the league against the run, and 26th defending passes to tight ends.

When Dallas gets the ball, the situation is reversed: Dallas has a strong passing game but has struggled to run the ball behind an injury-riddled offensive line. That's good news for the inconsistent Washington run defense, which has been terrible against the league's best running games -- 852 yards given up to the Giants, Chargers, Seahawks, and Broncos -- but has barely allowed three yards per carry in nine other games.

Dallas quarterback Drew Bledsoe should have time to throw the ball -- Washington has just 23 sacks, 27th in the NFL -- and find his favorite target, Terry Glenn, who had 157 yards and a touchdown when these teams first met. It helps that Redskins cornerbacks Walt Harris, Carlos Rogers, and Shawm Springs appear on the injury report, though all three will likely play.

These bitter rivals have had many strange finishes and close games during the 2005 season. Each team has had seven games decided by four points or less and ten games decided by seven points or less. The only safe prediction about this game is that it will be close.


(Sunday, 8:30pm)

When the Steelers beat the Bears in Pittsburgh last Sunday, 21-9, they were not just the first team in nine weeks to beat Chicago. They were also the first team in nine weeks to make Chicago's defense look mortal.

Suddenly, the Bears were having trouble tackling, allowing huge gains on screen plays. Once Pittsburgh took the lead, running back Jerome Bettis took the ball and bowled over the smaller, speedier Chicago defenders.

Atlanta's offense is built around the running game, which leads the league with 5.3 yards per carry and 24.5% DVOA. Like the Steelers, they feature a running back tandem with a smaller agile back (Warrick Dunn) and a big bruiser (T.J. Duckett). That sounds like a recipe for success after Pittsburgh beat Chicago with a heavy dose of the run, but before that game, Chicago had allowed just 3.6 yards per carry. (The Bears are third in run defense DVOA.) It is more likely that the Bears are the team that stopped the run in the first 12 games, not the team that allowed Pittsburgh to run in the 13th.

Atlanta's running game is not just about running backs, of course; it also features the greatest running quarterback of all time, Michael Vick. But the Bears have allowed given up just 45 rushing yards to quarterbacks all season, less than two yards per carry. Both figures are the lowest in the NFL. According to DVOA, the three most valuable rushing quarterbacks per carry, other than Vick, have been Aaron Brooks, Daunte Culpepper, and Anthony Wright. The Bears held these three players to a combined 20 yards on just six carries.

Vick's ability to run will also be limited by the bruised ribs that caused him to leave Monday's victory over New Orleans in the fourth quarter. And it is worth noting that the two defenses that held Vick to the fewest rushing yards this year were the Buccaneers and Jets. Like the Bears, both are built around strong linebackers.

When Vick passes, he prefers to pass to his tight ends, particularly the strong and agile Alge Crumpler. Compared to Atlanta, only two teams, Washington and Tennessee, have thrown a higher percentage of passes to their tight ends. But DVOA ranks Chicago fourth in the league defending tight ends. While the average tight end reception goes for 10.5 yards, the average tight end reception against Chicago goes for just 8.3 yards. The difference is particularly strong on first down, when less than half of tight end passes against Chicago are completed (compared to a league average of two-thirds).

These strong matchups on defense, of course, do nothing to solve Chicago's real problem -- the terrible play of rookie quarterback Kyle Orton, who averages just 5.2 yards per pass attempt. But Orton may not have to throw much at all. With veteran running back Thomas Jones having a career year, the Bears rank sixth in the league with 4.5 yards per carry. Atlanta allows 4.8 yards per carry, more than any other team except Houston and St. Louis.

Another point in Chicago's favor: the temperature Sunday night is forecast to be in the teens, and the Falcons are used to playing indoors. Since 2000, dome teams are 3-20 when the temperature at kickoff is 35 degrees or less, although one of those three wins came when Vick and the Falcons won at Green Bay in the playoffs three years ago.

The Bears need a win to keep the lead in the race for a first-round bye; if the Falcons lose, they effectively drop out of the playoff race. The Bears won't be able to stop Dunn and Vick completely. But the Falcons may not be able to stop Jones at all.

If Size Matters, the Bears' Defensive Line Didn't Get the Memo

by Michael David Smith

Chicago Bears coach Lovie Smith has built the best defense in football by relying on a simple philosophy: Speed is more important than size. Chicago's four starting defensive linemen -- 302-pound right tackle Ian Scott, 300-pound left tackle Tommie Harris, 260-pound defensive end Adewale Ogunleye, and 262-pound defensive end Alex Brown -- average 281 pounds, making them one of the smallest units in the league, but opposing offensive linemen haven't blocked them very often this season.

At least until Sunday, when the Pittsburgh Steelers' big offensive line (combined weight: 1,583 pounds) dominated the Bears' defense. Pittsburgh finished the day with 190 rushing yards, controlling the ball for more than 37 minutes. Because the Steelers matched up so well with the Bears, it would be tempting to conclude that, at the end of the day, small defensive lines simply can't stop the run. It would also be wrong.

Although some of the best run-stopping tackles in football -- including San Diego's 348-pound Jamal Williams, the Giants' 325-pound Fred Robbins, and Pittsburgh's 325-pound Casey Hampton -- are enormous, an examination of the size of the defensive linemen throughout the league shows that size and ability to stop the run do not go hand in hand.

The basic statistical measurement of how well a defense stops the run is average yards allowed per rushing attempt. (The NFL ranks rushing defenses not by average but by total rushing yards allowed -- a flawed statistic because it rewards good teams unfairly when they race out to early leads and force opponents to pass in order to catch up.) The defensive linemen on the top five teams in yards per rushing attempt weigh an average of 296 pounds, a mere two pounds heavier than the average weight of the defensive linemen on the five teams allowing the most yards per rushing attempt. The Bears rank fifth in the league, allowing 3.51 yards a carry, and their defensive line (small, at least, by NFL standards) weighs in at an average of 281 pounds.

Teams that use three defensive linemen and four linebackers generally have larger linemen than teams that use four defensive linemen and three linebackers, but even when 3-4 teams are separated from 4-3 teams, bigger lines are no better at stopping the run. The San Diego Chargers rank fourth in the league at 3.48 yards allowed per rushing attempt and have a huge 3-4 front of 303-pound end Luis Castillo, 348-pound Williams, and 309-pound end Igor Olshansky. But the Houston Texans are 30th in the league, giving up an average of 4.62 yards per run, even though they employ an even larger defensive line, with ends Gary Walker and Robaire Smith weighing 324 and 328 pounds, respectively, and tackle Seth Payne weighing 315.

The Pittsburgh offensive linemen who got the better of the Bears on Sunday weigh an average of 319 pounds, which puts them at the large end of the scale (NFL average is 310 pounds). The ease with which the Steelers ran against Chicago might hint that large offenses run well against small defenses, but the Steelers' performance two weeks earlier showed they can't always control the ball against small defenses. Against the Indianapolis Colts in Week 12, Steelers back Jerome Bettis had only nine yards on six carries, and the Steelers as a team combined for just 86 yards on the day. Indianapolis has a defensive line even smaller than Chicago's.

Despite the lack of evidence that bigger defensive linemen are more effective against opposing rushing attacks, NFL general managers looking to shore up their run defenses increasingly draft huge defensive tackles. The early buzz indicates that the top defensive tackle prospects for the 2006 draft are Oregon's 345-pound Haloti Ngata, Michigan's 335-pound Gabe Watson, and Texas's 330-pound Rodrique Wright.

Chicago's biggest problem on defense against Pittsburgh was bad luck with injuries (both starting safeties missed the game) and with fumbles (the Bears forced the Steelers to fumble four times, but Pittsburgh recovered all of them). Fans don't want to hear about bad luck, though, and if Chicago loses in the playoffs to another run-first team, some fans will say the Bears need to draft one of those players. Smith shouldn't give in to that pressure. He's built the best defense in the NFL for a reason: He knows that for defensive linemen, bigger isn't necessarily better.

These articles appeared in Friday's edition of the New York Sun.

Posted by: admin on 17 Dec 2005

19 comments, Last at 20 Dec 2005, 9:00pm by Rick


by admin :: Sat, 12/17/2005 - 3:00pm

One addendum: You might remember that a few weeks ago, I wrote that Marvin Lewis was wrong to deactivate his biggest defensive lineman, Shaun Smith, against Pittsburgh. When Lewis’s Bengals got a rematch against Pittsburgh, Smith started, played well, and Cincinnati did a much better job stopping the Steelers’ running attack. Then, when the Steelers ran all over the Bears, I really thought I had found something.

But once I started researching it I failed to find any evidence that the bigger lines are better at stopping the run, so this article went in the exact opposite direction that I expected.

I still think Lewis was right to use Smith in the second game against the Steelers, but I no longer think Smith’s size is his most important trait.

by admin :: Sat, 12/17/2005 - 3:00pm

First, a huge shout-out to Football Outsiders. My two fantasy QBs are Vick and Bledsoe, and I was having a devil of a time deciding between them. That info on Chicago being the best team against running QBs in the NFL decided it for me.

Re #1: I think in a lot of instances it helps a lot to have smaller defensive linemen in run support. Smaller linemen are faster, so if you have 4 fast DLs and the play is a run to the right, all 4 have a chance to make the play, because they have backside pursuit speed. I can tell you, for instance, that the University of Florida’s leading tackler this season was a DEFENSIVE END (which is the only time I have EVER heard of a DL leading his team in tackles without significant injuries to the LB corps). Anyway, the DE has great speed and catches a LOT of running plays from the backside, often for a loss. Also, intuitively, it seems like smaller DLs are the bane of the scrambling QB, since they’re better able to keep up with him and get him for a loss. That said, all other things being equal, if my defense is facing 4th and 1 and I’m given a choice between a 300 pound line and a 260 pound line, I’m going to pick the 300 pound one. It’s entirely anecdotal, but I just feel like bigger guys would be harder to move that one yard. Facing 1st and 10? I don’t know that it makes much difference.

Hey, MDS, I don’t know if this is the place to do it, but could I request an EPC on the Cleveland Browncos? Just curious because at least Warren and Brown are either free agents this offseason, or will have a prohibitively high cap cost (and will either renegotiate or get cut). I’m also a bit puzzled, because my observation is that Denver keeps getting TREMENDOUS pressure, but very few sacks- and that a lot of the pressure comes from the D-line, yet Denver still blitzes like it’s going out of style, including 3 straight 9-man blitzes against Philadelphia.

by JonL (not verified) :: Sat, 12/17/2005 - 4:46pm

The Redskins should have more success with running back Clinton Portis and tight end Chris Cooley.

That depends, of course, on Cooley not dropping the ball. Which is a lot.

by Jon (not verified) :: Sat, 12/17/2005 - 5:35pm

#3 You are very correct. Cooley makes some awesome catches then drops them, or misses balls right in his hands. Very frustrating...

by NFC Central Freak (not verified) :: Sat, 12/17/2005 - 5:46pm

The Atlanta win over Green Bay must be ignored. Green Bay's defense was decimated by injury and Ahman Greene left the game mid first quarter. The Packers had no chance against anyone much less a solid Falcon team.

Weather was not a critical factor in that game.

by top flight security (not verified) :: Sun, 12/18/2005 - 2:23am

the most interesting thing about the dallas-d.c. games was that the cowboys kept beating them no matter how bad dallas was. but now that i've seen a recent redskins team beat the cowboys i can watch this game with much more relaxed. I really think too many redskins fans put too much stock in this rivalry. Some people say they'd be happy to be 2-14 as long as the skins beat dallas twice, but i say i'd rather see washington go 14-2 and win the superbowl with both losses to dallas than that. All that being said, i've watched enough of these games too know how it's going to turn out. Whoever has a better game between moss and glenn, that team will win and if they both have bad games, then whoever forces a fumble from a blindside hit on bledsoe or brunell in the second half will win. That blindside hit will definitely happen.

by Liam (not verified) :: Sun, 12/18/2005 - 8:05am

Last season, much of Atlanta's success on defense was built around the ability of its small defensive line (averaging 281 pounds) to stop the run.

According to defensive line stats ATL's real strength was between the tackes, where they ranked 3rd. 16th and 8th (LT, mid-guard & RT). The dropoff this year is astronomical - they rank 32nd, 32nd and 24th.

The philosophy of the line was to get penetration - something that Rod Colemen and Ed Jasper were particularly good at.

However, there was one game last year where that line stunk beyond belief: Kansas City, despite not breaking a run greater than 16 yards, rushed for 271 yards and 8 TD's on 49 carries.

That particular game had something in common with this season in regards to the defensive line. Against the Chiefs, ATL were missing two players that epitomsied the philosiphy (Coleman and Kerney) and their replacements clearly weren't up to the task.

This has been the case this year, with Brady Smith being out of action for most of the year and Ed Jasper having departed.

It suggests to me that a small, disruptive line can have success against the run, provided that everyone along the line is quick enough to get the penetration. If one or two players can't do this, they are just creating lanes for the RB's to run through.

This is a very small sample size and it's hard to get any more data without a fair bit of research (not just finding small lines, but ones with injuries - and does the same drop-off occur with larger lines???). Can anyone back this up with more examples?

Also, it's worth noting that this year, ATL's defense are tackling like girls.

by MitchW (not verified) :: Sun, 12/18/2005 - 10:17am

Quick question - could the condition of the Pittsburgh field (it was more mud than grass) have contributed to their effectiveness in running the ball against the Bears?

Can't help but think that for a quick defensive line like the Bear's, playing on a sloppy field doesn't improve their chances.

by Dr Ryan (not verified) :: Sun, 12/18/2005 - 12:19pm

Does the NFL / teams have any rules regarding listing playing weights ?

Ted Washington is listed at 365 but I recon he's closer to 400 lbs

This could have a big impact on your hypothesis

by Sid (not verified) :: Sun, 12/18/2005 - 5:15pm

It helps that Redskins cornerbacks Walt Harris, Carlos Rogers, and Shawm [sic] Springs appear on the injury report, though all three will likely play.

Rogers is inactive, as is LaVar Arrington.

In the 46 year history of this series, Washington has swept only 3 times.

by Sid (not verified) :: Sun, 12/18/2005 - 5:48pm

The Chicago defense is basically great at everything. The question is whether they can fix whatever the problems they had against Pitt were. There is a limit to what they can do, because both safeties are still missing.

by Sid (not verified) :: Sun, 12/18/2005 - 8:00pm

RE: 10

And the 2005 season marks the 4th time.

by Patrick Bateman (not verified) :: Sun, 12/18/2005 - 10:18pm

So much for that safe bet in the WASH/DAL game.

by Sid (not verified) :: Mon, 12/19/2005 - 12:22am

This article didn't do much to dissuade me of the notion that bigger DLs tend to be better against the run. Bigger DLs are harder to move out of the way. The best NTs are all massive.
BTW, comparing the top 5 to the bottom 5 in weight isn't a very extensive comparison. The LBs make a difference as well.
I'd compare the weights of all the front 7s in the NFL, and see how they do against the run, on average.

by jake (not verified) :: Mon, 12/19/2005 - 5:42am

"The only safe prediction about this game is that it will be close."

i guess you meant "close to a blowout" didn't you?

by Jeesh! (not verified) :: Mon, 12/19/2005 - 9:46am

Hey - somebody better tell Parcells that Santana Moss just might go deep ! D'oh!

The Cowboys looked like one of the worst coached teams, and parcells has got to be the most over-rated coach ever. If excessive, dumb penalties aren't a mark of a poorly coached team, I don't know what is.

by mactbone0 (not verified) :: Mon, 12/19/2005 - 11:20am

See the thing is, the Bears have had both kinds of defenses. In 2001 with Dick Jauron in charge, the Bears had a front four of Philip Daniels (Sack Specialist!), Keith Traylor, Ted Washington, and Bryan Robinson (now a DT). They were massive and let the linebackers - Warrick Holdman, Rosy Colvin and Urlacher roam free. Now the Bears rely on getting upfield pressure to disrupt both running and passing. In the 2001 scheme nobody on D-Line could get any pressure, Colvin had double-digit sacks and Urlacher had a few as well.

I think the biggest difference is that small lines can be overpowered in Power situations. On the other hand, big lines generally don't get as much pressure. I know Big Ted and Traylor could get pressure sometimes but not in a consistent way.

I'd also like to think that massive athletic guys are much harder to find and keep healthy.

by Jeff J. (not verified) :: Mon, 12/19/2005 - 5:06pm

Can't add much more than what #15 & #16 said. What a travesty for Dallas and their fans!

I loved every bloody mintue of it. When it was over, I restarted my recording and enjoyed it all over again.

by Rick (not verified) :: Tue, 12/20/2005 - 9:00pm


You might want to rethink your opinion of Chris Cooley.
The man has been great all year.
Maybe last week was just a defence that was taking care of TE recievers.
Although I admit.... those drops were pretty bad.