Bill Connelly takes a look at what we can learn from defensive box score stats and general rates of havoc.
14 Jan 2011
by Aaron Schatz
The word of the week in the AFC is "rubber." Rubber, as in "rubber match," two sets of division rivals who split their regular season games this year. Rubber, as in "I'm rubber and you're glue," which is what Tom Brady would likely say to Antonio Cromartie and Terrell Suggs were he 10 years old. And rubber as in "when the rubber meets the road," because the losers of this week's games go home.
For those who may be unfamiliar with the Football Outsiders stats, they are explained at the bottom of the page. Scroll down or click this link. Please remember that all stats represent regular season only, except for weighted DVOA and anything else specifically noted.
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.
There are some interesting trends coming into this game, primarily the recent improvement of both the Pittsburgh offense and the Baltimore defense. The Pittsburgh offense ranked 20th in DVOA during the first four Big Ben-free weeks of the season, then 14th during Big Ben's first four games back. But since Week 10, the Steelers have the second-best offense in the league, behind only New England. The Baltimore defense has likewise improved, from 14th in the first half of the year to second (behind, of course, the Steelers) since Week 10. Over the second half of the season, the Ravens have the best run defense in the NFL. Pittsburgh is third in that time-period, and first if we look at the entire season.
That defensive improvement by Baltimore is one of the reasons why previewing this game is rather difficult. Finding weaknesses is tough when two teams spend the season playing "anything you can do, I can do better" with pretty much every defensive statistic.
Ray Rice is almost assuredly not going to be the biggest factor for Baltimore in this game. The Steelers have allowed a remarkable 2.8 yards per carry to running backs on first down, and a league-low 0.17 Open Field Yards per carry. Meanwhile, the Ravens' running game -- allegedly a team strength -- has dropped from 13th in DVOA during the first half of the year to 21st in the second half. The Ravens rank only 29th in third-and-short situations and 23rd in the red zone. And because of these issues with the running game, the Ravens offense often struggled when it had a big lead (26th in DVOA when leading by more than eight points).
So how will the Ravens do if forced to go to the air more often than they would prefer? Watch for the Ravens throwing to whichever receiver is lined up on the right side. (The Ravens generally move their receivers around, so we can't say who that receiver will be.) Our charting numbers show a big difference between the two Pittsburgh cornerbacks.
|Pittsburgh Cornerback Charting Stats
(Games Charted as of 1/14/11)
|Player|| Targets per
Of course, Joe Flacco will have to stay upright in order to find those receivers on the right side. The Pittsburgh defense was third in Adjusted Sack Rate, while the Ravens offense was 25th and got significantly worse as the season progressed. Baltimore's Adjusted Sack Rate on offense went from 5.3 percent in the first half of the season to 10.7 percent since Week 10. No matter who the Ravens throw to, they aren't likely to get big plays with lots of yards after the catch. The Steelers led the league with just 3.9 average YAC. (The Ravens were sixth, allowing 4.6 YAC.)
However, trends that looked like weaknesses for Pittsburgh before the Week 13 Steelers-Ravens game no longer look like weaknesses after the final five weeks of the season. The Steelers defense dipped a bit at midseason after starting the year strong, but they ended the year strong as well. Back in Week 13, the Steelers were weaker against slot receivers and tight ends than against starting receivers; both of those numbers have improved. However, Baltimore could have success with running back screen passes; the Ravens gained 7.9 yards per screen pass while the Steelers allowed 8.3 yards per screen pass (27th).
First of all, you may notice the impressive week-to-week graph over to the right. Overall, the Steelers were not as strong as the Patriots this season, but they were more consistent. The Patriots had two bad games, losses to Cleveland and the Jets. But the Steelers didn't have any stinkers. Even their losses had DVOA ratings around or even above zero percent. Their worst game of the year according to DVOA was -5.4% against this week's opponent, Baltimore, back in Week 4.
Of course, Ben Roethlisberger was not active for that game, and that highlights the biggest difference between Pittsburgh and Baltimore in 2010. While both defenses are strong, the Pittsburgh offense was simply better than the Baltimore offense this year. That's been true about the passing game since Roethlisberger returned in Week 6, and it's been true about the running game for the last two months.
Pittsburgh will want to use Rashard Mendenhall to attack the Baltimore defense on the right side, where the Ravens ranked just 28th in Adjusted Line Yards against runs behind right tackle and 31st in ALY against runs around right end. Still, "big gains" are relative against these defenses. Pittsburgh didn't have a run over 12 yards in either game against Baltimore, and had only ten runs over four yards. Eight of those ten were to the right side (seven by Mendenhall, one by Ike Redman).
When the Steelers go to the air, we'll all be watching for big plays by Mike Wallace. Wallace was the best receiver in the league this year according to both DYAR and DVOA, so obviously the Ravens are worried about him making big plays deep. However, the Ravens haven't been particularly susceptible to these plays since Ed Reed came back and they demoted Fabian Washington from the starting lineup. Lardarius Webb did a pretty good job covering Wallace in the first two Baltimore-Pittsburgh games, including two big passes defensed on deep throws to Wallace in the second half of the Week 4 game. Hines Ward declined a bit at the age of 34, but still had a good year and was particularly strong with 44.1% DVOA on third and fourth downs. The Ravens were very strong against slot receivers and tight ends, second in DVOA covering both positions.
The Ravens' pass rush isn't too strong -- just 24th in the league in Adjusted Sack Rate -- but it's hard not to get sacks against Ben Roethlisberger, especially when Jonathan Scott is "protecting" his left side. What's important is when those sacks come. Sacks on Roethlisberger do not tend to put Pittsburgh into tough second-and-long situations. Instead, Roethlisberger takes most of his sacks when he's desperate to make something happen that will extend the drive. The Steelers had the highest offensive Adjusted Sack Rate on third down -- and remember, ASR adjusts for the fact that sacks are more common on third down in general. The Ravens' sack rate also goes up with each down, so expect a couple of drives to end with Roethlisberger on the ground.
|Adjusted Sack Rate by Down:
Pittsburgh Offense vs. Baltimore Defense
|PIT OFF||BAL DEF|
The Ravens don't like to send as much pressure as you might expect, and they might want to use more blitzes against Roethlisberger. Baltimore sent five- and six-man pass rushes this year at rates that matched the NFL averages. However, they were awesome when sending a big blitz of six or more, with just 4.2 yards per pass allowed. Meanwhile, Ben Roethlisberger did get a little bit weaker with each additional pass rusher. It's not a massive difference, but he did go from 7.6 yards per pass (3-4 pass rushers) to 6.9 yards per pass (five pass rushers) and then 6.3 yards per pass (six or more pass rushers). Against Roethlisberger, pass pressure means bringing more guys, not just different guys. Roethlisberger was no worse against "zone blitzes" than he was against regular four-man and five-man rushes, and Baltimore was the team that sent the highest rate of zone blitzes this season, 13.2 percent. (Note that we changed the definition of a zone blitz in game charting this year to get a truer picture of complex 3-4 defenses. For a 3-4 defense, the definition now includes plays where both outside linebackers drop out but a defensive back or inside linebacker rushes the passer, not just plays where a defensive lineman drops out.)
The Ravens had the league's best defense against a play the Steelers rarely use, the running back screen, but also one of the league's top defenses against a play the Steelers use all the time, the wide receiver screen.
In my Numbers Crunching article on ESPN, and when guesting on a couple of podcasts this week, I mentioned that special teams would likely lead to longer field position for both teams, because both Pittsburgh and Baltimore rank very high in punt value but very low in punt return value. It turns out this is not quite true. I forgot about the injury to Pittsburgh's Dan "Robopunter" Sepulveda. Replacement Jon Kapinos was the worst punter in the league with Green Bay in 2009; he's been better in his short time with Pittsburgh this year, but still has just 32.3 net yards per attempt compared to Sepulveda's 39.1 net yards per attempt.
Baltimore has in advantage in special teams overall thanks to the spectacular (and, to be honest, fluky) season from kicker Billy Cundiff. Those touchbacks can mean a lot, although the advantage in field goals that appears in the season-long stats may no longer exist. Jeff Reed was worth -8.3 points compared to average when he kicked for Pittsburgh; replacement Shaun Suisham has been very strong, with 7.7 points of value above average on field goals (14-of-15).
One of these teams will win by three points because these games seem to always be decided by three points. Given their advantages over the Baltimore offense in important splits like third-and-short and the red zone, there's a better-than-even chance that the winner will be Pittsburgh.
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.
Feet feet feet. Feet-y feet feet. That's why they call it "football."
It's no secret that the Patriots have the advantage in this game. In fact, they may have the advantage on both offense and defense, as long as they continue to play like they did in December. After the 45-3 Monday Night Massacre, I wrote that the Patriots' defense had not improved as much as conventional wisdom seemed to believe. But it turns out that week, and that game against the Jets, was when it actually did start to improve. Through Week 12, the Pats ranked 27th in the league in defensive DVOA. In Weeks 13-17, the Pats rank second on defense, including first against the pass.
|New England Defense, 2010|
|Weeks 1-12||Weeks 13-17|
If we want to look for weaknesses in the pass defense, we might as well assume that the improvement of the last five weeks is (mostly) for real. It's less important to break down their pass defense from the first 12 weeks of the season; if the Patriots come out and play defense like that, their weaknesses will include pretty much any pass of any kind, provided that Mark Sanchez can keep the ball less than a foot above his receivers' heads.
So, where are the places where the defense did not improve over the last five weeks, or improved less? The Patriots have been particularly weak against running backs in the passing game this year (30th in DVOA),, and that DVOA actually got slightly worse over the past five weeks. It's not a specific strength of the Jets, but it is something LaDainian Tomlinson was known for once upon a time. The Patriots also didn't get any better against third receivers -- they were average before this five-week run, and average during that run, even though defense against the opposition's top two receivers improved dramatically. Then again, this is mostly a product of the final four games, because it wasn't in evidence during the 45-3 game. Santonio Holmes was the one Jets receiver who did manage a good game that day (seven catches, 72 yards), and Jerricho Cotchery, the Jets' number-three with a history of burning New England, had just two catches for nine yards as part of his Loser League MVP season.
Looking at direction doesn't help much. The Patriots' defense against passes to the deep right portion of the field declined over the last five weeks, but that's almost entirely due to just two passes: a meaningless Ryan Fitzpatrick-to-Naaman Roosevelt bomb with Buffalo down 31, and the awesome Matt Flynn-to-James Jones 60-yarder from Green Bay.
One positive for the Jets' passing game is that the Patriots do not blitz much. Mark Sanchez struggled against heavy pressure this season, but that likely won't be a big issue in this game.
Of course, the Jets will come out trying to run the ball as often as they throw it, at least until they fall behind. The Jets ran the ball on 47.7 percent of first-half plays this season -- in other words, before the score usually was an issue. That percentage was sixth in the league. The Jets should be able to get consistent running gains. Their hogs up front are a lot stronger than the guys who lead the Patriots' 3-4 defense, especially with the Patriots dealing with some injuries on the line (although those injuries are somewhat mitigated by the return of their best run-stopping linebacker, Brandon Spikes, from suspension). The Jets are third in Adjusted Line Yards, while the Pats defense is 21st. The Jets have higher ALY than the Patriots in all five directions, although you have to wonder a little bit about the and a huge advantage when it comes to short-yardage runs. The Jets converted 76 percent of the time, second in the league behind Miami, while the Patriots allowed conversions 71 percent of the time, 30th in the league. Rex Ryan needs to be gutsy on fourth down in this game.
He also needs to use Shonn Greene more than he uses Tomlinson. If you look at the Jets' rushing numbers before and after their Week 7 bye, you can see just how much Tomlinson has fallen off after a hot start. Greene is averaging fewer yards per carry as well, but with a better success rate because he tends to get the ball in those short-yardage situations.
|New York Jets Running Backs, Before and After Week 7 Bye|
|Tomlinson Weeks 1-6||92||490||5.3||24.3%||52%|
|Tomlinson Weeks 8-17||128||418||3.3||-12.6%||38%|
|Greene Weeks 1-6||71||323||4.5||5.4%||51%|
|Greene Weeks 8-17||114||444||3.9||9.4%||58%|
The Patriots offense wasn't just great when you look at the total performance according to DVOA. This offense was great in pretty much every way. The Patriots were second in rushing DVOA, but otherwise had the best offense in nearly every statistical split. They had the best offense when losing and the best offense when winning. They had the best offense in the red zone, and in any other 20-yard split on the field except when backed up behind their own 20. They had the best offense at home, and the best offense on the road. They had the best offense from shotgun, and the best offense with the quarterback under center. They had the best offense before halftime, and the best offense after halftime. They had the best offense in a boat, and the best offense with a goat. They had the best offense on second down and third down, although they were -- gasp! -- second behind the Giants on first down.
So you can see why stopping them can be somewhat difficult. The Jets defense is pretty good -- fourth on first down, and first in the league on second down -- but it completely broke down on third down this year. The Jets were dead last on third-down defense after being number one in the NFL a year ago. The "third-down rebound effect" makes that a great sign that the Jets will continue playing well in 2011, but it's not a great sign that the Jets will be keeping the New England offense off the field this Sunday.
Making it even more likely the Patriots will score points is the way the Jets defense gets worse as you get closer to their goal line. The Jets rank first in DVOA when the opponent is behind its own 40, 10th when between the 40s, 24th when the opponent gets the ball between the Jets' 20 and 40, and 19th when the opponent is in the red zone.
And the new-look Patriots don't really fit the way the Jets like to play defense. Who is the number-one receiver for Darrelle Revis to shut down? The Patriots' two best downfield receivers are guys that Revis and Antonio Cromartie are unlikely to be covering, tight ends Aaron Hernandez and Rob Gronkowski. The Patriots have flipped the usual script by using their tight ends to stretch the field for the wide receivers running underneath, rather than the other way around. Here's a look at quarterbacks who threw fewer than 75 percent of their deep passes (16+ yards through the air) to wide receivers. The leaders are Brady and the guys from Cleveland (mostly due to Evan Moore, who was a wideout in college) and San Francisco (because of Vernon Davis).
|2010 Quarterbacks with < 75% of Deep Passes
Thrown to Wide Receivers
|* 60.4 percent after Randy Moss traded.|
The other problem for the Jets is the shotgun spread. As I've written numerous times this season, the Jets are far better against offenses that use standard formations. The final numbers have the Jets giving up 6.5 yards per play to teams in shotgun (29th in the NFL) and 4.0 yards per play to teams with quarterbacks under center (best in the NFL). New England's use of shotgun has gone down a bit compared to recent years (only 40 percent of plays) but it still worked wonders against the Jets in the 45-3 game. The Patriots went shotgun in only 20 of 48 plays (42 percent) against the Jets in Week 13, but a lot of that was the effect of running out the clock. If we only look through the touchdown on the first play of the fourth quarter which made the game 38-3, the Patriots used shotgun on 19 of 37 plays (51 percent). They averaged 10.5 yards on those 19 shotgun plays, compared to 6.7 yards on 19 non-shotgun plays (18 standard plays plus a 36-yard Defensive Pass Interference) during the same time period.
Watch for pass pressure on Brady on second downs. The Jets' defense has a strange split with 13.1 percent Adjusted Sack Rate on second down -- the only team above 10 percent -- but 6.9 percent ASR on first down (14th) and 1.0 percent ASR on third downs (dead last).
The Patriots running game is strong as well; other than the absence of Randy Moss, the biggest difference between the 2007 Patriots and the 2010 Patriots is that the current team runs the ball more often. Look for the Pats to run the ball around left end, where they rank fifth in ALY while the Jets defense is 29th.
The Jets are better on special teams, although not by much. They have a big advantage on kickoff returns; the Jets had the best kick returns in the league, while the Patriots have been poor on kickoffs since the injury to Stephen Gostkowski. Shayne Graham averages 6.5 yards less per kick than Gostkowski did. Of course, the Patriots had pretty good kick returns as well, and better punt returns than the Jets. Both Steve Weatherford and Zoltan Mesko are above-average punters.
This is pretty simple. The Patriots are one of the two or three best teams of the last decade. They dismantled the Jets less than two months ago. Before that game, their offense was far superior to that of the Jets; since that game, they've been better than New York on both offense and defense. They've had single-game DVOA above 40 percent in eight straight games. During that same period, only one team had as many as four games with DVOA above 40 percent (Tampa Bay, surprisingly). The Jets were one of the top ten teams in the NFL this year, yet they have had only four games above 40 percent all season.
There are ways for the Jets to win -- a couple big special teams plays, or perhaps the Patriots' defensive improvements turn out to be a mirage. Maybe the Patriots wake up on the same wrong side of the bed that they woke up on for their Week 9 game with Cleveland. But most likely, they will move on to host Pittsburgh or Baltimore in the AFC Championship.
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.
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. Those numbers are explained here.
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.) We also list red zone DVOA and WEIGHTED DVOA (WEI 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).
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. I know some of you don't like the trendline, but you are far outnumbered by people who do like it, so tough nouggies.
65 comments, Last at 19 Jan 2011, 9:43pm by Aquanarc