No defense generated more pressure last year than Connor Barwin and the Eagles, but did that pressure do them any good?
13 Jan 2012
by Danny Tuccitto (NO-SF) and Aaron Schatz (NYG-GB)
Offense, baby. Points! Scoring! That's what the best NFC teams were about this year. Well, most of them anyway. New Orleans, Green Bay, and the New York Giants all combined strong offense with porous defense. San Francisco, on the other hand, had a running-and-defense model and will try to prove that defense still wins championships. That is, if defense ever won more championships than offense in the first place, which it probably didn't.
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. Game charting data is still incomplete, but represents most of Weeks 1-16.
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.
When this divisional round game was set after New Orleans defeated Detroit last Saturday, the first thought that crossed my mind was, "What does DVOA history say about playoff matchups between an elite offense and an elite defense?" I didn't think this happened all that often, but when I searched our database, I found a whopping 85 playoff games since 1992 with a top-eight offense going against a top-eight defense. In those games, the elite offense went 45-40. Of course, New Orleans will be on the road this Saturday, so what about that subset of games? The elite offense went 15-21 on the road. Now, better teams according to DVOA usually end up having a divisional round home game, so that record might just be a statistical artifact of home field advantage. So what about the subset of games where the elite offense is on the road but is actually the better team overall? In those 15 games, the elite offense went 8-7. All of this suggests that, in playoff matchups between immovable objects and irresistible forces, neither side has a historical advantage.
Obviously, though, this game won't be getting played in a historical vacuum; the details of why New Orleans and San Francisco actually win (and lose) games in 2011 matter more. Looking at the 49ers single-game DVOAs, the pattern is clear, albeit somewhat counterintuitive. Their run defense is so consistently great that the number of points they allow in a given week ebbs and flows based on the performance of their pass defense. Indeed, in San Francisco's worst five games this season according to pass defense DVOA, they suffered all three of their losses, eked out a 3-point win at Seattle, and won at Philadelphia thanks to fumble-recovery luck.
For New Orleans, the counterintuitive straw that seems to stir the drink is their No. 1 run offense, which hasn't gotten much publicity on account of Drew Brees' record-setting season. In the Saints' five games this season with a negative run offense DVOA -- all on the road, by the way -- they suffered all three of their losses, eked out a 3-point win at Carolina, and won at Atlanta when the Falcons failed on fourth-and-1 in overtime.
Putting these two (eerily symmetric) trends together, it's apparent that this game will probably be decided by what happens when New Orleans' offense is on the field. Speaking of which…
From a personnel standpoint, it's no secret that the Saints rely heavily on multiple-receiver formations. According to the game charting data we've amassed so far, they have three or more receivers on the field about 50 percent of the time. Based on the 49ers' defensive tendencies, that means we'll be seeing a lot of nickel and dime: In their three games against pass-oriented teams with good quarterbacks (vs. Dallas, vs. Pittsburgh, at Detroit), the 49ers used five or more defensive backs on over 90 percent of plays with three or more receivers in the formation.
What's interesting about the 49ers' nickel package, and different from most teams, is that they don't sub out one of their starting inside linebackers for a better coverage linebacker. That's because Navorro Bowman and Patrick Willis are both athletic enough to cover running backs, and big enough to cover tight ends. To wit, on 53 charted passes to backs, Bowman and Willis (and Larry Grant as Willis' injury replacement) had primary coverage on 28 of them, with a success rate of 64.3 percent. In 102 charted passes to tight ends, they had primary coverage 52 times, with a success rate of 61.5 percent.
This begs the question, "Then why is San Francisco's pass defense ranked only 20th against running backs?" Well, the answer is basically that teams have found considerable success throwing to running backs when they split a running back or a tight end out wide. In essence, they counteract Willis' and Bowman's coverage prowess by forcing the Niners to cover running backs with someone else. Jahvid Best, for instance, had 51 yards on two catches while lined up at receiver because San Francisco ended up covering him with Dashon Goldson and Aldon Smith.
Of course, perhaps no team splits their running backs and tight ends out wide more than New Orleans. Between the Saints two offensive catalysts, Jimmy Graham and Darren Sproles, our charting data to date shows that they've been lined up at wide receiver on 41.9 percent of pass plays. In fact, both have lined up at wide receiver on the same pass play almost five percent of the time. Therefore, insomuch as the outcome of this game depends on San Francisco's pass defense, the Saints have exactly the type of offense to exploit its vulnerabilities.
Many analysts have mentioned how a defense's best bet to slow down the Saints passing game is by being able to get pressure on Brees with only a four-man rush. Those analysts are wrong: Our charting data indicates that New Orleans is No. 1 in success rate against both three- and four-man pass rushes. For San Francisco, this also doesn't bode well. Specifically, Vic Fangio's defense ranks 29th in the league with respect to their frequency of rushing five or more. Furthermore, one of Fangio's favorite disguised rushes -- a stunt with Aldon Smith looping around to the inside of Justin Smith -- will have to somehow flummox left tackle Jermon Bushrod, who's given up only one blown block sack all year, and left guard Carl Nicks, who we just named to the Football Outsiders' All-Pro team.
Based on the above, it's increasingly clear that, in order to slow down the Saints offense, the 49ers are going to have to rely more on their run defense than their pass defense. Luckily for them, the matchups are much more in their favor despite the fact that New Orleans has the No. 1 run offense DVOA, and the No. 1 offensive ALY. According to our directional running stats, the Saints mostly run to the outside left and (surprisingly) up the middle. Defending these two directions, the 49ers rank 11th and sixth, respectively. Furthermore, although the Saints are No. 1 in second-level yards, the 49ers are No. 1 in limiting them.
There are three questions San Francisco's offense is going to have to answer on Saturday. First, can the offensive line -- ranked 25th in Adjusted Sack Rate -- protect Alex Smith? Second, can they exploit the Saints' 26th-ranked pass defense? Finally, and (surprisingly) importantly, how much is tight end Delanie Walker's absence going to adversely affect their deceptive schemes? Let's take these in reverse order.
A little-known fact about the 49ers offense, which runs entirely contrary to the conventional wisdom, is that they line up with either one running back or an empty backfield about 60 percent of the time, which is essentially as often as -- wait for it -- New Orleans does. The difference with San Francisco is that their empty and one-back sets include an extra tight end, whereas New Orleans' include an extra wide receiver.
Before he got his jaw broken by Seattle linebacker LeRoy Hill in Week 16, Walker was that second tight end, and his skill set allowed San Francisco to line him up all over the place in their formations: wide receiver, flexed tight end, H-back, lead-blocking fullback, you name it. This allowed the 49ers' offensive coaching staff to exploit mismatches that resulted from having to account for two complete tight ends. Indeed, given how much the 49ers offense relies on deception and formation diversity (they've used 22 different combinations of running backs, wide receivers, tight ends, and linemen according to our charting data), one could argue Walker was the most important piece in their Jenga tower. Therefore, Walker being ruled inactive this week will likely limit what the 49ers are able to do on offense. If it's any indication, they used the far-less-talented Justin Peelle against the Rams in Week 17, and only lined up with two or more tight ends 33 percent of the time as opposed to the 45 percent they had been previously.
Nevertheless, the Saints' pass defense is so bad that San Francisco can still be successful even if it's with a more conventional scheme. One thing we're likely to see based on the statistical matchups is a heavy dose of Kyle Williams and Ted Ginn. That's because the Saints rank 27th in pass defense against slot receivers, they rank 30th in passes to the short-left portion of the field, and right cornerback Tracy Porter is by far the Saints' worst cornerback (30 percent pass Success Rate) according to our numbers. Before his concussion in Week 16, Williams in particular had been coming on as a viable No. 2 receiving option from the slot, ending the season ranked eighth in DVOA among wide receivers with less than 50 targets. Therefore, he'll likely be the main beneficiary.
Another way the 49ers are probably going to attack the Saints pass defense is by throwing on first down, which has the added bonus of running entirely contrary to their perception as a run-heavy offense. The 49ers were passing a lot more on first downs toward the end of the season, and on first-down passes San Francisco ranked fourth in DVOA, whereas New Orleans ranked 30th on defense.
But really, all of the above will be rendered moot if San Francisco's offensive line can't give Smith time to throw, and this matchup doesn't favor them at all. Gregg Williams is a well-known devotee of blitzing, and his defense has had all the hallmarks once again this season: They rush six or more a league-leading 26 percent of the time. Meanwhile, the 49ers pass offense has been mediocre at best against six-or-more rushers, ranking 23rd in success rate. Furthermore, the 49ers offensive line has given up a sack or pressure on 37 percent of plays in our charting database, with an average of 4.7 rushers per sack/pressure. This plays right into the Saints hands given that their 23 percent sack/pressure rate has come with an average of 5.1 rushers. In other words, New Orleans brings the heat, and San Francisco's offense line generally gets out of the kitchen.
The one type of blitz in particular that the Niners offensive line has had problems with is one in which the defense sends two blitzers at the interior line, and forces left guard Mike Iupati, center Jonathan Goodwin, and right guard Adam Snyder to figure out who to pick up. Oftentimes, it's resulted in a comedy of mental errors, especially in their losses against Baltimore and Arizona. No doubt the Saints have seen this on tape, so look for many of those signature big blitzes to attack this weakness.
There's one thing San Francisco can do to combat this, and it's actually a staple of their offense. Earlier this season, Chris Brown of Smart Football (via Grantland) broke down how Jim Harbaugh incorporated "hot routes" into every pass play such that they're a natural part of Alex Smith's progression rather than a pre-snap read relying on a mind meld between receiver and quarterback.
So where are the hot routes likely to be this week? Based on the stats, I'd put my money on passes to Kendall Hunter in the left flat and passes to Vernon Davis in the short middle. I already mentioned how bad the Saints are on short-left passes, but it should also be noted that Hunter finished the season ranked 13th in receiving DVOA among running backs. With respect to Davis, the Saints are also ranked dead last in passes to the short middle thanks to abysmal coverage from their linebackers (i.e., 37.5 percent Success Rate among starters).
In games as close as this one is likely to be, field position oftent makes the difference. If that's the case on Saturday, it likely means a San Francisco win. This season, the 49ers hit the trifecta of being No. 1 in average starting field position on offense, No. 1 in average starting field position on defense, and -- because of math -- No. 1 in net average starting field position. It's true that much of their offensive field position was due to turnovers created by their defense. However, it was also due in no small part to Ginn's kick returns. And after missing Week 17 with a sprained ankle, Ginn will resume those duties this week. With New Orleans ranking 26th in net expected points from kickoffs, he could end up being a big factor.
Much has been made this week about how the 49ers need to slow the game down and run a plodding offense lest they get in a track meet with the high-octane Saints. What the statistical matchups seem to suggest, however, is that the Saints offense is probably going to get theirs, moreso with their No. 3 pass offense than their No. 1 run offense. Therefore, it will be incumbent on the 49ers pass offense to respond in kind. Whether or not they do depends on how well the offensive line is able to pick up New Orleans' big blitzes.
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.
Every year, the Giants go into a second-half funk. This year was no exception. Actually, 2007 wasn't an exception either, even though the Giants won the Super Bowl that season. What makes 2007 and 2011 different? In both years, the Giants reversed their second-half collapse near the end of the regular season and carried that strong play into the postseason. In 2007, the reversal started in Week 17. This year, it started with the Week 16 win over the Jets. Check out the Giants' DVOA in Weeks 10-15 compared to both before and after. Ironically, their special teams got better during the weeks their offense and defense were much worse.
|New York Giants DVOA by Week, 2011|
As you can see on the week-to-week graph below, the close Week 13 game against Green Bay was the strongest game the Giants played during that midseason period of struggle.
The Packers offense also went through a midseason period of decline, although it was only a decline from otherworldly to excellent. Before their Week 8 bye, the Packers had 40.8% offensive DVOA. From Week 9 through Week 15, they had 27.9% offensive DVOA -- still excellent, but below New Orleans and New England. Then Aaron Rodgers blew the doors off the building in Week 16, and Matt Flynn had his big Week 17, so a strict "first half/second half" split doesn't show any change.
The big story with the Giants offense this season is that their running game faltered, but they overcame that thanks to Eli Manning having his best season and a huge breakout from wide receiver Victor Cruz. Cruz was great all year, and even better in recent weeks. Even more important, he was best when it mattered most. Cruz had a 72 percent catch rate on third down. He caught 28 of 39 passes for 708 yards and five touchdowns, including four plays over 60 yards and six over 40 yards. It all works out to 254 DYAR. The next highest receiver was Marques Colston, with 170 DYAR on third down.
While the Packers defensive backs will dedicate the most effort to stopping Cruz and Hakeem Nicks, Manning should look to use his tight ends as weapons in this game. Green Bay ranked 26th in DVOA against tight ends, and gave up a league-leading (opponent-adjusted) 79 yards per game to tight ends. In the first game, Jake Ballard and Travis Beckum combined for four receptions, 114 yards, and a touchdown (plus an almost-touchdown by Ballard that was caught just barely out of bounds). The Giants don't really need to keep tight ends in to block against the surprisingly impotent Green Bay pass rush, which ranked dead last with 4.8 percent Adjusted Sack Rate. The Packers didn't have much pass rush after Clay Matthews. Matthews may have had only six sacks this year, but he did lead the league with 19 quarterback hits. (We don't have full hurry numbers to give yet.)
The lack of a pass rush is surely part of the reason why the Green Bay secondary struggled so much this year compared to last year. Based on our game charting stats, Charles Woodson didn't decline very much. We have him at 58 percent Success Rate and 6.6 yards per play. Tramon Williams was one of the best corners in the league in 2010, but dropped to 55 percent Success Rate and 8.0 yards per play in 2011. And Sam Shields really suffered from a sophomore slump. His current charting numbers give him 43 percent Success Rate and 8.9 yards allowed per play. Shields has been particularly susceptible to quick hitches and quick slants this season. He'll be over on the offense's left side when the Packers are using three defensive backs.
Although the Packers have a reputation for bringing heat, that heat is pretty much always just one extra guy. The Packers rushed six or more on just six percent of pass plays, but led the league by rushing five on 37 percent of pass plays. That's good because Manning just destroyed big blitzes this year. In current charting numbers, he leads the league with 11.9 yards per play against big blitzes. However, he averages fewer yards per play against a five-man rush (6.7) than he does against a four-man rush (7.7).
The running game has been better since Ahmad Bradshaw came back from injury in Week 13, although Bradshaw himself hasn't been that great. Bradshaw is averaging 3.81 yards per carry since Week 15, including the wild card game. Brandon Jacobs is averaging 5.44 yards per carry over that same timeframe. Overall, the Giants' rushing DVOA has gone from -5.6% in Weeks 1-12 to 9.8% in Weeks 13-18. The Packers' run defense has been as bad as their pass defense this year, and less surprising, since their run defense was bad last year as well. The Packers are 26th in run defense DVOA and 30th in Adjusted Line Yards. The Giants will be able to gain yards on the ground. However, they still may not be able to gain "yard" on the ground when they need to. The Giants were 27th with a 53 percent conversion rate on power runs (third/fourth down or goal line, 1-2 yards to go). The Packers defense was 24th at 67 percent.
The Giants also shouldn't feel the need to "establish the run" by handing the ball off on first down. The Packers give up big yardage right from the start of every series, ranking 31st in defensive DVOA on first downs.
As many know, the Giants offense has been excellent at the end of games this year. The Giants are fourth in DVOA in the fourth quarter and Manning set a new NFL record with 15 fourth-quarter touchdowns. Green Bay ranked 30th in defensive DVOA in the fourth quarter, but a lot of that is playing prevent defenses with big leads. The Packers' defensive DVOA was 36.5% in the fourth quarter with a lead of more than a touchdown, but 11.5% in the fourth quarter when the score was within a touchdown.
Finally, there's the question of whether the Giants will have trouble punching it into the end zone when they get down near the opposing goal line. After some of the research we've done this year, we're a bit unsure about whether there's real consistency to a team's performance in the red zone. If there is, the Giants may run into a problem kicking field goals instead of scoring touchdowns, because their offense had a lower DVOA in the red zone while the Packers' defense improved significantly when opponents were in the red zone.
When I wrote an ESPN column on the Packers around midseason, I suggested that the Giants would be the team that ended their unbeaten streak (and I was almost right). The logic was simple. The Giants have one of the few quarterbacks with a reasonable chance of keeping up with Aaron Rodgers. On defense, the Giants have a strong front four, which gives them the ability to pressure the quarterback without blitzing. You do not want to blitz Aaron Rodgers, period. Every year, Rodgers has better numbers against the blitz than he does against a standard pass rush. For 2011 in charting we've compiled so far, Rodgers has 7.8 yards per play against three or four pass rushers, 9.2 yards per play against five, and 10.8 yards per play against six or more.
According to our game charting, the Giants only big-blitzed three times in the first game. Rodgers completed two of those passes for a 21-yard gain and a 20-yard touchdown. They sent five men on nine plays, and Rodgers gained 9.2 yards per play: 5-of-6 passing plus a sack and two third-down scrambles that each picked up a new first down. When the Giants sent four, Rodgers averaged "only" 7.4 yards per play. The moral, once again: DO NOT BLITZ Aaron Rodgers.
Despite all the talent on their front four, the Giants did blitz a bit this year. They blitzed five on 20 percent of pass plays, and six or more on nine percent. Both figures were about league average. And Adjusted Sack Rate suggests that the Giants pass rush may have been a little bit overrated this year: good, but not stellar. The Giants actually finished just 10th in ASR at 7.4 percent despite being tied for third in the NFL with 48 sacks. They had lots of sacks in part because they faced so many pass plays (635, sixth in the league). They also played a schedule that was slightly tipped towards teams that gave up more sacks than average. This is not to take anything away from Jason Pierre-Paul, but the rest of the Giants' pass rushers had stats that weren't quite up to his standard. JPP had 16.5 sacks, and Osi Umenyiora was second at 9.0. In the game charting we've compiled so far, JPP has 17 hurries, and Umenyiora is again second with nine. The gap is bigger in quarterback hits, where JPP had 14 and Mathias Kiwanuka was second with six.
Conventional wisdom says that injuries have held back the Giants pass rush. If that's true, the numbers suggest that missing Umenyiora was a bigger deal than missing Justin Tuck. Umenyiora had more sacks and more hurries in fewer games. The Giants had only a 5.9 percent Adjusted Sack Rate with Umenyiora hurt but 8.6 percent ASR with Umenyiora healthy. On ther other hand, the Giants had a 10.9 percent ASR when Tuck was out, but just 6.2 percent ASR in the games he played.
Umenyiora didn't play in the first Green Bay-New York game, but neither did two Packers starters on the offensive line, left tackle Chad Clifton and right guard Josh Sitton. They both returned by the end of the regular season and should be good to go. Wide receiver Greg Jennings should also be back from the sprained MCL that cost him the last three games of the year. Right tackle Bryan Bulaga has been out two games with a knee injury and may play but still isn't 100 percent; you've got to figure the Packers need to give him tight end help, especially if JPP lines up on him.
It's tough to pick out specific trends when it comes to the Packers pass offense, which was excellent in all ways, on all occasions, with all receivers, and anywhere on the field Rodgers wanted to throw. Fifteen Packers were thrown passes this year and only one of them, tight end D.J. Williams, had a negative receiving DVOA -- on four targets. In this game, the best target is going to be whoever is covered by Aaron Ross. Corey Webster had a pretty good season, but Ross was forced into the starting lineup by injuries and was not very good (10.1 yards per play, 43 percent Success Rate) by our numbers. In the first game, seven of the nine passes where we have Ross listed in coverage were to Jennings. You do not want to put Aaron Ross on Greg Jennings, but of course you don't want to have to put Aaron Ross on any of these guys so the Giants are sort of stuck.
Green Bay's running attack has been better than generally reported. Much like with New Orleans and New England, conventional wisdom has confused high yardage totals with the ability to gain yardage on running plays. Green Bay's numbers are even more messed up because they won so many games and had so many kneeldowns. The Packers knelt down 23 times for -22 yards. The official stats say that the Packers gained 3.94 yards per carry. Take out the kneeldowns and a couple of aborted shotgun snaps, and the Packers actually gained 4.26 yards per carry. Rodgers is a very efficient scrambler, and both James Starks and Ryan Grant had positive rushing DVOA. The Packers aren't so great starting off with a run on first down -- their running backs averaged 4.2 yards per carry on first-down runs, 19th in the NFL -- but they were pretty good running on later downs, ranking eighth in rushing DVOA on second down and second in rushing DVOA on third down. Those are mostly medium-length situations, not short-yardage; the Packers were 21st in "power running."
As on offense, the Giants don't have strong red zone stats on defense. The Giants ranked just 27th in the league in DVOA in the red zone. In the first matchup of these teams, the Packers made it into the red zone five times. Four times they scored a touchdown, and the fifth time they kicked the game-winning field goal.
Green Bay has a small advantage here, but this is not a special teams mismatch. The strengths of the Packers (kickoff and punt returns) match the strengths of the Giants (kickoffs and punts). Giants punt returns were poor, and otherwise these two teams are pretty mediocre on special teams.
The Giants are a good team, they've played very well over the last couple weeks, and they provide particularly difficult matchup issues for the Packers. And yet, it seems almost like people are underestimating the Packers just because they haven't seen Rodgers on the field since December. This was still the best team in the league in 2011. This offense is ridiculously powerful. The Giants are supposed to have a defense that can slow down the Packers and yet the Packers still scored 38 points when these teams first played. This is probably going to be another close game and the Giants will definitely put Aaron Rodgers on his back a few times. Manning will get his points too. But in the end, the Packers are just better, and they are at home. They have to be favored.
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 rolling average of the last five games. Note that even though the chart appears in the section for when each team has the ball, it represents total performance, not just offense.
41 comments, Last at 17 Jan 2012, 11:07am by Eddo