Saturday Divisional Round Preview 2015

Saturday Divisional Round Preview 2015
Saturday Divisional Round Preview 2015
Photo: USA Today Sports Images

by Andrew Healy (BAL-NE) and Vincent Verhei (CAR-SEA)

This year's two No. 1 playoff seeds take the field on Saturday looking to avoid upsets against teams that have given them fits in the past. But how much does the past truly matter when so many players on an NFL team change from year to year? For the New England Patriots, the reason to be scared of Baltimore is not that the Ravens have beaten them twice in the playoffs since 2009 but rather that the Ravens are a better team than most people realize, No. 5 in DVOA and almost tied with the Patriots themselves over the course of the entire regular season. For the Seattle Seahawks, there may be worry about a Panthers team that has played them close in meetings for three straight seasons. But do we learn more from the huge gap between these teams in their performance over the entire 2014 season?

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.

Any game charting data that appears with an asterisk appears courtesy of the ESPN Stats & Information Group and is complete through the end of the season. Other game charting data (such as defensive back coverage stats) is roughly 80 percent complete. Please remember that all stats represent regular season only, except for weighted DVOA and anything else specifically noted.

Baltimore at New England

by Andrew Healy

Ravens on Offense
DVOA 9.7% (9) -3.4% (11)
WEI DVOA 3.0% (9) -7.6% (11)
PASS 32.6% (6) 2.0% (12)
RUSH -6.0% (18) -10.4% (14)
RED ZONE -15.0% (22) -11.4% (10)

Patriots on Offense
DVOA 13.6% (6) -4.6% (8)
WEI DVOA 19.5% (3) -12.1% (6)
PASS 35.0% (5) 5.9% (15)
RUSH -3.6% (14) -19.3% (5)
RED ZONE 5.2% (11) -42.4% (1)

Special Teams
DVOA 8.0% (2) 5.5% (5)
BAL kickoff +8.3 (4) +0.5 (13)
NE kickoff +10.2 (3) +8.4 (3)
BAL punts +17.9 (1) +8.4 (5)
NE punts -1.9 (14) -0.6 (18)
FG/XP +5.3 (7) +10.8 (1)

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.

Under John Harbaugh, the Baltimore Ravens have played 14 playoff games. Adding up the point spreads, the Ravens' expected point differential in those games has been -32. To say that they have done better than expected is like saying that Johnny Manziel was worse than expected in his first NFL start. The Ravens have outperformed Vegas's expectations by more than a touchdown per game, going 10-4 with a point differential of +83. If the point spreads accurately captured the Ravens' prospects, there was only about a 1.4 percent chance that they would do that well.

On Saturday, the Ravens come into Foxboro as seven-point underdogs, almost exactly the same spread as their previous two playoff meetings against the Patriots. The Ravens were underdogs by seven points in 2011 and 7.5 points in 2012. Despite being big underdogs, the Ravens controlled the first game and dominated the second. There are enough constants from those two earlier contests to think that the line might again be undervaluing the Ravens' chances. Baltimore's under-the-radar coach and less-heralded quarterback remain in place, as does New England's more celebrated duo. The Ravens' strong front seven that has physically won against the Patriots' offensive line has different pieces, but remains similarly good.

Nonetheless, Saturday's game is not going to be a replay of postseason games past. This game will be more about the things that have changed from the previous meetings more than it will be about those things which remain the same.


Perhaps the most dramatic change for the Ravens from last year to this one has been the improvement on the offensive line. In 2013, the Ravens had one of the worst lines in football. They ranked dead last both in Adjusted Line Yards and yards per carry. The Ravens were not just the worst running team in football last year; they averaged just 3.1 yards per carry, the lowest single-season yards per carry average since 2000.

This year, the Ravens did not just regress towards mediocrity. They had one of the most remarkable line play turnarounds in NFL history, increasing their yards per carry by more than 1.4 yards, rising all the way to sixth in the league. Adjusting for down, distance, and opponent, the Ravens were even better. They ranked third in the NFL in Adjusted Line Yards. Most of these improvements came about due to changes in personnel. The Ravens had Eugene Monroe for a full season at left tackle, Jeremy Zuttah at center, and got the return from injury of Kelechi Osemele at left guard. Most importantly, the Ravens replaced heartwarming-story-but-football-liability Michael Oher with the effective Rick Wagner at right tackle. The Ravens improved all the way to No. 1 in the league in ALY on runs listed as "right tackle," although it's worth noting that they were also only 18th in ALY on runs listed as "right end."

Moreover, the Patriots will not see Wagner and Yanda on the right side of the Ravens' line. With Wagner on injured reserve after a Week 16 injury, they will likely see Yanda at right tackle and rookie John Urschel at right guard, though Urschel was effective against the Steelers in that role. (Ben Muth covered this and Baltimore's other adjustments on the line earlier this week.

Against the Steelers, the Ravens were also missing Monroe. James Hurst filled in, and the experiment did not go particularly well. The aging James Harrison mostly won the matchup against Hurst despite not accumulating much in the way of stats. Osemele notably bailed him out with a great assist on what became Joe Flacco's touchdown pass to Torrey Smith. Monroe's status this week will be a key against a healthy Chandler Jones. Monroe has been practicing and looks likely to get the start.

While Gary Kubiak used scheme adjustments to keep the Ravens effective against the Steelers' 30th-ranked defense, doing so against the Patriots is a taller order. The Patriots come into Saturday with the 11th-ranked defense by DVOA, and that rank is probably a bit low since the Week 17 scrimmage against the Bills gets treated like any other game. Using the same line as last week against the 12th-ranked Browns' defense in Week 17, the Ravens' offense struggled mightily for most of the game. They had just three points and 179 yards through three quarters before coming alive in the fourth.

For Saturday's game, Harbaugh's coaching staff has demonstrated that they will do what is possible to put their players in position to prosper. They are still lacking key pieces on the offensive line, though. Worse, it is easy to overrate both Flacco's transformation this year and his qualities as a clutch quarterback. First, as I wrote about in this week's Any Given Sunday, Flacco's statistical improvement this season came primarily from playing much better against bad defenses than he had in the past. He played in line with his historical averages (which is to say, pretty average) against good defenses.

Even accounting for that, some would argue that Flacco is a money player, one who comes through the most when the stakes are the highest. That argument seems to make sense on the surface given his playoff record, but it largely falls apart on closer examination. Flacco has two big playoff games against above-average pass defenses. Those were the games against Denver and San Francisco during the 2012 Super Bowl run. His numbers in the former game were aided by one of the worst secondary blunders we will ever see. Other than that, all of his good playoff games have come against bad pass defenses.

To put the random nature of Flacco's "money quarterback" reputation in perspective, take a look at his overall playoff numbers, and then see what those numbers would look like had Rahim Moore knocked that ball in Denver away (which would mean that the rest of his 2012 run wouldn't have happened):

Would Joe Flacco Be Clutch Without Rahim Moore's Mistake?
Flacco's Actual Playoff Stats 225 402 56.0% 2931 21 8 28 184 6.53
Flacco's Stats Without Moore's Mistake 178 326 54.6% 2292 14 8 23 159 5.88

Flacco's overall numbers are nothing special. Without the miracle play to Jacoby Jones, he would have an average net yards per attempt (ANY/A) of 5.88. That number would make his closest playoff comparison among active quarterbacks none other than Tony Romo, who only now might be shedding the unfair choker label he has worn for so long. There is not a lot of fairness, and an awful lot of randomness, in who gets the clutch label and who does not. Flacco deserves credit for his playoff successes, but his postseason reputation is a little overblown.

I don't think it's fair to think of Flacco as a quarterback who is somehow special in the postseason, but against the Patriots, he will certainly continue to benefit from having a new offensive coordinator and scheme that puts him in more favorable positions than those he faced in earlier seasons. His offensive line would likewise be a new asset -- if he had the group that ranked fourth in Adjusted Sack Rate during the regular season. That earlier group might have been particularly well-suited to deal with one of the Patriots' most effective defensive tactics from the regular season, too: the Jamie Collins A-gap blitz. It only seems like the Patriots have a DVOA of -100.0% on those blitzes, but they have consistently yielded sacks or turnovers. With a healthy offensive line, the Ravens would have had two excellent guards in Osemele and Yanda to help center Jeremy Zuttah deal with the A-gap blitz. Now, with Yanda forced to play tackle, Collins may choose to blitz through the gap between Zuttah and Urschel. (We can only hope that if Urschel gives up a sack, he turns to Flacco and says "Did I do that?"

In addition to the emergence of Collins, the Patriots have an elite corner who is more durable than the one they lost early in their last two playoff losses. Before Aqib Talib went down in 2012, the Patriots had held Joe Flacco to one completion on four attempts and two quick punts. After Talib's injury, Flacco and the Ravens scored 28 points, including two touchdowns to the receiver who Talib was born to cover. These kinds of in-game injuries can negate all the planning that great coaches do during the week to overcome personnel shortcomings. Bill Belichick will have a game plan this week that revolves around all the flexibility that having Darrelle Revis affords, and he will likely be able to keep with that plan for the whole game this time.


While the Patriots no longer depend defensively on a player with a long injury history, their offense revolves around the league's preeminent hold-your-breath player. As in, "hold your breath that he might do something incredible," and "hold your breath because the next hit might end his season." The Ravens have physically dominated the Patriots in previous playoff games and will almost certainly attempt to physically go after Rob Gronkowski on Saturday. Of course, the Colts and other teams have seen how Gronkowski responds to physical play by opposing defenses when those hits leave him unaffected. As famous Baltimore resident Omar Little once said, "You come at the king, you best not miss."

Assuming that Gronkowski withstands the assault, the Patriots will be very hard to stop. The Patriots' offense has been uniformly dominant by DVOA for four years whenever their tight end king has been on the field. The Patriots fifth-ranked passing offense this year understates where they and Gronkowski are now. Since Gronkowski returned to playing most of the offensive snaps in Week 5, and excluding the meaningless Week 17 game against the Bills (which he did not play in), the Patriots have been the best offense in football by overall DVOA (24.0%) and passing DVOA (50.6%).

The Ravens, at first glance, do not seem particularly well-suited to slow Gronkowski down. Despite being the eighth-ranked defense overall, the Ravens ranked just 20th against opposing tight ends. Still, just as Bill Belichick schemes to take a weapon away from his opponent, Ravens defensive coordinator Dean Pees showed against Jimmy Graham the kind of strategy he might employ to shut down a receiving tight end. Against the Saints, Graham caught six of ten targets for a relatively low 47 yards, albeit with two touchdowns. Our game charting listed linebacker C.J. Mosley as the primary defender on three of those plays, with safety Will Hill the primary defender on six others. On the plays where Hill was the primary defender, Graham caught one pass for two yards. Though our charting is not quite complete, Hill is listed in the data as the primary defender on just seven other passes the entire season. Slowing Graham down with a defender who played little other man coverage this season speaks to the opponent-specific game planning in which Pees engages. While Hill will likely match up against Gronkowski some of the time, Pees is likely to vary the looks he shows. In addition to using Mosley and double-teams in man coverage, Pees will pair man coverage on Gronk with zone elsewhere.

The Ravens have needed to rely on castoffs due to a carousel of players rotating through their secondary. Five defensive backs who have gotten substantial playing time for the Ravens are now on injured reserve. They are now down to castoffs who have replaced castoffs, plucking Rashaan Melvin off the Dolphins' practice squad in December and turning him into a starter. Still, the secondary mostly held its own against the league's top-ranked pass offense from the regular season. The Patriots are likely to do better than Pittsburgh in calling plays designed to test that secondary (they could hardly do worse), but the Patriots also can't test the Ravens corners with a wide receiver nearly as talented as Antonio Brown.

While Pees has gotten remarkably competent play from the likes of Melvin and Anthony Levine, there are clear limits to what these players can do. The Ravens' pass defense succeeded against Pittsburgh, and will succeed or fail against New England, largely based on whether it can generate a pass rush without sending additional rushers. Against Pittsburgh, the Ravens got their biggest plays from creative zone blitzes with defenders who delayed their drops into short zones. On the Terrell Suggs interception, safety Darian Stewart got a free rush even though only three pass rushers ended up coming.

Pees will undoubtedly throw in some of these zone blitzes against Brady too, but they may not be all that effective. This year, the Patriots' offensive line has looked up to the task of handling pressure from unexpected places, so long as that offensive line consists of Nate Solder, Ryan Wendell, Bryan Stork, Dan Connolly, and Sebastian Vollmer. In contrast to previous years, however, the line has looked very shaky in pass protection with personnel groupings other than the starting five. Overall, the line still finished second in Adjusted Sack Rate, which is remarkable given how they played in the first four weeks.

To get pressure, the Ravens may choose to rely on the blitz even less than usual (they blitz on 30 percent of opposition pass plays, compared to the 31 percent league average) against the Patriots.* While the Ravens allow about half a yard less per play when they blitz than when they do not (the league average is about 0.2 yards less), Brady kills the blitz. The Patriots average almost a full yard more per play when opponents send five or more pass rushers than when they send four or less. In addition, while most teams give up more sacks to the blitz, the Patriots give up fewer.

Why the Ravens Will Blitz Brady Sparingly
Yds Per
Pass Play
Sack %
Patriots' Offense Blitz 7.02 1.19% 2.98%
No Blitz 5.99 1.36% 4.28%
Rest of League Blitz 6.13 2.16% 7.84%
No Blitz 6.22 2.52% 5.83%

Moreover, over the last five years, teams have been more likely to intercept Brady when they send fewer pass rushers, not more. To get a turnover, there are few quarterbacks in the league it makes less sense to blitz than Brady. The Patriots' relative strength against the blitz may reduce the Ravens' tendency to utilize it even below its normal level.

Though they don't rely on the blitz, the Ravens ranked second in sacks for the season and fifth in Adjusted Sack Rate. While Terrell Suggs and Elvis Dumervil rightly get attention for the pressure they bring, the Ravens are dangerous across the defensive line with the return of Haloti Ngata. Ngata and fellow tackle Brandon Williams each picked up one of Baltimore's five sacks against Pittsburgh, and more importantly their pressure up the middle limited Ben Roethlisberger's ability to step up in the pocket when Dumervil and Suggs brought pressure from the outside. The Ravens are one of the best teams in the league at getting home without sending additional pass rushers, a formula that has proven to be the Patriots' undoing in previous playoff losses.

While the Patriots' offensive line will need to at least battle the Ravens' front four to a draw, they may not need to do that much in terms of run blocking. Baltimore is fifth against the run by DVOA and only 15th against the pass. The Patriots' relative offensive weakness is in running, where they are ranked just 14th. Given Baltimore's weak spot on defense being the secondary, the Patriots would be smart to minimize any attempt at equity between running and passing. The Patriots have had the best passing game in football with a full-speed Gronkowski. They would be wise to throw, throw, throw.


Not surprisingly with two smart teams in terms of coaching and personnel, both special teams units are excellent. The Patriots have the fifth-ranked unit by DVOA, while the Ravens rank second. The Patriots have the best field-goal kicking game this year. The Ravens have the best punting game.

While this matchup is close to a wash, each team has its own strength in a place where it might produce a big play. Julian Edelman is tied with Devin Hester for the highest career punt-return average of the last fifty years. The Patriots were just fifth in our punt return rankings because Edelman received just 61 percent of opponents' punts this year. He matched his career average of 12.3 yards on his 25 returns, more than 4 yards more than Danny Amendola. Edelman should be back to receive, and dangerous, on Sunday.

If the Ravens are to get a big play on special teams, it is likely to come on a kickoff return, where they are finished third in our rankings. Returner Jacoby Jones ranks in the top ten in career yards per kickoff return with 27.4 and career kickoff return touchdowns with five.


Unlike the last two times the Patriots and Ravens met in the playoffs, the Patriots do not have a clear edge in overall DVOA. This time, however, the Patriots have health on their side, and that may make the difference. With a healthy secondary, the Ravens' defense would be truly scary with their dominant front seven. If their improved offensive line was intact, the Ravens would likely have found success with runs to the right. But the Ravens are not at full strength in either place, and the Patriots are likely to find ways to exploit those weaknesses. A healthy Rob Gronkowski will stretch the Ravens at the back and Jamie Collins may break through the weakened center of the Ravens' line. We know that Vegas' numbers are usually right, and they have the Patriots as seven-point favorites. That gives the Patriots about a 75 percent chance of winning on Saturday, which seems crazy given their recent history with not-as-good Ravens teams. However, while the Ravens have shown how unpredictable the NFL playoffs can be, the Patriots do have the odds and the injury report in their favor on Saturday.

Carolina at Seattle

by Vince Verhei

Panthers on Offense
DVOA -5.0% (20) -16.3% (1)
WEI DVOA -6.6% (18) -21.4% (1)
PASS 4.1% (23) -9.3% (3)
RUSH -6.0% (17) -25.1% (2)
RED ZONE -14.2% (20) 10.0% (28)
Seahawks on Offense
DVOA 16.7% (5) -1.6% (15)
WEI DVOA 16.6% (4) -18.2% (3)
PASS 19.6% (10) 0.2% (9)
RUSH 29.9% (1) -4.0% (22)
RED ZONE 23.2% (4) 21.2% (31)
Special Teams
DVOA -5.5% (30) -1.7% (19)
CAR kickoff -2.2 (20) -7.5 (31)
SEA kickoff +2.3 (9) +4.5 (11)
CAR punts -23.1 (32) -2.4 (15)
SEA punts -4.1 (22) -4.0 (21)
FG/XP -0.2 (18) +0.8 (13)

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.

This weekend's matchup between the Seattle Seahawks and the Carolina Panthers will be just the fifth NFL playoff game since 1989 in which one team finished the regular season with a DVOA at least 40% higher than its opponent. In the first four such matchups, the 1991 Washington Redskins (56.9% DVOA) defeated the Atlanta Falcons (5.8%) 24-7; a week later, that same Washington team defeated the Detroit Lions (-1.2%) 41-10; the 1998 Minnesota Vikings (23.1%) defeated the Arizona Cardinals (-17.1%) 41-21; and the 2007 New England Patriots (52.9%) fell to the New York Giants (1.9%) 17-14 in Super Bowl XLII. That latter game shows that nothing in the postseason is ever a sure thing, but the prior three show why Seattle is a prohibitive favorite to advance to the NFC Championship Game. Even if we expand our sample to all playoff games where one team's DVOA was 30% higher than that of its opponent, the odds don't change; the favorites in those games have gone 18-6, a .750 winning percentage.

It's not hard to find reasons to favor Seattle in this game; the challenge is to seek out areas where the Panthers might have an advantage, or even a stalemate. We can start with recent history; the Seahawks have played the Panthers in each of the past three years, and though Seattle went 3-0 in those games, all were low-scoring nail-biters: a 16-12 win in 2012, a 12-7 win in 2013, and a 13-9 win in Week 8 of this season. On the other hand, all three of those games were in Carolina, and if the Panthers couldn't seal the deal in three home games against the Seahawks, it's hard to like their odds better knowing the game will be under the lights in Seattle. Weather reports are calling for mild temperatures and rain at gametime, so we should expect the Seahawks to be comfortable, but it's possible the wet conditions will lead to a sloppy play or two that goes the Panthers way. They had better hope so; they're going to need some lucky bounces to escape this weekend with a win.


The Carolina offense is going through a transition phase. More and more, this is becoming Jonathan Stewart's team. The 27-year-old has notched at least 20 carries four times in the past five games. That's as many 20-carry games as he had in his first six seasons. And he's not just slamming into the line for no gain; he's averaging 5.1 yards per carry over those five games. These days, the first step in stopping the Panthers is stopping Stewart.

On the other hand, Stewart's DVOA over those five games is just 0.4%, and the Panthers' average offensive DVOA in those contests was 6.2%, so it's not as if the new attack has been steamrolling opponents. That 5.1-yard average is mainly the product of 15 runs that have gained 10 or more yards, especially his 69-yard touchdown in Week 14 against the Saints. Stewart is averaging 1.37 Open Field Yards per carry in his last five games; to put that into perspective, the Baltimore Ravens led the league this year with 1.30 OFY per carry. At the same time, though, Stewart has been stuffed for no gain or a loss on more than 20 percent of his carries (which is a little worse than average), and his Success Rate has been hovering right around the league-average rate of 45 percent. So though Stewart has been explosive, he has not been particularly reliable or consistent.

How does that match up with Seattle's run defense? Over the course of the season, Seattle's front seven allowed 3.42 Adjusted Line Yards per carry (fifth-best in the league), with a Stuff Rate of 24 percent (sixth), while giving up just 0.59 OFY per carry (eighth). The key words in that sentence, though, are "over the course of the season." The interior of Seattle's defense had been dinged up all year, but then Kam Chancellor returned to the field in Week 11 against Kansas City, and middle linebacker Bobby Wagner returned against Arizona in Week 12. The two had missed a combined seven games in the first part of the season, but since they've been back, the defense has been playing its best football of the year by almost any measure. For six games in a row now, the Seahawks have posted a defensive DVOA of -20.0% or better. (Remember that negative DVOAs mean that the defense is playing well.) They only reached that threshold four times in the first 11 weeks of the season. And since Week 12, the Seahawks lead the league in Adjusted Line Yards (2.86) and OFY (0.22), and they are fifth in Stuff Rate (25 percent). So no, success is not guaranteed for Stewart, no matter how many carries he gets.

Carolina's other weapon on the ground is Cam Newton, who finished second to Russell Wilson in rushing DYAR among quarterbacks this year, after finishing first, first, and fifth in 2011, 2012, and 2013. Gauging a defense's ability to stop running quarterbacks is always tricky, simply because the sample sizes are so low. Seattle's defense, for example, only faced 38 quarterback runs all season, a middle-of-the-pack total. Those 38 runs, however, produced a league-low 45 percent Success Rate, and averaged just 3.95 yards per carry, better than every other defense but one. The evidence we have, then, says that the Seahawks should be able to limit the damage Newton does on the ground.

That leaves Carolina with no option but to try and throw against Seattle's outstanding pass defense. The Panthers' passing game this season has largely been a two-man show, dominated by rookie wideout Kelvin Benjamin (145 targets) and veteran tight end Greg Olsen (123 targets). No other player on the team even hit the 80-target threshold. Benjamin had a big game against the Seahawks in Week 8, catching four passes for 94 yards, including a 51-yard grab that saw Benjamin beat Seattle's two best players, cornerback Richard Sherman and free safety Earl Thomas, outleaping both of them to catch the ball in traffic. That might have been the most impressive play any offensive player made against Seattle all season, and Benjamin was one of only five men to cross the 90-yard threshold against the Seahawks. Olsen, though, had little impact in that game, collecting one catch for 16 yards on only three targets.

That, however, is a sample of just one game. Through 16 games, Benjamin was 67th among wide receivers in both DYAR and DVOA -- above replacement level, but below average. Olsen was a star, though. He was 11th among tight ends in DVOA, and only Rob Gronkowski and Antonio Gates had more DYAR.

How does Seattle's defense match up against that duo? Well, they were dominant against wide receivers, ranking fourth in coverage against No. 1 wideouts, sixth against No. 2s, and fourth against all other wide receivers. All in all, only Cincinnati had a better defense this year on passes to wide receivers. They were only 18th in covering running backs, but that shouldn't matter, because as we noted the Panthers don't throw to their running backs much. The real issue is covering tight ends, and there Seattle ranks ... 18th? Below average?


Or have we? Granted, Antonio Gates shredded Seattle in Week 2, catching each of the seven passes thrown his way for 96 yards and three touchdowns. Otherwise, it's hard to find a tight end who actually had a big day against the Seahawks. Philadelphia's Zach Ertz gained 51 yards in Week 14, but no other tight end even broke 40 yards in a game against Seattle. The Seahawks' struggles against tight ends, such as they were, came in small pieces spread out over many games, not in big chunks at a time. If the Panthers are able to beat Seattle with an offense built around their tight end, they'll be the first to do it since the Chargers some 16 weeks ago.

And that, of course, is assuming the Panthers will be able to keep Newton upright. Carolina's offense finished the season 22nd in Adjusted Sack Rate, while Seattle's defense was 14th in the same category. Both teams, though, showed positive trends towards the end of the year. In the last four weeks of the regular season, after Jonathan Stewart became a workhorse, Carolina's ASR fell to 4.3 percent. Seattle's defensive ASR, however, climbed to 11.5 percent since that loss to Kansas City.

[ad placeholder 4]

It doesn't seem to matter terribly much how many pass rushers Seattle uses. The Panthers averaged 6.2 yards per play in non-blitz situations, and 6.2 against five or more pass rushers, ranking in the low 20s in both categories.* The Seahawks defense was fifth-best when rushing four or less at 6.0 yards per play, and best when blitzing at 4.5 yards per play. However, they were 22nd in blitz frequency, only rushing five or more on 28 percent of opponents' pass plays. If we figure Cam Newton will drop back 40 times on Saturday night, that means he'll probably see about 11 or 12 blitzes. Safety blitzes shouldn't be an issue; Seattle was one of three teams that blitzed a defensive back less than 30 times this year.*

In the charting data we have so far, Richard Sherman is in the top 10 cornerbacks in Success Rate, and in the top 20 in yards per play. However, teams are definitely avoiding him. We've charted more passes thrown at Byron Maxwell than we have at Sherman, even though Maxwell missed three games in the middle of the year. Maxwell gave up 9.0 yards per target and only had a 41 percent Success Rate, so he is beatable. Nickelbacks Marcus Burley and Jeremy Lane both had decent charting stats in limited action, so it seems unlikely that Carolina will get much help from the Philly Browns and Brenton Bursins of the world.

Last year, Sherman took a little heat from some critics because he played almost exclusively on the defense's left side and rarely followed top receivers to the other side of the field. This year, the Seahawks have moved Sherman around more, especially when Maxwell was injured. The last time Seattle played Carolina, Sherman shadowed Benjamin the entire game. Even though Maxwell is back and healthy, it still makes sense for Seattle to stick Sherman on Benjamin and leave Maxwell to handle Jerricho Cotchery.

So what's the best game plan for Carolina's offense against Seattle? Balance will be key. No, Stewart is not likely to dominate the game on the ground, but the more carries he gets, the more likely he is to break a long run. And even a couple of 2-yard gains will keep Carolina out of obvious passing situations where the balance really swings Seattle's way. (Carolina's offensive DVOA on third-and-long: -42.4%. Seattle's defensive DVOA on third-and-long: a jaw-dropping -93.4%.) When the Panthers do pass, they really should think about forcing the ball to Greg Olsen. Pass for pass, Seattle has been much worse against tight ends than they have against wide receivers, and few tight ends are better capable of exploiting that weakness than Olsen. Carolina will have to weather the storm when Seattle blitzes, and look for big plays against four-man rushes. That will be the best time to look for a deep ball to Benjamin, especially if they can somehow get him in coverage against Maxwell. It's almost impossible to see how Carolina can with this game without at least one big Benjamin catch.

And finally, they must hope that their defense plays lights-out, and that ten or 13 points will be enough to win this game.

(One more note on Seattle's defense. The Seahawks have gotten better as they have gotten healthy, but at the same time, they have also been fortunate to play a lot of beat-up teams. Four of their last six games came against backup quarterbacks: Arizona's Drew Stanton and Ryan Lindley, Philadelphia's Mark Sanchez, and St. Louis' Shaun Hill. The other two came against Colin Kaepernick and a 49ers team that was in the midst of a four-game tailspin that cost its coach his job. It's clear that this low level of competition played a part in Seattle's defensive improvement; the only question is to what degree. It's safe to say that the real quality of the Seattle defense lies somewhere in between their six-game closing surge, when their DVOA was -37.9%, and what they did in the first two-thirds of the season, when it was -5.0%. Of course, that would still make them one of the best defenses in the league.)


Perhaps you've heard, Seattle is very, very good at running the football. They finished the season with the fifth-best rushing DVOA we've ever measured. Marshawn Lynch led all running backs in rushing DVOA and was second in rushing DYAR, and Russell Wilson set a new record for quarterbacks with 284 rushing DYAR. And while those two players certainly deserve the lion's share of the credit for Seattle's rushing success, they weren't the only weapons on the team. When any player other than Wilson or Lynch carried the ball for the Seahawks, the team's rushing DVOA was 12.3%. That is much worse than their full-season mark of 29.9% -- but it would still have left them as the best rushing offense in the league.

Given Wilson's record-setting performance, it seems fair to say that containing his rushing ability is the key to stopping Seattle, and there's some truth to that. Wilson gained at least 70 yards on the ground six times this year, and Seattle's offensive DVOA in those games was 24.2%. (For the sake of perspective, Green Bay led the league this season with an offensive DVOA of 24.6%.) In the five games when Wilson ran for 30 to 69 yards, that DVOA fell to 13.1%; and the five times he failed to run for 30 yards, Seattle's offensive DVOA was 10.0%. Now, a 10.0% offensive DVOA is still pretty good -- only eight offenses were that good this season -- but limiting Wilson's running plays certainly seems to cut down on Seattle's production.

There is a third key cog to the rushing game, though, besides Wilson and Lynch. The Seahawks managed to put up this dominant performance even though one of their most important blockers, center Max Unger, missed about two-thirds of the season. There was not much difference in the passing game when he was on the sidelines, but the running game, oh, the running game. In the ten games Unger missed, Seattle had a rushing offense DVOA of 22.6%, with a 44 percent Success Rate, and averaging 5.1 yards per carry. With Unger, however, this year's Seahawks were the NFL's equivalent of Tom Osbourne's Nebraska Cornhuskers, running over teams without mercy or restraint. Seattle's rush offense DVOA in Unger's six starts was 40.3%, with a 60 percent Success Rate and 6.3 yards per carry. Why is this relevant? Unger, who hasn't played since Week 11 due to a high ankle sprain, is expected back in the starting lineup on Saturday night.

Does Carolina stand a chance of slowing Seattle down? Examining the Panthers defense is hard to do, because the unit was in a state of flux pretty much all year long. Only six defenders started more than 11 games, and only five starters in the wild-card win over Arizona were also starters in Week 1 against Tampa Bay (and one of those, defensive tackle Star Lotululei, will miss this weekend's game and the rest of the playoffs after breaking his foot in practice). It has been a long transition process, but it seems to be working. Carolina's defense has progressed steadily throughout the year, improving against the run and the pass, and has played its best football at the end of the season.

The one constant for Carolina: middle linebacker Luke Kuechly, probably one of the best half-dozen defensive players in the league. The third-year linebacker led the NFL this year in percentage of team plays, total Successes, and passing Successes. With Kuechly in the middle, the Panthers' run defense has been fairly steady all season. And generally, they have been pretty good, ranking 10th in Adjusted Line Yards and eight in Stuff Rate. In Open-Field Yards, though, which measures a defense's ability to prevent long runs, they were worst in the league. And though Carolina's defense improved in so many ways at the end of the year, this wasn't one of them. Through Week 13, they were giving up 1.21 OFY per carry. Since Week 14, that number actually got worse, climbing to 1.50. In the last four weeks of the year, they gave up a 63-yard run to Doug Martin, a 26-yarder to Isaiah Crowell, an 18-yarder to Charles Sims, a 17-yarder to Jacquizz Rodgers, and 16-yard runs to Mark Ingram, Pierre Thomas, and even Josh McCown.

Which brings us to Carolina's ability to defend against running quarterbacks. As we mentioned earlier, evidence on this sort of thing is sketchy at best, but the evidence we have says that Carolina, well, sucks at it. Quarterbacks have gained successful yardage against Carolina on 77 percent of their carries, the highest rate in the league, and averaged 7.0 yards per carry, sixth-worst. Now, that's a sample of only 31 runs, so let's not take it as gospel truth. But we probably shouldn't dismiss it entirely, either. In their first meeting this year, Wilson carried the ball four times: scrambles of 14 and 15 yards; a 7-yard gain on second-and-5; and a 3-yard loss and a fumble. So no, he didn't kill Carolina every time he ran -- only three-quarters of the time.

All in all, we might expect Seattle to struggle a little more than usual at running the ball consistently, though they're more likely than ever to rip off long runs. That's really not good news for Carolina, but we can assume that at some points, the Seahawks will want to pass. We noted earlier this season how difficult it is to predict which Seattle receivers will get the bulk of the targets in any given week. Seattle's passing scheme isn't built around forcing the ball to a specific player or two, nor is it built around picking on specific defenders in coverage. By and large, it is built around Russell Wilson scrambling for time and finding receivers who come free as coverage breaks down.

Unfortunately for Wilson, that's going to be more difficult against Carolina now than it would have been earlier in the year. The Panthers' biggest overhaul has been in the secondary. Seventh-year cornerback Antoine Cason, seventh-year safety Thomas DeCoud, and second-year corner Melvin White were all starters on opening day, but DeCoud and White have been benched, while Cason was released in early December. (He signed with Baltimore and will also be playing on Saturday.) In their places are third-year corner Josh Norman and a pair of rookies: safety Tre Boston, a fourth-round draft pick out of North Carolina, and cornerback Bene' Benwikere, a fifth-rounder out of San Jose State.

Our charting data is still incomplete, but the numbers we do have for Carolina show that their new starting corners have been much more effective than the old ones:

Panthers Cornerbacks Charting Stats, 2014
Suc. Rate
Antoine Cason 56 9.1 27%
Melvin White 37 8.7 41%
Josh Norman 36 6.0 56%
Bene' Benwikere 17 5.9 65%
Approximately 20 percent of the season has not been charted yet.

Norman has been starting since midseason, while the two rookies entered the starting lineup in Week 14. If there was a turning point for the Carolina defense, that might have been it. In the first 13 weeks of the year, Carolina's defensive DVOA was 6.5% (14.1% passing, -3.1% rushing). From Week 14 to the end of the regular season, those numbers have improved drastically, to -27.7% overall (-39.0% passing, -7.8% rushing). That does not include last week's destruction of Arizona, which would make these numbers even more impressive, but that probably had more to do with Ryan Lindley than anything Carolina did.

Even still, Carolina's pass coverage in recent games has been outstanding. Their full-season "defense vs. types of receivers" numbers aren't very impressive, but the Panthers have been much better since the Week 14 switch. In the final four games of the season, Carolina ranked sixth in DVOA against wide receivers, seventh against tight ends, and fourth against running backs. It's that last number, which probably undersells Carolina's strength here, that may be most critical against Seattle. Through 16 games, only Cleveland had a better DVOA on passes to running backs than Carolina, and remember that Kuechly led the league in Pass Stops -- in his case, that usually meant tackles on running backs after short catches. Why is this so important against Seattle? The Seahawks didn't have a Le'veon Bell or Matt Forte tallying 100-plus targets, but when Wilson did throw to his running backs, he was outstanding -- no quarterback with at least eight starts had a higher DVOA on passes to running backs. Norman and Benwikere, if their charting numbers can be trusted, might have an advantage over Doug Baldwin and Jermaine Kearse. If that's the case, it might come down to Marshawn Lynch and Robert Turbin to convert critical third downs.

In terms of yards per play, Wilson and the Seahawks have been middle-of-the-pack against three or four pass rushers, third-best against exactly five rushers, and 22nd against six or more. So if the Panthers are going to blitz, they'd better bring the house. That's not likely to be a huge issue for Carolina, which blitzed only 25 percent of the time, ninth-lowest rate in the league.*

And finally, there's the matter of pass rush, where the Panthers' defensive line had the seventh-highest Adjusted Sack Rate, while Seattle's offensive line was 24th. It's quite likely that Wilson will hit the turf more than once on Saturday night.

All told, there's reason to believe that the Panthers defense might be able to hold their own against Seattle's passing attack, especially if you believe that their numbers over the past month are a more accurate measurement of their ability than their full-season statistics. That's not likely to matter, though, if Seattle is running up and down the field with little resistance.


[ad placeholder 3]

The Seahawks were counting on Percy Harvin to return kicks this year, and when that didn't work out, they didn't really have a Plan B. Rookie Paul Richardson has handled the bulk of the kickoff returns since Harvin left -- not very well, mind you. Steven Hauschka has been outstanding on kickoffs (top three in total value), but Seattle's coverage teams have been nothing special. Punter Jon Ryan specializes in not outkicking his coverage -- the Seahawks allowed a league-low 17 punt returns, though those returns averaged 11.5 yards each, third-worst in the NFL.

That's nothing, though, combined to Carolina's horror-show punt team. Brad Nortman's punts have been decent enough (well, except the two that were blocked), but the Panthers have given up 15.5 yards per punt return. That's 3 full yards per return more than any other club. This could be a problem, in a game where they figure to be punting a lot. They are also giving up 32.4 yards per kick return, a few decimal points better than Oakland and worse than anyone else. Fortunately that isn't often an issue, because Graham Gano leads the NFL in touchback percentage on kickoffs. Richardson and Bryan Walters may not be the most dangerous threats to exploit these weaknesses, but the opportunities to make plays should be there.


By DVOA, a Carolina win would not be the biggest playoff upset we have ever recorded -- it would be the second-biggest playoff upset we have ever recorded. The Panthers will need to play a nearly perfect game and get more than their share of breaks just to keep the score close. Nothing is guaranteed until the final gun fires, but according to all available evidence, we shouldn't expect anything other than a comfortable Seattle victory.


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.

Team DVOA numbers incorporate all plays; since passing is generally more efficient than rushing, the average for passing is actually above 0% while the average for rushing is below 0%.

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.


18 comments, Last at 11 Jan 2015, 4:51am

1 Re: Saturday Divisional Round Preview 2015

"Given Baltimore's weak spot on defense being the secondary, the Patriots would be smart to minimize any attempt at equity between running and passing. The Patriots have had the best passing game in football with a full-speed Gronkowski. They would be wise to throw, throw, throw."

To be fair, the Ravens secondary plays short passes very well, and are 3rd in the league in pass DVOA on passes under 15 yards (as mentioned in Film Room). If the Pats are going to throw, then they'll need to throw deep. That's why the Steelers matchup scared Ravens fans more than the Pats, the Steelers have many more big play WR's.

It remains to be seen whether the weather will allow effective deep throws by Brady. I can't imagine an all pass strategy would work great, because it would allow the Ravens to play Dumervil and McPhee full time to rush the passer, normally they play about 50% of the snaps.

2 Re: Saturday Divisional Round Preview 2015

Er... why are you suggesting we judge Flacco's numbers by what they'd look like if he hadn't played the extra two games, again? The Moore play was random but his performances in the following two games sure weren't.

I don't think there's any reason Flacco has been good in the postseason besides small sample size, but it's wrong to say he HASN'T been realy, really good just because a blown defensive play helped them advance.

4 Re: Saturday Divisional Round Preview 2015

I think the point is to show how the narrative can change due randomness. Think of every one and done qb. They never get to play another 2-3 games to inform our views of their play.

I think Flacco is better than the average numbers suggest(the ravens never seem to have qb friendly schemes). He throws medium much more than the other qbs do and outside of this year, his receivers have largely been mediocre(yes boldin was good, but the rest were jacoby jones types, and you could argue they are still an average receiving core this year).

6 Re: Saturday Divisional Round Preview 2015

Well, okay, HISTORY changed due to randomness, but the narrative was invented because of how he actually played in the rest of those games. If he had gone 11 for 35 in the AFCCG, we wouldn't be giving him the benefit of the doubt because a random play got him there, right?

Point is, if you want to say he hasn't played that great of competition in the playoffs, that's fine. And I have no idea how he'll play tomorrow. But to say he hasn't been incredibly good in recent playoff chances is just incorrect.

7 Re: Saturday Divisional Round Preview 2015

This is exactly right. Go ahead and take away the 70 passing yards and the 1 TD that came on the Rahim Moore play, if you must. But then to entirely ignore 2 outstanding games that followed because they "shouldn't have happened" is really, really faulty thinking. It's not like they didn't happen, and don't reflect on Flacco's ability to play.

Though it is kind of fun to imagine "It's A Wonderful Life" playoff scenarios based on things that shouldn't have happened. Take away the Tuck Rule and Brady never won Super Bowl 36. But he made up for that with the perfect season resulting when Tyree's helmet catch disappears. Fun, but not useful analysis.

9 Re: Saturday Divisional Round Preview 2015

Echoing this. People who are reading this preview on this site shouldn't have to have it explained to them that Flacco's outstandingly good postseason numbers are probably a result of small sample size randomness rather than any 'clutch' ability. Stick to telling us what DVOA indicates about the matchups and leave the silly narratives to others.

10 Re: Saturday Divisional Round Preview 2015

I think you missed the point that the analysis isn't about Flacco, it is about the NARRATIVE.

The "clutch qb" narrative emerged from the media due to some luck, and that label would not have happened without it. Its a criticism of how the media generally labels QBs, not a statement that Flacco isn't good in the playoffs. (that argument is one of small sample size).

Although luck directly involving the QB (tuck rule, helmet catch, massive blown coverage) is sometimes involved, other times its the defense or other things the QB does not control such as an otherwise reliable kicker missing a short FG for a win or a game-clinching interception turning into a fumble and new set of downs for the opponent (Sorry, Charger fan bitterness is oozing out).

The "Playoff QB" narrative is almost always just a result of fluky, random events.

Joe Montana went three years straight with one-and-done in the playoffs and people wanted to fire Walsh because they couldn't get it done in the playoffs. Then they won two in a row. Those three years in isolation and Montana's stats in the playoffs look poor. The two years after, and he looks like a playoff god. Average all of his playoff games together, and -- wait for it -- his stats look pretty much just like the regular season ones. Brady and Manning are trending the same way as the number of playoff games piles up.

Flacco, and Romo, if given another 15+ playoff games, would probably look indistinguishable in the playoffs from the regular season.

12 Re: Saturday Divisional Round Preview 2015

That was definitely all I was trying to say. The narrative turns entirely on one crazy play. Flacco deserves credit for everything that came after, as I said in the article. In Flacco's case, it's pretty neat how one play has changed how he's defined. Our readers certainly get small sample size, but I don't know that most people appreciate that the story on Flacco depends so much on the opportunities that came from one very fluky play. The argument on Flacco being clutch is silly and I wanted to point that out as simply as possible.

By the way, I wouldn't rule out Flacco being better this year even. I'd just like to see more good games against good defenses before overruling his much bigger sample of mediocre play.

And thanks Scott for filling in my gaps.

13 Re: Saturday Divisional Round Preview 2015

Exactly right, and I should have been clearer. Flacco had to do something afterwards to make the narrative happen, but the narrative still wouldn't exist without that bit of wild luck. It's interesting to think about. And the randomness of that narrative means we should use regular-season-and-playoffs (including 2012) Flacco to project him and not some imaginary clutch playoffs-only Flacco.

There was a lot of Ravens' love elsewhere in the analysis. Hope you guys didn't miss it with my Flacco thought experiment getting in the way.

11 Re: Saturday Divisional Round Preview 2015

Mostly they seem to be arguing that you can make any quarterback look bad if you selectively manipulate his statistics with an eye towards doing so.

The more common version of this is the obnoxious "if you take out monster run X team Y only put up 3.8 YPC" and the like that pops out on cue every time it's suggested that a team has a good running game. Ok, 3.8 YPC isn't good on average... but how does it compare to the average "YPC after the best carry is removed"? If you don't give context you're just cherrypicking.

14 Re: Saturday Divisional Round Preview 2015

I think Baltimore was the worst possible draw for New England (well, Seattle, then Green Bay, are probably worse, but you expect the Super Bowl opponent to be tough), but obviously 2009, 2011, and 2012 aren't relevant. Isn't this the first postseason Gronk's been healthy? Baltimore still matches up better (corner injuries are less devastating than lb injuries would be against New England), but this is New England's best team since 2007, regardless of dvoa. Their defense won't be pushed around. If the offensive line holds its own, they have a great chance to make the Super Bowl

16 Re: Saturday Divisional Round Preview 2015

2010 was a great season, no doubt. The fact that they lost against the Jets really changes the narrative. They did have some legitimate concerns, they were relying on a great deal of youth (Gronk, Ahern, Spikes, the secondary and DLs) and their receivers were athletically challenged (Branch and Wes coming off ACL). Of course, some of this is flavored by hindsight, I underestimated these issues at the time.

Another more obvious problem was the fact that the team had one healthy DL on the roster, and they had to bring him back early from team suspension to be available for the game. That I was very concerned about. Of course, the OL was better in 2010 than it has been this year, so that balances things out a little.

This team is much healthier and more experienced; it seems likely that we go into the game with more certainty around level of performance. Who knows, though, maybe tomorrow we'll be talking about some "tragic flaw" that should have been apparent all along.

18 Re: Saturday Divisional Round Preview 2015

Flacco's reputation at this point seems to be defined by the one or two clunkers he delivers every year, but he played well enough to win. I think the 'elite' debate is stupid, but I do think that a guy who consistently gives you a chance to contend for Super Bowls is as good as you could want