by Vincent Verhei (SEA-ATL) and Aaron Schatz (GB-DAL)
Our NFC divisional round games feature a pair of rematches from Week 6 of the regular season. The Saturday afternoon game pits the Seattle Seahawks against the Atlanta Falcons in a rematch of Seattle's thrilling win, and then on Sunday we'll watch the Green Bay Packers try to get revenge on the Dallas Cowboys for their two-touchdown loss.
The first Seahawks-Falcons matchup saw the Seahawks jump out to a 17-3 halftime lead on a pair of rushing scores, but then Atlanta rallied with touchdown drives of 75, 79, and 97 yards in their first three drives of the second half. The Seahawks appeared to tie the score in the fourth quarter, but Steven Hauschka missed an extra point. Fortunately for him, he got another try shortly thereafter and hit a field goal to put the Seahawks up 26-24. The Falcons had one more chance to win, but the game ended on fourth down on a controversial non-call.
At the exact same time, the Cowboys were taking care of business in Dallas. Dak Prescott threw three touchdowns and Ezekiel Elliott rushed for 157 yards as the Cowboys led 17-6 at halftime and then 27-9 early in the fourth. The Packers never got closer than 11 points after that, and eventually lost 30-16.
Hopefully we'll get two games like that Atlanta-Seattle showdown, and not so much like the pedestrian Green Bay-Dallas contest. 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.
Game charting data appears courtesy Sports Info Solutions, unless noted. Please remember that all stats represent regular season only, except for weighted DVOA and anything else specifically noted.
Seattle at Atlanta
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.
Very quietly, the Atlanta Falcons have put together one of the NFL's all-time highest-scoring offenses this season. Only eight other teams have ever scored 540 points in the season, and all of them have higher profiles than this year's Falcons. Five of those teams featured all-time great quarterbacks (Peyton Manning, Tom Brady, Aaron Rodgers, or Drew Brees) at the top of their games. One team was fresh off a Super Bowl win, and would play in another Super Bowl the next year. One paired a Hall of Fame wideout in Cris Carter with one of the greatest rookies of all time in Randy Moss and took the league by storm. The eighth was coming off a Super Bowl win, and was on its way to a second-straight appearance.
So why are these Falcons so anonymous despite their production? It could be that for all those points, no individual Falcons are threatening any offensive records. Matt Ryan will probably win the MVP, but he didn't throw for 5,000 yards or 40 touchdowns this year, when Drew Brees and Aaron Rodgers did. Julio Jones' numbers are actually down significantly from 2015. Both Ryan's and Jones' statistics are suppressed because the Falcons were a run-heavy team this year, ranking 26th in pass attempts. And while Atlanta had a great rushing offense this year, they used a committee approach at running back, with Tevin Coleman getting about half as many carries as Devonta Freeman. That's partly why Freeman finished ninth in rushing yards instead of challenging Ezekiel Elliott for the rushing title.
It also must be said that these Falcons scored those 540 points in one of the league's highest-scoring eras. DVOA adjusts for that, and doesn't rank these Falcons anywhere near the best teams on record. Though they were first in both total offense and passing offense DVOA in 2016, they would not finish among the top 25 teams on record in the first category, and not in the top 10 in the latter. Even casual football fans probably realize it's easier to score 500-plus points now than ever before, and adjust their opinions of this Falcons offense appropriately.
The biggest reason the Falcons are being taken lightly, though, is more simple than that: they haven't won enough games, not in the postseason in years past, nor in the regular season this year. Since Ryan and general manager Thomas Dimitroff arrived in 2008, the Falcons have gone 85-59. Only five teams in the NFL have won more regular-season games in that timespan, including just one NFC team, the Packers. However, the Falcons have been almost totally unable to translate that regular-season dominance into postseason victory. They have gone just 1-4 in the playoffs since 2008, and that one win was one of the more unimpressive playoff victories you'll see: a 30-28 win against the Seahawks in 2012, when they blew a 20-0 halftime lead at home against then-rookie Russell Wilson and were bailed out by a last-second Matt Bryant field goal. As for those four defeats, two were nail-biters decided late in the fourth quarter, but two were blowouts lost by 20 points or more. Wins are a team stat, not a quarterback stat, but the unfortunate fact is that Matt Ryan and the Falcons have as many playoff wins as T.J. Yates and the Texans or Tim Tebow and the Broncos, and three fewer than Mark Sanchez and the Jets.
As for 2016, you would be hearing a lot more about those 540 points if the Falcons had won 15 games and a No. 1 seed. Instead they went "only" 11-5, held back by a defense that was 27th in points allowed and often dug holes so deep even the league's best offense couldn't escape. Atlanta gave up at least 24 points in all five of its losses, and the defense was hardly any better in victory than it was in defeat -- the Falcons were the first team in league history to win six games in which it allowed 28 points or more. Overall, the losing team in Atlanta's 16 games this year averaged exactly 24 points. Three touchdowns and a field goal will win you a lot of games in the NFL, but that's just where they get started in Atlanta.
Stepping into battle against this team come the Seattle Seahawks, one of the league's more schizophrenic clubs, and also one of its more dangerous. The Seahawks have won 11 games this year, beating playoff teams like Miami, New England, and then Detroit last week, as well as these Falcons. They also had five losses and a tie, and five of those games came against teams that missed the playoffs. While Atlanta was consistent as most any team in the NFL, Seattle was 29th in both overall and offensive variance. At their worst, they would be one of the weakest teams Atlanta (or anyone else) has played all season. At their best, they are more than capable of upsetting the Falcons on the road.
WHEN THE FALCONS HAVE THE BALL
Earlier we called the Falcons a run-heavy team, and while that is technically true, it is not true all the time. They ran the ball 39 percent of the time in the first quarter, which ranked 27th. But they ran the ball 51 percent of the time in the fourth quarter and overtime, which ranked third. In other words, they come out throwing to build early leads, then start running to kill clock when they are ahead late in games.
And when they pass, they have plenty of weapons. The Falcons had ten players this year with at least 200 yards receiving and 13 players with at least one touchdown catch, most in the league in both categories. In years past, if you stopped Julio Jones, you stopped the Falcons, but that is not even close to true anymore. Jones missed two games this year, and was held to 35 yards or less in three others. The Falcons won all five of those games, with Matt Ryan completing 68 percent of his passes with a 9.2-yard average, 11 touchdowns, and no interceptions. Julio Jones is just a cog in the machine this year.
Mind you, he's a really valuable cog. Though he was just 17th in the league in catches, he was second in receiving yards, and his 17.0 yards per catch was second-best behind DeSean Jackson's 17.9 among thousand-yard receivers this year. He was fourth in catches resulting in first downs, second in catches that gained at least 10 yards, and second in catches that gained at least 20 yards -- all on a team that was in the bottom ten in total pass attempts. This is how you finish first among all wide receivers in DYAR, and second in DVOA.
Yes, Jones was second in DVOA -- behind his own teammate, Taylor Gabriel. This is something of a sample size fluke, because Taylor finished with exactly the 50 targets needed to qualify for our wide receiver tables, but there's no doubt that Gabriel made the most of those 50 targets -- a 70 percent catch rate, 16.5 yards per catch, and a half-dozen touchdowns, tied with Jones for the team lead. And he wasn't just catching red zone slants, either -- those touchdowns went for 9, 25, 35, 47, 64, and 76 yards, a 42.7-yard average on six scoring plays.
While Gabriel was a hit in his first season in Atlanta, Mohamed Sanu, the Falcon's other free-agent wideout acquisition, was a more subdued success, ranking 43rd in DYAR and 34th in DVOA. Mind you, with 32 teams in the league, there's no shame in having a No. 2 receiver finish 34th. The Falcons are deep here, too -- with 32 and 31 targets respectively, neither Aldrick Robinson nor Justin Hardy came terribly close to qualifying for our tables, but both had DVOAs that would have been in the top 20. This shows again that even if you take away Julio Jones, Matt Ryan can still beat you with his fourth and fifth receivers.
Unless, of course, he decides to beat you with his running backs or tight ends instead. Freeman and Coleman both finished with more than 400 yards receiving, and were fifth and sixth among running backs in receiving DYAR this year. Coleman was particularly impressive, leading all running backs in receiving DVOA and producing a passel of explosive plays. Coleman, Atlanta's backup running back, had eight catches this year that gained 20 yards or more. (Jermaine Kearse, Seattle's starting wide receiver, had seven.) As for the tight ends, though Atlanta didn't use them very frequently (as a group, Falcons tight ends had 84 targets, tied for 29th), Austin Hooper still finished 11th among tight ends in DYAR and second in DVOA. That comes with a healthy amount of sample size skepticism -- Hooper had 27 targets, barely reaching the 25-target threshold for tight ends -- but the fact remains that by receiving DVOA, the Falcons finished the year with the top two wide receivers, the top running back, and the second-ranked tight end. Yikes.
How can the Seahawks match up against an offense like that? It's tempting to guess that we'll get 60 minutes of a Richard Sherman-Julio Jones one-on-one matchup, but that's not likely. Current Seahawks defensive coordinator Kris Richard has let Sherman follow top receivers across the field far more often than predecessors Dan Quinn and Gus Bradley, but that's not a primary tactic. More than half of Sherman's targets this year came on the defense's left side, with about one-third on the right side and the leftovers up the middle. When he did go to the right side, it was almost always to cover an elite wideout like Jordy Nelson (three targets to the right), Brandon Marshall (eight), or Jones himself. Jones had ten targets when Atlanta played in Seattle in Week 6, and Sherman was the primary defender on five of them, including four on the defense's right side. So we'll certainly see a fair number of Jones-Sherman battles, but it won't be an all-day thing.
Instead, we're likely to get a heavy dose of the usual Cover-3 scheme the Seahawks have been using for most of the Pete Carroll regime, with Sherman usually manning the left side and DeShawn Shead taking over on the right. Whoever lines up on that side, that's not a good matchup for Seattle -- see last week's Detroit-Seattle preview for a look at Shead's horrible charting numbers. The worse news, though, is still at free safety, where the dropoff from Earl Thomas to Steven Terrell has been as massive as you might think.
We went over Seattle's pass defense DVOA with and without Thomas last week -- minus-6.8% with Thomas, 28.6% without him (including the 17 snaps he played against Carolina in Week 13). On further review, that's a little misleading and based on a few big plays -- nine touchdowns and nine interceptions with Earl, eight touchdowns and two interceptions without him (including the playoff win over Detroit). The differences in yards allowed per pass (7.5, up from 7.0) and sack rate (6.5 percent, down from 7.8 percent) were much smaller, and the Seahawks actually improved in completion percentage allowed (60.6 percent, down from 61.9 percent). Where Seattle really misses Thomas is on deep middle passes. With Thomas playing centerfield, opponents completed only 2-of-11 passes to the deep middle area of the field for 53 yards. That's 4.8 yards per pass, and 5.3 yards per game. Since Thomas went out, opponents have completed 7-of-12 deep middle passes for 248 yards -- 20.7 yards per pass, 35.4 yards per game.
And it just so happens that Matt Ryan loves passing to the deep middle of the field -- only four quarterbacks threw more passes to that region than Ryan's 32, and remember that Ryan threw very few passes overall this year. And he was good at it too, with a league-high 20 completions for 490 yards. His leading receiver to that part of the field, naturally, was Jones, who had nine deep middle catches (nobody else on the team had more than three). So the key matchup in the secondary probably won't be Julio Jones vs. Richard Sherman. More likely, it will be Julio Jones vs. Steven Terrell -- and that is bad news for Seattle.
What's good news for Seattle is that while their secondary looks more overmatched than it has in years, their defensive front should have a big edge over Atlanta's offensive line. The one potentially fatal flaw in the Falcons' passing game is an inability to keep Matt Ryan safe and sound. The Falcons were only 23rd in adjusted sack rate, and 18th in pressure rate. Seattle's defense, meanwhile, was 10th in adjusted sack rate and 11th in pressure rate. Perhaps due to the problems in the secondary, they were more aggressive on defense this year. They used at least five pass rushers on 26.9 percent of opposing plays this year, 16th in the league and their highest rate in the past five seasons. And they really cranked up the pressure on third downs with a blitz rate of 39.4 percent, ninth-highest in the league. Seahawks defensive coordinator Kris Richard has said that to beat Matt Ryan, "you put him on the ground." Does that mean he'll be even more aggressive on Saturday?
While Seattle's front has done a good job of disrupting pass protection, they have been even better at shutting down opposing running games, with linebackers Bobby Wagner and K.J. Wright keying one of the best run defenses we have ever measured. They weren't particularly dominant in any one area of our defensive line stats, but they were good at all of them. They also forced eight fumbles on runs, and they led the league with just 3.4 yards allowed per rush. Seventeen players have carried the ball at least ten times in a single game against Seattle this year, and more of them have averaged fewer than 3.0 yards per carry (five), than have averaged more than 4.0 (three). Freeman (14th in rushing DVOA) and Coleman (12th) had excellent seasons for Atlanta, and their offensive line had some good numbers as well, but the Falcons look overmatched here.
If not, though, it could be a double whammy for Seattle, because success on the ground would give the Falcons ample opportunity to throw play-action passes. Atlanta used play-action on 26 percent of their passing plays, highest rate in the league, though it should be noted that they were second in yards per play with play-action, first without it. The Seahawks, meanwhile, were killed by play-action -- 10th in yards allowed per play without it, but 23rd in yards allowed per play with it. However, as noted, it's likely that Seattle's front will cut down Atlanta's options on the ground. The Falcons will likely need to rely on their edge in the dropback passing game to beat the Seahawks on Saturday.
WHEN THE SEAHAWKS HAVE THE BALL
In a very literal sense, the Seattle offense this season has looked at times like one of the best in the league -- and at times like one of the worst. The top five offenses this season all had full-season DVOAs of at least 16.1%, a mark Seattle surpassed in six games this year. The bottom five offenses, though, all had DVOAs of minus-12.9% or lower, and Seattle's offense fell below that mark five times.
Is there a personnel reason for these wild swings? As we mentioned last week, the constant shuffling on the offensive line can't have helped, and their offensive output has gone way up when the current starting five of George Fant, Mark Glowinski, Justin Britt, Germain Ifedi, and Garry Gilliam has taken the field. However, closer inspection shows that those five may not have made as big an impact on the offense's statistics as it appears on first glance, at least not where we would expect superior blocking to be noticed. When those five men start, Seattle's sack rate is actually higher than when one or more is missing (7.6 percent, up from 6.6 percent), and their knockdown rate has also gone up (19.3 percent from 19.0 percent). The Seahawks were also among the bottom three teams in pressure rate no matter who was starting on the line. While the pass pressure didn't change much, though, Seattle's overall passing attack sure did -- they averaged 7.4 yards per pass play in games with the current starting line, with 14 touchdowns and one interception. When any of those five lineman was not starting, Seattle averaged 6.3 yards per pass play with 11 touchdowns and 11 interceptions. So has the offensive line helped carry Russell Wilson to his biggest games? Or have they just been lucky enough to be on the field when Wilson has played his best? It's probably not a coincidence that this five-man unit didn't see the field until the second half of the year, when Wilson was starting to heal from early-season knee and ankle injuries.
The image painted by the running game is just as murky, and again a rotating cast of characters is likely to blame. Eighteen different players had at least one carry for Seattle this year. That includes five running backs with at least ten runs, plus a backup quarterback who was forced to play tailback late in one game. Their leading rusher was cut in November and now plays for Green Bay. Their second-leading rusher came into the year recovering from a broken ankle and then missed seven games with a broken leg. Their third leading rusher was their gimpy quarterback. Their fourth-leading rusher missed four games with a broken hand, then six more with a broken shoulder blade. Get the picture?
All these ups and downs and ins and outs make it very hard to get a grasp on what Seattle has been for the past four months, so the best we can do might be to look at what the Seahawks have done in the past few weeks. In his past four games, since his five-interception meltdown in Green Bay, Wilson has completed 67.7 percent of his passes for 8.0 yards per toss, with 10 touchdowns and only one interception, though he has been sacked 12 times. Granted, that includes games against the Rams, 49ers, and Lions, but it's not as if he'll be playing the 1985 Bears this weekend either. In those four games, as has been the case for most of the season, Doug Baldwin has been the leading receiver, catching 31-of-41 passes for 354 yards and three touchdowns. In the absence of Tyler Lockett, Paul Richardson has come on strong, catching 11-of-16 passes for 130 yards and two scores. On the other hand, Jimmy Graham (10-16-175-1) and Jermaine Kearse (8-21-95-1) have seen their roles in the offense decrease.
And then there's Thomas Rawls. Half of his rushing yards came in two games this year, in Week 13 against Carolina and last week against Detroit. In those two games, he ran 42 times for 267 yards (6.4 per carry) and three touchdowns. In his other eight games, he ran 94 times for 243 yards (2.6 per carry) with just one touchdown. It's hard to reconcile the guys who just broke Marshawn Lynch's team record for rushing yards in a playoff game with the guy who had eight carries for 8 yards against Arizona two weeks ago, or seven carries for minus-7 yards in Week 2 against the Rams.
While Seattle's offense is a mystery, Atlanta's defense is consistent and simple -- they have consistently been simply awful. True, there was some statistical improvement at the end of the year -- the Falcons were 28th in defensive DVOA in Weeks 1 to 13, and 13th in defense in Weeks 14 to 17 -- but given the nature of some of those games it's reasonable to skeptical about that improvement. Two of the Falcons' three best defensive games of the year came in Weeks 16 and 17 against Carolina and New Orleans -- two teams that had been eliminated from the postseason and had nothing to play for. Another came against the Rams and Jared Goff, who showed in half a season why he had spent September and October nailed to the bench. Besides those three games, Atlanta only had two other outings with a negative defensive DVOA (which means good defense): against Paxton Lynch and the Broncos in Week 5, and against the Cardinals in Week 12.
So no, this is not a good unit, and they are particularly vulnerable to the run. That may be surprising, considering the Falcons ranked 17th in rushing yards allowed -- Atlanta gave up only two 100-yard rushers this year, and never more than Ryan Mathews' 109 in November -- but remember that opponents were often forced to pass to keep up with Matt Ryan and the explosive offense. The Falcons rotate players on their defensive front quite frequently. Only four were on the field for even half of Atlanta's defensive snaps: linebackers Deion Jones (81 percent) and Vic Beasley (60 percent), defensive tackle Grady Jarrett (57 percent), and defensive end Adrian Clayborn (53 percent). All that shuffling, though, has failed to produce a quality run defense. The Falcons were 25th in adjusted line yards and 26th in second-level yards.
The pass defense, then is the relative strength of this unit, and it's built on a very conservative scheme. Atlanta blitzed just 17.2 percent of the time (30th), and used a big blitz of six or more pass rushers on only 2.1 percent of all pass plays (31st). They rely on their defensive front to pressure the quarterback, which means they rely almost exclusively on Vic Beasley to pressure the quarterback. The eighth overall pick in 2015 led the NFL with 15.5 sacks this year, but no other Falcon had more than five. Beasley was also 11th in the league with 30.0 hurries, but nobody else on the team had more than 19.5, and the Falcons were just 22nd overall in rate of pressure.
That put a heavy burden on Atlanta's cornerbacks, who played well under the circumstances, even with Desmond Trufant out for the year with a torn pec. Robert Alford has stepped into the top cornerback role and played admirably, finishing 28th out of 87 qualifying corners in success rate and 39th in yards allowed per pass. Second-year man Jalen Collins was brought along slowly after a four-game suspension to start the year for performance enhancing drugs, but he has been a starter since mid-November and finished 29th and 46th in the same categories. Nickelback Brian Poole was worst of the bunch in success rate (64th) but best in yards allowed per pass (27th). The Falcons have close to zero depth beyond those three -- no other Atlanta corner played more than 116 defensive snaps this season, and they finished 29th in coverage against "other" wide receivers this year -- but Seattle's not the kind of team that can go four-wide and exploit that, unless you want to rely on Tanner McEvoy or Devin Hester to win games in the playoffs. (Then again, this is the same coaching staff and quarterback that nearly made Chris Matthews a Super Bowl MVP.)
The other area where Atlanta has been weak in coverage has been against running backs, where they rank 26th. This means the health of C.J. Prosise could be critical. Prosise led Seattle running backs with 208 receiving yards this year despite playing only six games, but hasn't played since breaking his shoulder blade against Philadelphia in Week 11. Prosise was limited in practice this week and will be a game-time decision against Atlanta.
The Falcons defense has had the luxury this season of playing with the league's best offense. This means they haven't needed to dominate opponents to win games. They have only needed to prevent big plays, force opponents to make long drives down the field, and wait for a dropped pass or a penalty or some other mistake to set up a long-yardage situation and a stop. This seems like a fine strategy against Seattle, an offense that often shoots itself in the foot and relies on big plays to score. Seattle's average touchdown drive this season lasted 6.6 plays. Only the Chiefs (!) and Dolphins averaged fewer plays per touchdown drive. That could change this week, though. Seattle will probably need to target Atlanta's soft run defense to win. That would extend the length of each drive, while suppressing the number of possessions for each team -- always a good strategy against an offensive juggernaut.
We covered Seattle's special teams last week, and little has changed since then. Short version: they're not very good, and worse than their full-season numbers appear because Tyler Lockett is out for the year with a broken leg. Devin Hester was a non-factor in the wild-card win over Detroit, while Steven Hauschka missed his seventh extra point of the year, and Jon Ryan had a 34-yard punt that was fair caught at the Detroit 23.
On the other hand, there figure to be a lot of kickoffs in this game, and Seattle should have the edge there whether kicking or receiving. Eric Weems has been nothing special for most of the season on either kick or punt returns for Atlanta (though he did have a 73-yard punt return against Oakland). The Falcons' kickers, though, have been significantly better than Seattle's this year. Matt Bosher's average punt traveled 2.8 yards further than Jon Ryan's, and he had only one touchback all season. And if the game comes down to one kick, you've got to like Matt Bryant's chances (92 percent on field goals this year, 98 percent on extra points) more than Hauschka's (89 percent, 83 percent).
The Falcons have not been a particularly good home team this season, going 5-3 in Atlanta, including losses to non-playoff teams in Tampa Bay and San Diego. In fact, their offense has played slightly better on the road (27.4% DVOA) than at home (22.9%), as has their defense (4.4% on the road, 12.5% at home). Unfortunately for Seahawks fans, Seattle's home/road splits go in the opposite direction, and to a greater degree. They went 7-1 in Seattle, but just 3-4-1 on the road, including losses or a tie to four non-playoff clubs: the Rams, Cardinals, Saints, and Buccaneers. The Seahawks had a 13.7% offensive DVOA at home, but that free-falled to minus-16.6% on the road. The defensive also suffered away from Century Link Field, going from a minus-16.7% DVOA at home to a minus-5.8% DVOA on road trips. The Falcons have been a better team than Seattle overall this season, but they definitely have flaws, flaws that the Seahawks would have a better chance of exploiting were this game in Puget Sound. In Atlanta, though, they lose that edge they will probably need. Odds are that the Seahawks will become the second football team from Seattle in the last two weeks to see their playoff dreams come to an end in the Georgia Dome.
Green Bay at Dallas
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.
Like the other three games this weekend, this last game is also a rematch from the regular season. Back in Week 6, the Cowboys went on the road and beat Green Bay, 30-16. Although this wasn't the start of a losing streak -- the Packers did beat the Bears the following Thursday night -- it was essentially the start of Green Bay's midseason dip, the period that had so many people asking "what happened to Aaron Rodgers?"
At first glance, the final score of this game looks to be primarily based on turnovers. The Packers turned the ball over four times, losing three of five fumbles plus a Rodgers interception. This has not been a problem in recent weeks, as the Packers have committed just one turnover in their current seven-game winning streak. But dismissing the first matchup is not as easy as saying "OK, well, the turnovers are fixed now." The Packers only committed one of those turnovers in the first half, and Dallas had a turnover too, yet the Cowboys went into halftime up 17-6. Rodgers' first-half totals (15-for-22, 135 yards) don't look so bad, but those totals are padded by five passes for 61 yards in the final 27 seconds of the half, a drive that ran out of time right when it reached field-goal range. The biggest story of the first matchup wasn't Rodgers but rather Cowboys running back Ezekiel Elliott. The Packers run defense had kept running backs to an absurd 2.2 yards per carry through the first four games, but Elliott came in and picked up 157 yards on 5.6 yards per carry.
So here we are three months later, in Dallas this time instead of Green Bay. The Cowboys are essentially the same team they were back in Week 6, and the same team they've been all year. The Packers are playing much better on offense, but worse on defense, setting up a very likely shootout at American Telephone and Telegraph Stadium. Now with extra telegraphs!
WHEN THE COWBOYS HAVE THE BALL
We normally start these playoff game previews by looking at pass offense vs. pass defense, but of course the Dallas offense is really built around Ezekiel Elliott and the running game. And the Green Bay run defense did not come out of that first Dallas matchup anywhere near the same way it came in. You can split the numbers and look at the Packers before Elliott and after Elliott, or you can look at the first half of the season compared to the second half. Each split will show a decline:
|Green Bay Run Defense, 2016|
|Weeks||Run Def DVOA||RB Yd/Carry||Adj Line Yards||Stuff Rate|
The Cowboys' advantage here is biggest at the start of each series. The Cowboys have the No. 3 offensive DVOA in the league on first downs, including No. 1 on running plays. Green Bay, on the other hand, is 28th in first-down defense. However, it's the Packers pass defense that's most vulnerable on first downs. The Packers are 12th in run defense on first downs, but 31st in pass defense. That makes first down the perfect time for the Cowboys to pull out their play-action passes. As you might imagine, the run-first Cowboys use a lot of them. Only Atlanta and Cincinnati used a play fake on a higher percentage of passes. Green Bay significantly struggled against play-action passes this year, allowing 9.2 yards per pass against play-action but only 7.0 yards per pass otherwise. That was one of the ten largest gaps in the league.
The Cowboys' run DVOA was actually quite disappointing on third and fourth downs, ranked only 26th in the league. However, the Packers are just 29th in run defense on third and fourth downs. Also, there's a bit of a fluke here with the Cowboys' rating. Dallas actually converted 62 percent of runs on third or fourth down, better than the NFL average of 54 percent. They converted 73 percent of short-yardage runs. The Cowboys end up with a lower DVOA because a) they tied for the league lead with three fumbles on third-down runs, two on botched handoffs, and b) they have a low average on third-down runs because they didn't have a lot of long third-down running plays. Elliott had no runs longer than 7 yards on third or fourth down, and the Cowboys had only five: a Chris Jones fake punt, a Lucky Whitehead end-around, and three Dak Prescott scrambles. The Packers' poor rating is much more "real," as they forced no fumbles on third or fourth down and allowed more than 7 yards on 14 of 44 runs.
When it's time to pass, all eyes should be on the Cowboys receivers against the Green Bay secondary. As noted last week,
we track three categories of receivers, but if you combine them all, Green Bay comes out 31st in the league covering wide receivers. The Packers allowed 8.86 yards per pass to wideouts, worst in the league and a full yard above the NFL average. Things are better against tight ends (seventh) and running backs as receivers (15th). Jason Witten caught just four passes for 42 yards in the first matchup, while Elliott had just two catches for 17 yards.
Quinten Rollins has been practicing all week and should be returning from a concussion this week, but according to SIS charting metrics he was the worst of the three primary Packers cornerbacks although he generally covered shorter routes. In general, the Packers use Damarious Randall on the left side (offensive right), Ladarius Gunter on the right side (offensive left), and Rollins and Micah Hyde in the slot. The players here are ranked out of 84 qualifying corners:
|Green Bay Cornerback Charting Metrics, 2016|
|Player||G||GS||Tgts||Yd/Pass||Rank||Suc Rate||Rank||Avg Air Yds||Rank||YAC||Rank|
|Micah Hyde (CB/S hybrid)||16||11||45||6.8||--||49%||--||8.8||--||2.1||--|
Rollins' charting metrics suggest that Cole Beasley, not Dez Bryant, may be Prescott's No. 1 target in this game. Beasley was the Cowboys' top receiving target on both first and third downs this year, and particularly stood out with 50.1% receiving DVOA on third down. Bryant is the most frequent target on second down, with nearly half his targets. That probably is related to the use of play-action, but as I note above, the Cowboys might want to switch and use play-action more on first down than on second.
As I've written a couple of times this season, the lauded Cowboys offensive line allowed a surprising amount of pass pressure, rakning 22nd in the league by SIS charting numbers (subscription required). It does look like a lot of that pressure has to do with blitzing Prescott, but he wasn't blitzed as much as you might expect: 25 percent of pass plays, slightly above average. The problem is that opposing pressure was much more likely to get to Prescott with a blitz. SIS recorded pass pressure against Prescott on 35 percent of blitzes, the fifth highest rate among 34 qualifying quarterbacks. (By comparison, Aaron Rodgers was pressured on 27 percent of blitzes, even though the overall pressure rate for the Green Bay offense was roughly equal to the pressure rate for the Dallas offense.)
The Packers allowed 8.0 yards per play when rushing the standard four men, the worst figure in the league. They were better blitzing, although it was a high-risk, high-reward proposition: the Packers allowed 5.9 yards per play on the blitz, sixth in the NFL, but their 53 percent success rate ranked 17th. Dak Prescott's numbers were fairly consistent no matter the number of pass rushers, which might be another testament to the Dallas offensive line (and the ability of Prescott and center Travis Frederick to identify blitzers.) The Packers bring their pressure from all around the defense, and SIS counted five different defenders with at least 10 hurries. Datone Jones led with 19.5, Nick Perry had 17.5, Mike Daniels 15.5, Clay Matthews 15, and Julius Peppers 14. SIS charting suggests that right tackle is the weak link of the Dallas offensive line, as Doug Free was near the top of the league in blown blocks with the rest of the Dallas line far behind him. (Blown blocks here incorporates blown blocks leading to sacks and hurries but also blown blocks leading to run stuffs.)
There's one other issue worth noting as we get into the playoffs. We still have very little clue how the Dak Prescott will do if the Cowboys fall behind early, especially if the Cowboys are still behind in the second half and the threat of using Elliott is minimal. Prescott only threw 16 passes this year with the Cowboys losing by more than a touchdown, and only half of those came after halftime: Week 8 against Philadelphia, when the Cowboys made up a 10-point deficit and went on to win in overtime.
WHEN THE PACKERS HAVE THE BALL
One of the confusing questions about the Green Bay offense is figuring out how much Jordy Nelson's ACL injury last season contributed to the "struggles" of Aaron Rodgers. (Aaron Rodgers struggling, of course, is still an above-average NFL quarterback, but not the MVP candidate of the past seven weeks.) Last year, it certainly seemed like removing Nelson from the Green Bay offense caused the entire Jenga board to collapse, especially given how brutal Davante Adams was. But this year, Rodgers was down for a few weeks at midseason again, despite the presence of Nelson in the lineup, not to mention the fact that Adams played better than he did a year ago. (Better, but he still has trouble getting open and drops a lot of passes.) So it's hard to tell just how much to discount the Packers offense with Nelson absent due to broken ribs suffered in the wild-card game last week. Even though he didn't seem to fully recover his speed from before the ACL injury, Nelson still ranked third in the league in receiving DYAR this season. Nelson will be replaced as an outside receiver by rookie Geronimo Allison, who had 15.3% receiving DVOA on 22 targets during the regular season.
When two teams have a playoff rematch, the football media generally makes too much out of the numbers from the first meeting. However, the strategy Dallas used to slow down Aaron Rodgers is very important because a) it worked all season for Packers opponents and therefore b) they are likely to use it again this week. The strategy here is to send fewer pass rushers and throw as many defenders into coverage as possible.
According to Sports Info Solutions charting, Dallas rushed only three men on 13 of 46 pass plays in Week 6, holding Rodgers to 5.8 yards per play. It was a particularly good strategy on third and fourth downs, with Rodgers converting just 1-of-6 on those downs when SIS lists three pass rushers. By contrast, SIS lists only four Dallas blitzes all game. Rodgers converted all four of those passes for 35 yards and two first downs.
It looks like this has been the best way to play against Rodgers all season. For the entire year, Rodgers only averaged 5.0 yards per play with three pass rushers; only Case Keenum and Eli Manning were lower. Compare that to 7.3 yards per pass against four pass rushers and 7.7 yards per pass against five pass rushers. The Cowboys also were better the smaller their pass rush:
|Aaron Rodgers and Dallas Defense, Based on No. of Pass Rushers, 2016|
|2-3 pass rushers||4 pass rushers||5+ pass rushers|
|Rodgers ranked out of 34 QB; all success rates are from the defensive perspective.|
Rodgers has continued to gain less yardage against three-man rushes even during his current hot streak. Since Week 12, and including the wild-card game, he has gained 4.9 yards per pass on 31 plays where SIS has marked a two- or three-man rush, compared to 9.0 yards per play against a four-man rush and 7.4 yards per play against a blitz.
Like Green Bay, Dallas will have a cornerback returning from injury. However, the Cowboys are getting back the player who was their best cornerback this season, Morris Claiborne. For the first half of this season, Claiborne finally fulfilled the potential he had when the Cowboys chose him out of LSU with the sixth overall pick in the 2012 draft. But halfway throught he year, he suffered a groin injury and missed the rest of the regular season. This week he'll finally be back in the lineup. The other Dallas cornerbacks are close enough to each other that it's hard to definitely say any one of them had a better year than any other. However, all of the Dallas cornerbacks have something in common. They all rank better in yards allowed per pass than they rank in success rate, by about 20 places in the rankings. Similarly, the Cowboys as a whole rank 29th in DVOA against short passes (up to 15 yards through the air) but fourth in DVOA against deep passes. This all fits the pattern of a team with a lot of late-game leads: you let your opponent complete passes underneath as long as you don't get burned deep in a way that lets the opponent start a big comeback. It also fits the style that worked for the Cowboys against the Packers in Week 6: don't blitz, rush only three a lot of the time, and flood the field with coverage to prevent big gains even when Rodgers can find an open receiver.
|Dallas Cornerback Charting Metrics, 2016|
|Player||G||GS||Tgts||Yd/Pass||Rank||Suc Rate||Rank||Avg Air Yds||Rank||YAC||Rank|
|Byron Jones (CB/S hybrid)||16||16||63||6.2||17||54%||35||10.8||62||2.0||49|
The biggest weakness Dallas has in pass coverage is against tight ends, where the Cowboys rank just 30th in DVOA. Tight ends caught 78 percent of passes against Dallas this year, the highest catch rate against any defense in the league. Jared Cook missed the first Dallas-Green Bay matchup. Richard Rodgers had just two catches for 12 yards, but I would expect more than that from Cook.
There's another important pattern from the first matchup, but unlike the Cowboys' pass-rush philosophy, this one is unlikely to carry over for this week's rematch. One reason that the Cowboys held down the Packers in Week 6 is that they got a number of important stops on third and fourth down in the first half:
- The Packers could not convert a third-and-1 run from the Dallas 19, leading to a 37-yard field goal.
- Jordy Nelson couldn't get to the sticks with a 4-yard gain on third-and-6 from the Dallas 29, leading to a 43-yard field goal.
- Near the end of the first half, the Packers stalled out with incomplete passes on third-and-5 and fourth-and-5 in no man's land, at the Dallas 38.
The turnovers mostly came in the second half, but the halftime lead was built with these stops. However, over the course of the season, the Packers haven't been particularly bad on third down (fifth in DVOA) and the Cowboys haven't been particularly good (18th in DVOA, including 25th on third-and-short). So the Packers should see fewer drives stall out and should settle for field goals less -- especially if they are willing to put the ball in Rodgers' hands when they need a conversion in short-yardage situations. The Packers converted just 49 percent of short-yardage runs this season, 30th in the NFL, and they were 2-for-4 on these runs in the wild-card game. (Of course, this wasn't particularly a Dallas strength, either. The Cowboys defense allowed conversions on 69 percent of these runs, 25th in the NFL.)
The Packers' running game improved at midseason, just as their passing game has. The Packers ranked 16th in run offense DVOA through Week 9, but were third in Weeks 10-17, trailing only Buffalo and Dallas. On the other hand, the Dallas run defense really is good. A couple weeks ago on Twitter, I made fun of one of the TV commentators for saying that the Cowboys had the best run defense in the league. I don't remember who it was, but I have a bit of a mea culpa. Yes, the Cowboys allowed the fewest rushing yards because they were always ahead, but they also ranked eighth in run defense DVOA and allowed just 3.9 yards per carry. The Packers may want to look to run on second down rather than first. On first down, the Packers offense ranks 12th running the ball while the Dallas run defense is third. On second down, the Packers rank No. 1 in run offense DVOA while the Dallas defense ranks 25th. Unfortunately for the Packers, their directional strength running the ball is to the left, which matches the Cowboys' strength defending the run.
Dallas had better special teams than Green Bay during the regular season, although the Packers' strong special teams performance in the wild-card round makes things closer. It's important to remember that special teams varies from game to game and season to season more than offense or defense, so last week's Packers special teams performance isn't any kind of indicator that they will suddenly have another good game this week. They got a spark from using Jeff Janis on kickoff returns after he only returned six kickoffs during the regular season, but Janis had negative value on those six returns. His positive value in the wild-card game isn't necessarily predictive.
The weakest part of the Packers' special teams was on kickoffs, where they ranked 30th, but that's not exactly a strength of the Cowboys either, as the Cowboys ranked 20th in kickoff return value. Lucky Whitehead is the primary return man on both kickoffs and punts.
Chris Jones had a fine season punting for the Cowboys; Jake Schum had a great game last week for the Packers, but was not as good during the regular season. The other Cowboys standout on special teams is placekicker Dan Bailey, who was seventh in value this year by our numbers.
Missing Jordy Nelson is a problem, but the other reasons for Green Bay's offensive improvement should still hold: more creative play calling, an improved running game, and spectacular throws from Rodgers. It's hard to imagine that the Cowboys will be able to keep the Packers offense as bottled up as they did back in Week 6. But their offense isn't exactly chopped liver. Even during this period when the Packers have been red-hot, the Cowboys have kept their offense on schedule and kept the chains moving at roughly the same rate. The big difference has been turnovers, as the Packers' only turnover since Week 12 came on a botched handoff while the Cowboys turned the ball over six times in Weeks 12-16. (Forget Week 17, which was stupid.)
The Packers aren't likely to suddenly start turning the ball over as frequently as they did back in Week 6, but the Cowboys can still use all that pass coverage to keep the Packers from too many big yardage gains. Their run defense will make it difficult for the Packers to run their way into advantageous down-and-distance situations. You can't hold Aaron Rodgers down, but this offense is good enough to outscore him given the holes on the Packers defense. It's going to be very hard to stop Ezekiel Elliott all day long, and it's going to be hard to cover all the Cowboys receivers well unless you can have constant success blitzing Dak Prescott mercilessly. And while that strategy would probably have some success, it probably wouldn't have enough success.
All year long, the Dallas Cowboys have been powered by their fabulous offense, but both teams in this game have a fabulous offense. This week, the Cowboys are the favorite to move on to the NFC Championship Game because they have the better defense, not the better offense. And a week of rest and home-field advantage certainly don't hurt.
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.