To figure out what's likely to happen in 2017, we have to take a step back and look at what really went down in 2016. Several underlying metrics have historically been effective in projecting whether teams are likely to improve or decline in the upcoming season. The games aren't played on paper, but the paper can tell us a lot before the games even begin.
Last year, we took a look at those predictive measures and found that the Chargers, Dallas Cowboys, New York Giants and Tennessee Titans were likely to improve, while the Denver Broncos and Carolina Panthers were probably going to decline. This year, we're going to take a deeper look at the individual teams likely to see their fortunes shift in 2017 and use the numbers to explain why.
Below are the five teams most likely to improve and five teams most likely to decline in 2017, sorted by the gap between their win total and their Pythagorean expectation from a year ago. You can find a primer on many of those stats here. I'll get into why the team is likely to improve and decline and then, at the end, address the most obvious comebacks as to why each team might defy the metrics.
Click the links below to read about each team:
Teams likely to improve
Jaguars | Chargers | Browns | Eagles | Cardinals
Teams likely to decline
Raiders | Texans | Dolphins | Giants | Cowboys
Five teams most likely to improve
1. Jacksonville Jaguars (3-13)
Point differential: -82
Pythagorean expectation: 5.9 wins
Record in games decided by seven points or less: 2-8 (.200)
Strength of schedule: 0.497 (14th-easiest in NFL)
I'll forgive you if you're sick of hearing that the Jaguars are about to take a leap forward. The reality is that they've already taken two modest leaps forward. In 2015, they jumped from 3.7 expected wins to 6.4 wins, a mark they mostly maintained last season with those 5.9 Pythagorean-projected victories.
While it was a season with an ugly record, the poor results are mostly explained by that ugly 2-8 mark in games decided by one touchdown or less. You might attribute that to some element of a young Jaguars team not knowing "how to win," but the Jaguars were a totally unremarkable 9-10 in one-score games over the previous three seasons under Gus Bradley.
Teams that are really bad in one-touchdown games often improve the following season. The Jaguars are one of 68 teams between 1989 and 2015 to post a winning percentage of .200 or below in one-score contests. Those teams were a combined 67-377-1 (.152) during their ugly season. The following season, their record in one-score games was 239-289-1 (.453). Their overall win-loss record jumped by an average of three full wins.
The popular story of the Jags in 2016 was the disappointing season from their offense, driven by the inexplicable mechanical breakdown of quarterback Blake Bortles. Indeed, while the hype machine surrounding Bortles was too loud heading into last season, the offense suffered dramatically in falling from 21st in DVOA in 2015 to 27th last season.
What's also true, though, is that the defense got good without anybody realizing. The Jaguars improved superficially, going from 31st in points allowed to 25th, but DVOA tells a clearer story. Bradley's defense jumped all the way from 26th in 2015 to 13th last season, the first time it has ranked in the top half of the league since 2011.
The Jaguars' improvement wasn't quite as obvious for reasons outside of the defense's control. A turnover-happy offense and poor kickoff coverage left the team to face the league's third-worst average starting field position. Paul Posluszny & Co. were left to defend an average of 69.2 yards, while the top-ranked Patriots had nearly six extra yards to work with on a typical defensive possession. Jacksonville also faced 182 meaningful possessions against a league average of 177, giving the opposition five extra trips to score.
The Jags can help their case by forcing a few turnovers. They were dead last in turnover rate on a per-possession basis in 2016 and made their way by forcing the league's second-highest rate of three-and-outs, behind the Texans. Jacksonville should be a little luckier with fumbles; it recovered just 39 percent of fumbles last season, the second-lowest rate in football and ahead of only the Lions. It was even worse on defense, with the Jags recovering just six of 18 fumbles. That percentage is totally random and unlikely to recur next season.
The Jaguars fielded the league's fifth-youngest defense last season, so they are likely to continue improving in 2017. Things are even more promising when you consider that Jacksonville added two of the best defensive players in free agency by signing Calais Campbell and A.J. Bouye.
Campbell continued to play at a Pro Bowl level for the Cardinals last season, racking up 8.0 sacks and 21 knockdowns during his age-30 season. Bouye came out of nowhere to have a mammoth season for the Texans, moving around the defense while looking like Houston's best corner for long stretches of the season. Not bad for a guy who was fourth on the depth chart heading into the year.
Bouye's riskier than Campbell, but if he plays at a similar level to the guy we saw in 2016, the Jaguars are suddenly set at cornerback for years with the 25-year-old Bouye and 22-year-old Jalen Ramsey, who was an instant star as a rookie. Underrated Cowboys safety Barry Church should be an upgrade over the perennially frustrating Johnathan Cyprien. There's a chance this defense could be great as early as this season.
Can the offense keep up? So much depends on Bortles, but there's enough infrastructure around for the former third overall pick to succeed. The Julius Thomas experiment failed, but the Jags go three deep at wideout with Allen Robinson, Allen Hurns and the surprisingly effective Marqise Lee, each of whom are still 25 or younger. They have their most promising No. 1 back since Maurice Jones-Drew in rookie No. 4 overall pick Leonard Fournette, who profiles to be a superstar back. The offensive line could be better by getting rid of the perennially disappointing Luke Joeckel and a less-than-100-percent Kelvin Beachum, but an offseason trade for left tackle Branden Albert didn't work out, as the former Dolphin retired before ever playing in Jacksonville.
Bortles is still the question mark. There are glimmers of hope in his 2016 once you realize that his 2015 season was grossly overrated -- his sack rate and interception rate both declined -- but his time has to be about up. Among the 37 quarterbacks since 1990 who have thrown 1,000 or more passes through their first three seasons, Bortles is 33rd in era-adjusted completion percentage, yards per attempt and adjusted net yards per attempt. The guys joining Bortles at the bottom of the ANY/A+ leaderboard are almost all disappointments, with David Carr, Jake Plummer and Joey Harrington to the north, and Christian Ponder, Rick Mirer and Jeff George to the south. Bortles doesn't even have to be good; if he can just stay out of the way, the Jaguars may have enough talent to win the South.
Comebacks: It's Bortles, right? It's hard to find a good reason to believe the light bulb is suddenly going to turn on for Bortles, who has barely developed or shown much more than pure athleticism during his time in the NFL. Chad Henne hasn't done a much better job of protecting the football as a pro passer, either, so there's not even a game-manager type to turn to on the bench if new coach Doug Marrone decides to change. Jacksonville will likely improve on its record regardless of how Bortles plays, but a mediocre quarterback would probably limit it to six wins.
2. Los Angeles Chargers (5-11)
Point differential: -13
Pythagorean expectation: 7.7 wins
Record in games decided by seven points or less: 1-8 (.111)
Strength of schedule: 0.505 (16th-toughest in NFL)
The numbers were mostly onto something last season. The one notable exception was the Chargers, who seemed likely to vault forward after a 4-12 season in 2015 in which they went 3-8 in one-score games.
While the 2016 Chargers did improve by one win, nobody could argue that they looked better in those close contests. San Diego was an incredible 1-8 in games decided by one touchdown or less. It's one thing for the Chargers to blow a league-high five games they led at halftime. It's another to give away four games in which they either led or were in a tie game with the ball at the two-minute warning. Some of these losses blend together, but it's worth reiterating how many ways the Chargers blew games last season:
In Week 1, they lost to the Chiefs after going up 27-10 with 13 minutes to go, aided by a 17-yard punt inside the two-minute warning that set up the Chiefs with excellent field position.
Two weeks later, the Chargers had a two-point lead against a Colts team facing fourth-and-7 on its own 20-yard line. They allowed a conversion and a 63-yard touchdown pass to T.Y. Hilton three plays later.
The next week, San Diego was up 13 points on the Saints with the ball and 6:50 to go. The Chargers fumbled away the ball on each of their next two plays from scrimmage, setting up the Saints for two short touchdowns.
A week later, the Chargers were stuffed on third-and-2 down three points with 3:05 to go. Punter Drew Kaser subsequently muffed the hold on the ensuing 36-yard field goal attempt, costing the Chargers a shot at tying the game.
Three weeks later, the Chargers couldn't punch the ball in with four chances from the 2-yard line down eight late against the Broncos.
With 1:13 left in a tie game against the Dolphins, the Chargers needed a few more yards to advance from the Miami 42-yard line and set up a game-winning field goal attempt. Philip Rivers threw a slant under duress to Kiko Alonso, who took it to the house for a game-winning Miami touchdown.
Josh Lambo missed a 45-yard field goal that would have sent the Chargers to overtime against the Browns.
Check that again: They lost to the Browns.
That's an unreal string of brutal losses, with three in the first month of the season alone. There was a decent team here. Four of their five wins came against teams with winning records, including a 33-30 victory over the Falcons in Atlanta. They were 13th in DVOA at the midway point of the season before injuries caught up to them, and Mike McCoy's team eventually finished just ahead of the Vikings and Bucs in 20th.
The Chargers are 4-16 (.200) in one-score games over the past two seasons. I understand if anybody who watched the Chargers collapse on a weekly basis in 2016 doesn't believe they're going to be better in close games this season. It's hard for me to believe. It's also extremely likely to happen.
I went ahead and looked at teams that posted particularly brutal records in seven-point games over a two-year stretch. No organization since 1989 has lost more one-score games over a two-year span than the 2015-16 Chargers. Maybe that's the way you describe a bad team. Their .200 winning percentage in those games is the 19th-worst mark, a figure topped by the 2001 Panthers and their 1-13 (.071) record over a two-year stretch. By any measure, the Chargers have been bad at this.
And by any measure, the teams who were bad at this got better. The 29 teams who lost 12 or more one-win games over a two-year stretch were a combined 154-423 (.267) during their period in the wilderness. The following year, those same teams posted a winning record in close games, going 123-113 (.521). They improved their overall win-loss record by an average of 2.2 wins.
If you prefer winning (or losing percentage), the 35 previous teams that won 25 percent of their close games or less over a two-year stretch went 105-433 (.195) in those seven-pointers and then followed that up with a 137-135 (.504) mark the following season. The latter subset of teams improved by 3.4 wins the following season.
No matter how you slice it, two years of bad luck in close games doesn't hold any predictive value. The 2011-12 Panthers went 2-12 in one-score games, lost their first two games in 2013 by a combined six points, and then went 15-3-1 in those same contests over the next three years. Having "learned how to win," Carolina subsequently went 2-6 in one-score games last season.
The Chargers should be in a much better place in terms of personnel. For one, they won't be without star pass-rusher Joey Bosa for the first four games of the season. By the time Bosa made his debut, injuries had sapped the Chargers of several key contributors. Pro Bowl cornerback Jason Verrett went down for the season after four games, while fellow starting corner Brandon Flowers missed 10 games with a pair of concussions.
The receiving corps was even further picked apart. Stevie Johnson hit injured reserve with a torn meniscus before the season even began. Star wideout Keenan Allen tore his ACL during the first half of Week 1. Receiving back Danny Woodhead suffered his own season-ending knee injury the following week. Rivers was throwing to afterthoughts like Dontrelle Inman, Tyrell Williams and Hunter Henry for most of the season.
Flowers, Johnson and Woodhead are gone, and it would be foolish to count on Verrett and Allen to play all 16 games, given that they've failed to do so even once across seven combined campaigns. It's also fair to expect them to make it out of September, though, and if they can last most of the season in combination with the players who emerged in their absence, the Chargers look awful deep at several critical positions. First-round pick Mike Williams would help here, but he's likely to miss all of training camp after injuring his back in May.
The Chargers will almost certainly be better in 2017 under new coach Anthony Lynn. Their ceiling is as high as that of any bad team from last season; it's hardly out of the question that everything comes together and they win 11 or 12 games. More plausibly, they should top .500 and push for a playoff berth. Just what the AFC West needed: another competitive team.
Comebacks: The move to Los Angeles could impact the Chargers. It will be interesting to see how their home-field advantage is affected by the move to a 27,000-seat stadium, the smallest in the league by nearly 30,000 seats. The 1981-82 Raiders are the most recent NFL team to relocate and either post a winning record or make the playoffs, although there's some selection bias there, as the other seven teams to skip town won an average of only 6.8 games the year before they left for greener pastures.
3. Cleveland Browns (1-15)
Point differential: -188
Pythagorean expectation: 3.5 wins
Record in games decided by seven points or less: 1-5 (.167)
Strength of schedule: 0.522 (Fifth-toughest in NFL)
I'm not going to waste your time by arguing that the Browns were secretly good last season. They weren't. Cleveland finished with the league's worst point differential, and only the Jets posted a worse DVOA. The Browns were bad.
I will argue, though, that they weren't quite as bad as that 1-15 record suggests. Cleveland played a brutally tough schedule; I have them facing the fifth-toughest slate in football, while DVOA pegs it as the most difficult schedule of any team last season. FPI estimates the Browns will have a league-average schedule this season.
And while the Browns didn't win until Week 15, they were competitive in other spots. They led the Ravens during the fourth quarter in September. They were a 46-yard field goal away from beating an eventual playoff team in Miami one week later. The Browns weren't historically awful. They were just a run-of-the-mill terrible team. There's a difference there, although it's academic for last year's team.
In part, the Browns were flummoxed by a nearly unprecedented revolving door of quarterbacks. Cleveland probably wasn't going to look good under any circumstances, but coach Hue Jackson was down to third-string rookie Cody Kessler by the end of Week 2 after both Robert Griffin and Josh McCown suffered injuries. In the end, five different quarterbacks threw 20 passes or more for the Browns last season.
The only other time that has happened in the modern NFL without a strike being involved is in 1984, when Mike Ditka cycled through five quarterbacks for a 10-win Bears team. A year later, he narrowed down his list to two and the Bears went 15-1. That isn't about to happen for the Browns, but the trio of Kessler, Brock Osweiler and rookie second-rounder DeShone Kizer should be able to make it through 16 games and outperform the likes of McCown and Kevin Hogan.
Unlike last season, when the Browns let go of several key free agents in an attempt to rack up compensatory picks and repeatedly traded for future draft picks, the 2017 offseason brought an influx of talent to Cleveland. The Browns spent in free agency, notably signing away the top guard on the market from Cincinnati, Kevin Zeitler. Generally useful veterans like Kenny Britt and Jason McCourty were added to the roster, although the former comes as a replacement for breakout wide receiver Terrelle Pryor.
More importantly for Cleveland's future, it imported a huge haul in the 2017 draft class. By Chase Stuart's chart, the Browns used 86.9 points of draft capital this April, which topped the league by a wide margin. The second-placed Saints spent 63.5 points on their picks, with the league average at 44.8 points. The analytically inclined Browns brain trust was able to pull this off while adding an additional 2018 first-round pick from the Texans. Myles Garrett & Co. might not make an enormous difference this season -- and there's no guarantee they'll succeed at all -- but ask the Falcons about how good draft classes can revitalize a defense overnight.
The Browns will be more talented. They'll be luckier. That's true of most 1-15 teams, who improve faster than you think. The Browns are the ninth one-win team since 1989. The previous eight squads improved dramatically overnight, winning an average of 6.6 games the following season. The 2008 Dolphins jumped from 1-15 to a division title and the playoffs. The AFC North will likely be too tough for the Browns to make that sort of leap, but remember that Miami beat out a Patriots team that had gone 16-0 the previous season for the most unlikely post-realignment division title. It took a season-ending injury to Tom Brady in the opener, but weird things happen. A five-win season might not sound like much, but it's a reasonable target for the Browns in 2017.
Comebacks: It's easy to make the lazy joke and say that this is Cleveland and good things don't happen to the Browns, but things could go wrong. Garrett is already struggling through a foot injury. There's no obviously good quarterback on the roster. The Browns shed their two best pass-catchers from last season, Pryor and Gary Barnidge. They may not be very good at developing draft picks. All of those are plausible problems, but even if they're true, sheer randomness should be enough for the Browns to triple their win total from a year ago.
4. Philadelphia Eagles (7-9)
Point differential: +36
Pythagorean expectation: 9.0 wins
Record in games decided by seven points or less: 1-8 (.111)
Strength of schedule: 0.544 (Toughest in the NFL)
No team has a better stat-nerd case for jumping into the postseason in 2017 than the Eagles. Advanced metrics suggest Doug Pederson's team was already playoff-caliber last season; the Eagles finished fourth in DVOA (just ahead of the Steelers) and had the sixth-best point differential in the NFC, which should have been enough to push them into a wild-card spot.
Instead, the Eagles became one of five teams in 2016 to post a losing record despite a positive point differential, which is a particularly weird feat because it hadn't happened once in the league across either of the previous two campaigns. Philadelphia's gap between expected wins and actual wins was the largest of those five, owing to that 1-8 record in games decided by a touchdown.
Carson Wentz & Co. had a serious shot at winning several of those games. They fumbled late with the lead while running out the clock against a Detroit team out of timeouts. They stalled in the red zone on their final drives against the Giants and Washington. Pederson chose to go for two after a late Wentz touchdown pass made it 28-27 against Baltimore and the Eagles didn't convert. They lost in overtime to the Cowboys when Dallas won the coin toss and scored a touchdown.
Those are five good teams, and indeed, the Eagles also had a few impressive victories. They blew out the Steelers by 31 points in Week 3. Jim Schwartz's defense shut down the league's best offense in a 24-15 victory over the Falcons. And they beat a playoff-bound Giants team in Week 16, although their win over the Cowboys in Week 17 was over a group of backups.
Philadelphia started the season against a pair of cupcakes, Chicago and Cleveland, but its schedule from that point forward was brutally difficult. I have them with the league's toughest schedule, while Football Outsiders pegs them for No. 2. Either way, it was rough. FPI believes the Eagles' schedule will be right around the league average this season, which should be markedly easier.
The obvious place for general manager Howie Roseman to invest this offseason was at wide receiver, and the Eagles did just that. Buying low on Alshon Jeffery and Torrey Smith gives Philadelphia a pair of useful weapons who should have plenty of life in their legs. Suddenly, with Jeffery, Smith, Jordan Matthews, Zach Ertz and Darren Sproles, Wentz has several viable targets. Those guys will have to stay healthy, but Wentz might never have to throw a pass to Nelson Agholor again, and that alone should satisfy Eagles fans.
Many observers are expecting a breakout sophomore campaign from Wentz, who got out to a hot start before struggling mightily after Philadelphia's Week 4 bye. The rookie got into a distressing habit of sailing passes over the middle, leading to tipped reception attempts and easy picks. The new wideouts will help matters, if only because teams will actually be worried about Jeffery and Smith beating their cornerbacks if left one-on-one on the outside.
Another subtler factor will be the return of Lane Johnson, who missed 10 games after a PED suspension. Some of this owes to the quirks of scheduling, but Wentz was a totally different player with his star right tackle on the field. With Johnson up front, the rookie quarterback posted a 72.3 QBR and a 97.5 passer rating. While Johnson was sidelined, though, Wentz could muster only a 48.5 QBR and 70.2 passer rating. Having your best players on the field helps.
Comebacks: Wentz has gotten plenty of positive reviews for his performance as a rookie, but the numbers suggest he was pretty bad. Wentz finished with a 52.8 QBR, which was 26th in the league and below Brock Osweiler. His 5.1 ANY/A was 27th in the league, which nestled him between Blake Bortles and Case Keenum. Wentz's average pass traveled just 7.3 yards in the air, which ranked 26th. You can get a sense of his neighborhood as a rookie. I suspect Wentz will improve in his second season, but there's also a chance the numbers are right and Wentz is a fringe-average passer.
Furthermore, the NFC East is a tough place to make a living these days. Even if the Cowboys and Giants regress some, the East was the only division in the league to deliver four teams with a positive point differential last season.
I'd also be worried about how the Eagles match up against their rivals, given that they are as thin as any team in the league at cornerback. Philadelphia used a pair of mid-round picks on cornerbacks, but Sidney Jones is recovering from a torn Achilles and is really a selection for 2018 and beyond. The veteran trio of Jalen Mills, Patrick Robinson, and Ron Brooks will end up covering players like Dez Bryant, Odell Beckham Jr., and Terrelle Pryor this season. That's going to be a weekly mismatch.
5. Arizona Cardinals (7-8-1)
Point differential: +56
Pythagorean expectation: 9.4 wins
Record in games decided by seven points or less: 2-5-1 (.313)
Strength of schedule: 0.472 (Fourth-easiest in NFL)
This should technically be the 49ers, but I hope you'll forgive me in discussing a team that might have slightly more relevance in terms of 2017 contention. (If you're a Niners fan, you can watch the conversation I had about the 49ers with ESPN's Louis Riddick on NFL Live here.)
The Cardinals are an example of why it's dangerous to believe that some coaches or teams have an innate ability to win a disproportionately high percentage of their close games. If you include his run as interim coach of the 2012 Colts, Bruce Arians had won 78.5 percent of his one-score games as a head coach, posting a 22-6 mark over four seasons in charge of Indianapolis and Arizona. Last season, Arians and the Cardinals were 2-5-1 in those same contests, and even those wins were frantic. They needed last-second field goals to beat the 49ers and Seahawks, the latter of which included Arizona's defense blowing a 13-point lead with four minutes to go.
It was a strange season in many ways for Arizona. The Cardinals faced one of the league's easiest schedules and somehow fell off significantly. Not only did their record fall by 5.5 wins, but their Pythagorean expectation declined from the dominant 11.9-win heights of 2015 to 9.4 victories last season. They did this despite returning virtually everyone of note from their 2015 team. In total, they fell from third in DVOA to 16th.
So what happened? The passing offense collapsed, going from third in DVOA in 2015 to 27th. Carson Palmer came back to earth after a career year in 2015. Palmer's sack rate rose behind an offensive line that struggled to protect him at times, as expected starters Jared Veldheer (out for eight games) and Evan Mathis (done for the year after four) missed significant time. Only one of his top three wideouts was useful (Larry Fitzgerald), as Michael Floyd was anonymous before being released after a DUI, while John Brown struggled with sickle cell issues and had a cyst removed from his spine after the season.
David Johnson picked up some of the slack in a remarkable individual effort, but the Cardinals were a totally different team throwing the football in 2016. Arians' offense is predicated upon throwing the ball downfield, but on "deep" passes (16+ yards downfield), Palmer's passer rating fell from 103.7 (seventh in 2015) to 68.3 (24th last season), even while the drop rate on those throws fell from 5.3 percent to 0.9 percent. If Palmer can't effectively throw deep, the Cardinals either need to fundamentally change their offense or find a new quarterback.
If there are questions on offense, there is even more to worry about on the defensive side of the ball. Arizona's defense continued to rank among the league's best last season, finishing third in DVOA, despite having a blinking neon sign screaming "Throw at Me!" on the side of the field across from Patrick Peterson. The Patriots won their season opener against Arizona by exploiting debuting rookie cornerback Brandon Williams, who was benched shortly thereafter. Marcus Cooper restored some level of civility, but special-teams dynamo Justin Bethel remained a mess to the point that Arians called him a failure in progress in December.
If Cooper were still around, cornerback would be settled. He's not, and Cooper isn't the only one out the door. Arizona had eight players rack up 800 snaps or more on defense last season. Six of those eight players were set to hit unrestricted free agency, and general manager Steve Keim was able to bring back only (admittedly the most important) one, Chandler Jones.
The Cardinals lost six key defenders who racked up a combined 4,423 snaps last season, and their salary-cap situation precluded them from investing much into replacing them. The biggest departure was longtime star Calais Campbell, who the Cards will try to replace with 2016 first-round pick (and Arians doghouse resident) Robert Nkemdiche. The secondary lost a pair of starters in Cooper and versatile safety Tony Jefferson, and while the Cardinals will try to replace Jefferson in the short term with Antoine Bethea and the long term with Budda Baker, Jefferson had rounded into one of the league's better safeties before leaving for Baltimore.
One has to wonder whether the Cardinals will consider changing their defensive style. They've lived and died by the blitz under Todd Bowles and then current defensive coordinator James Bettcher, but that philosophy came into play in part because the team had no dominant pass-rusher. They may have two now in Jones and Markus Golden, who combined for 23.5 sacks last season. Seattle was the only other defense in the league with two pass-rushers who racked up double-digit sacks.
Throw in the limping secondary and the Cardinals might need to play more coverage and trust in Jones and Golden to get after the quarterback one-on-one. Arizona will hope to get more out of Tyrann Mathieu, who was limited to free safety for stretches in 2016 while recovering from another torn ACL. If Mathieu can return to his 2015 form as a slot corner, the Cardinals can feel much more comfortable attacking the line of scrimmage. Arians runs hot and cold with rookies, so it'll also be critical to see if he favors first-round pick Haason Reddick, who could figure as a coverage linebacker on passing downs from Week 1.
Arizona might have a wider range of outcomes than just about any team in the league, and given that franchise icon Fitzgerald is set to hit free agency after the season, it may have as much to lose in 2017 as any team. If the offense reverts back to its usual self and Arizona's defensive stars stay healthy and play like Pro Bowlers, the Cardinals could be a juggernaut and Super Bowl contender.
Alternately, there's a chance Palmer is toast and the defense spends half the year trying to replace the players who left town. The first six weeks of the season might be reasonably tough, and Arizona ends with the Giants and Seahawks, but between Week 7 and Week 14, there's an easy slate. The Cardinals play the Seahawks during that period, but they otherwise get a bye, three games against the AFC South, and matchups against the Rams and 49ers. If the Cards can get to that stretch at 3-3, they'll probably go on enough of a run to make it to the postseason.
Comebacks: While players like Palmer, Mathieu and Brown weren't their usual selves last season, the Cardinals got excellent play from a few of their stars. Johnson might have been the best running back in the league. Fitzgerald led the league in receptions at age 33 and posted another 1,000-yard season, albeit while averaging just 9.6 yards per catch as a slot receiver. Jones and Golden were monster pass-rushers. Peterson was a shutdown corner. Each of those guys played all 16 games. It's tough to count on them to all be stars and healthy for the entire season in 2017, and if they're not, then it's even harder for the Cardinals to make one final run toward a Super Bowl with this core of players.
Five teams most likely to decline
1. Oakland Raiders (12-4)
Point differential: +31
Pythagorean expectation: 8.7 wins
Record in games decided by seven points or fewer: 8-1 (.889)
Strength of schedule: 0.526 (Fourth-hardest in NFL)
To put what the Raiders did last season in context, that gap of 3.3 wins between their actual win total and their expected win total is the fourth-largest since 1989. Of the 10 teams with the largest gaps between their actual win total and their Pythagorean win total between 1989 and 2015, seven declined, with three maintaining their previous record. The average drop was 3.4 wins:
Most of the decline for those teams came in close games. During the season in which they seemingly defied math, those 10 teams were a combined 60-12 (.833) in games decided by one touchdown or less. The following season, they didn't just regress toward the mean; they regressed precisely to the mean, going a combined 31-31 (.500) in close games. Their fall-off in the games that weren't close wasn't anywhere near as significant, as they dropped from a .671 win percentage to a .551 clip.
The Raiders, not coincidentally, were one of the best teams in close games in recent memory. Jack Del Rio's team went 8-1 in games decided by seven points or fewer. You won't be surprised to hear me say that teams with that sort of record also struggle to keep it up. The Raiders have one of the 25 best records in one-score games from 1989 on. During their standout seasons, those other 24 teams were a combined 131-6-1 (.953) in one-score games. The following year, those same teams -- stocked with quarterbacks such as Aaron Rodgers, Tom Brady, Ben Roethlisberger and Steve Young -- went a combined 87-88 (.497) in games decided by seven points or fewer. They won an average of three fewer games each.
Is it possible that the Raiders, as the phrase often goes, learned how to win? Maybe. When we talk about teams learning how to win, the idea is that a team will hold a lead in the fourth quarter with excellent defensive work and a running game that burns out the clock. There were examples of that later in the season, but early on, the Raiders won a few games that seem incredibly unsustainable:
In Week 1, Oakland beat the New Orleans Saints when the Raiders scored with 47 seconds to come within one point before successfully converting a two-pointer to make it 35-34. Del Rio's bravery was logical in a high-scoring contest, but even the right call isn't going to result in the ideal outcome every time.
Two weeks later, the Raiders nearly blew a 17-10 lead against the Tennessee Titans when Andre Johnson caught a game-tying touchdown pass on the 13-yard line, only to be flagged for a questionable offensive pass interference call. The Raiders held on for the win.
In Week 4, Oakland blew a nine-point fourth-quarter lead over the Baltimore Ravens in less than three minutes, allowing a touchdown drive before DeAndre Washington fumbled the ball away with 5:42 to go and set up a second Baltimore touchdown. Derek Carr saved them with a 66-yard, game-winning touchdown drive.
The next week, the Chargers were set to kick a 36-yard field goal that would have tied the game with 2:07 to go, only for holder Drew Kaser to fumble the snap.
In Week 8, the Raiders went to overtime with the Tampa Bay Buccaneers after Sebastian Janikowski missed a game-winning 50-yarder at the end of regulation. Oakland committed penalties on each of its first two drives in overtime to push itself out of comfortable field goal range, turning one drive into a 52-yard miss and another into a punt. This time, the defense came up with two stops, giving the offense a third chance, which Carr turned into a touchdown.
It's not to say the Raiders wouldn't have won any of those games otherwise, of course, but most fans would look at them as rather fortunate to come away with five wins in five tries. At the very least, it's hard to reconcile those performances with a team that knows how to win the close ones. Oakland also isn't a particularly young team -- its snap-adjusted age was 26.5 years per player, just above league average -- which interferes with the young-team-learning-how-to-win storyline.
The arrival of Marshawn Lynch might help the Raiders grind out the clock in the fourth quarter, but the perfectly timed offensive pass interference penalties and failed field goal holds won't stick around. Oakland also got a healthy season from its wildly expensive offensive line, with its five starters playing 74 of 80 games. Five of those missed games were from right tackle Austin Howard, who was generally considered to be the line's weakest link before being cut on Friday. Key backup Menelik Watson is also gone, to Denver, so a less effective season from the line could cancel out any improvements from luring Lynch out of retirement.
In so many ways, the Raiders were dependent upon Carr to bail them out, and he did with aplomb. After leading the league with four fourth-quarter comeback victories in 2015, the third-year passer elevated his game and conjured up seven such victories in 2016.
Seven would have tied the likes of Eli and Peyton Manning for the NFL single-season record if it weren't for the fact that Matthew Stafford raised the bar with eight such victories in 2016. It would also be an unprecedented feat to keep up those comebacks. Of the 25 previous quarterbacks with five or more fourth-quarter revivals since the merger, just one -- Steve Bartkowski -- was able to conjure up five comebacks the following season. Carr might have the same unflappable poise and confidence late in games in 2017 that he had last year, but it's unlikely to be as helpful.
Finally, although this seems counterintuitive, there have to be concerns about whether the Raiders can build upon their turnover margin from a year ago. Oakland's defense wasn't very good, finishing 23rd in DVOA, but it was opportunistic enough to produce the league's sixth-best takeaway rate on a per-possession basis. Combined with Carr's chopping his interception rate in half, the Raiders' turnover differential hit plus-16, which was tied with the Chiefs for the best mark in the league.
Winning the turnover battle is a great way to win games, but turnover margin from year to year is markedly inconsistent. As an example, again from 1989 to 2015, there were 41 teams to post a turnover differential between plus-15 and plus-20. Their average turnover margin was plus-17.3. The following year, those teams had an average turnover differential of plus-2.3. They fell off by an average of 15 turnovers, and with it they declined by an average of two wins.
Comebacks: The 2012 Colts loom as a team that might give Raiders fans hope. That team jumped from 2-14 to 11-5 and posted a remarkable 9-1 record in games decided by one touchdown or less. Unlike most teams, though, they kept up their 11-5 record the following season, with Andrew Luck leading them to a 5-1 mark in one-score contests. They got to 11-5 again in 2014 before Luck finally took too many hits and started to miss time, and the Colts have dropped off to .500 since.
The difference is that, while Luck appears to have some magic in one-score games -- the Indianapolis Colts have posted a .714 win percentage in seven-point contests since the 2012 season, the New England Patriots posted a .639 and nobody else even tops .600 -- that hadn't been the case for Carr before 2016. He and the Raiders were 2-5 in one-score contests in 2014 and 5-5 in 2015.
Even if the Raiders decline, though, they probably will still be good. The most plausible outcome is that Oakland takes a step back and ends up as a nine- to 10-win team, which should still keep it in line for both the division title and a wild-card spot.
2. Houston Texans (9-7)
Point differential: -49
Pythagorean expectation: 6.5 wins
Record in games decided by seven points or fewer: 8-2 (.800)
Strength of schedule: 0.512 (12th-hardest in NFL)
One of the way those Colts teams held on to their throne was by dominating a mostly putrid AFC South. Indianapolis was 12-0 vs. the division between 2013 and 2014, and the Texans filled the void once the Colts dropped off. The 2014 and 2015 Texans were genuinely good teams, but the 2016 Texans are about as bad as a division winner can get. Consider that they ranked a lowly 29th in DVOA, safely ensconced between two teams that fired their coaches, the San Francisco 49ers and Los Angeles Rams.
While Houston was in a relatively easy division, it didn't play a particularly charitable schedule; by my accounting, the Texans faced the 12th-toughest slate in the league. The Texans did beat the Kansas City Chiefs by seven points, and they had another seven-point victory over the Detroit Lions, although the latter team ranked 27th in DVOA. Otherwise, Houston was eking out wins against the AFC South and the likes of the Chicago Bears and Cincinnati Bengals.
Meanwhile, its average loss came by 13.3 points, including a 27-point loss to the Patriots with Jacoby Brissett at quarterback on a short week. The Texans were really 8-1 in meaningful one-score games, given that they lost a meaningless Week 17 game to a Matt Cassel-led Titans team -- and even that required a late Brock Osweiler rushing touchdown to make it close. Houston's largest win of the season was over the lowly Bears in the opener by nine points. Meanwhile, it had three losses of 18 points or more.
The Texans joined the privileged ranks of those teams that posted a winning record while being outscored over the season. Houston posted the fourth-worst point differential for a team with a winning record since 1989, and things don't often go well for teams in that bracket. The 10 winning teams with the worst point differentials declined by an average of 1.8 wins the following season. Squads that exceeded their Pythagorean expectation by two to three wins -- a group the Texans squeeze right into at 2.5 wins -- declined by an average of three wins the following season.
Neither of those numbers bode well for the Texans, nor does one element of the game that often regresses toward the mean from year to year. Houston was terrible on special teams last season, finishing dead last in DVOA while finishing well below league average on everything except punt returns. Normally, teams will make a change in their coaching staff or personnel, and that, combined with some randomness, will push a truly bad special-teams unit toward the middle of the pack.
The problem, though, is that the Texans were also last in special-teams DVOA last season, at which point they fired Bob Ligashesky and hired Larry Izzo. They were 28th under Ligashesky in 2014, 29th in 2013 and last in 2012. It has been five seasons since Houston peeked out from the bottom five in the league in special teams. You would think that might inspire personnel changes, but the Texans brought back kicker Nick Novak and punter Shane Lechler on one-year extensions and will return four of their six most frequent special-teams players from a year ago. None of this inspires any confidence that things will be different in 2017.
Things will likely be tougher in the AFC South. The Jacksonville Jaguars imported arguably the two best defensive players in free agency with Calais Campbell and A.J. Bouye, the latter of whom was Houston's top cornerback in 2016. Indianapolis hired general manager Chris Ballard and spent the offseason finally making coherent moves for its defense, although Andrew Luck's shoulder is a concern. The Titans profited from robbing the Rams last year by adding two first-round picks to their roster at positions of notable weakness.
The Texans weren't able to do much in free agency this offseason, thanks to their spending spree from a year ago and the $9 million in dead money they're eating on Osweiler's contract. The one downside to winning its division is that Houston will be stuck playing the Patriots and Chiefs in 2017, while the rest of their division plays lesser teams from the AFC East and West. That might be enough by itself to swing the South.
Comebacks: There are two strong personnel-related arguments for the Texans. One is the return of a Hall of Fame-caliber talent in J.J. Watt, who missed virtually all of the 2016 season with a back injury. Watt matters, of course, but he played at a superstar level in 2013 and a Texans team with many of these same concerns (notably a 5-0 record in one-score games in 2012) fell from 12-4 to 2-14. The Texans defense was also relatively good without Watt last season, finishing seventh in DVOA; even if Watt pushes them into the top five, they would have far more to gain from adding a star on offense.
Of course, the other argument revolves around a serious upgrade at quarterback, where Osweiler was deposed after one season. Will the Texans get better play from their passers at 2017? I find it hard to believe Tom Savage is likely to be better than Osweiler over any stretch of time.
Remember that the Texans drafted Savage in the fourth round of the 2014 draft and thought so highly of him that they signed Brian Hoyer and gave him and Ryan Mallett the passing reps in 2015, then gave Osweiler a huge deal during the subsequent offseason. After a 73-pass sample in which Savage averaged a pedestrian 6.3 yards per attempt, the Texans were sufficiently convinced by Savage to trade two first-round picks and move up for Deshaun Watson this April. If Savage is a viable pro quarterback, it will be a surprise to the franchise that keeps trying to replace him.
It's more plausible that Houston would get better quarterback play from Watson. Osweiler averaged 4.3 adjusted net yards per attempt last season. There have been 10 other passers since 1990 taken between picks 11 and 20 of the first round who have thrown passes as a rookie. Those passers have combined to average 4.9 ANY/A. Watson is certainly a higher-upside option than Osweiler or any of the replacement-level quarterbacks the Texans would have brought in on a backup's salary to compete.
As a rookie, Watson is a high-variance option, as is the case for any debuting quarterback. That's hardly unreasonable, but it doesn't necessarily push the Texans forward more often than it holds them back. As a team built to win now, the Texans should be able to rely upon a dominant defense and put together a passable offense. If they're right about Watson, they could angle for a Super Bowl spot, but it's more plausible that the team goes with Savage before turning to Watson, who mixes flashes of brilliance with the sort of mistakes you make when you're a rookie in a new league.
3. Miami Dolphins (10-6)
Point differential: -17
Pythagorean expectation: 7.6 wins
Record in games decided by seven points or fewer: 8-2 (.800)
Strength of schedule: 0.480 (Eighth-easiest in NFL)
The 2016 Dolphins were, more than most teams, defined by their schedule. It's not just that the Dolphins were 8-2 in one-score games; it's that they were 8-2 with the vast majority of those wins coming against the worst teams in the league.
Take the teams that finished with the 10 worst records in football, the ones that (ignoring trades) would have occupied the top 10 spots in the 2017 draft. Miami went 8-1 against those teams, winning seven of those games by seven points or fewer. The Dolphins went to overtime with the Cleveland Browns. They came within 2 yards of going to overtime with the 49ers. Miami needed a furious fourth-quarter comeback to beat the Jared Goff-led Rams by four. A fourth-quarter kickoff return was the margin of victory in one of their games against the New York Jets. Kiko Alonso picked off a Philip Rivers pass on the edge of what would have been game-winning field goal range for the Chargers and took it to the house for a win. The Dolphins got a 55-yard field goal to save a blown lead against the Buffalo Bills, then had to survive a 45-yard miss from Dan Carpenter in overtime.
Against upper-echelon competition, the Dolphins looked ordinary. They were 1-4 against teams with a winning record, with a notable 15-point victory over the Pittsburgh Steelers in Jay Ajayi's breakout game. The Lions were the only playoff team with a worse record in those games. Several of their losses were hardly close: It's one thing to lose by 17 points against New England, but the Dolphins also lost by 32 points to the Ravens, 15 points to the Bengals and 13 points to the Titans. Oddly enough for a team that rode a second-half winning streak to the playoffs, their best loss was in Week 1, when they couldn't come up with one of two fourth-down stops to beat the Seattle Seahawks.
The slate will be far tougher for the Dolphins in 2017. ESPN's Football Power Index projects that they will face the second-toughest slate of opponents in the league. The schedule for the AFC East as a whole will be tough -- the division swaps out the NFC West and AFC North for the NFC South and AFC West -- but the Dolphins also have to face the Patriots twice and match up with the Titans and Ravens.
They spent all kinds of money this offseason to bring back as much of their playoff team as possible, which seems reasonable if you look at the roster like it's a true 10-win team, but not so much if it's closer to the 7.5-win figure that Pythagoras suggests. Players such as Kenny Stills, Jermon Bushrod, and Andre Branch were more fungible than their new deals might indicate, while veteran additions Lawrence Timmons and Julius Thomas will need to step in and play at a high level despite struggling in spots in 2016.
The offensive line is a particular point of concern. The Dolphins traded Branden Albert to the Jags as part of the pair of transactions that brought Thomas to Miami. Last year's first-round pick, Laremy Tunsil, will move to his natural position at left tackle, and center Mike Pouncey appears to be healthy after missing most of 2016 with a hip injury, but guard is a major concern with journeymen such as Bushrod, Kraig Urbik, Anthony Steen and Ted Larsen in line to compete for jobs.
If Pouncey can't stay on the field -- and he hasn't played a full season since 2012 -- the interior of Miami's offensive line could be really ugly. That could derail the Miami offense despite its many weapons, given that no quarterback struggles more handling pass pressure than Ryan Tannehill.
Comebacks: Dolphins fans are likely pinning their hopes on head coach Adam Gase, who struggled to figure out his running back situation (both in attempting to pay serious money in a restricted offer sheet for C.J. Anderson, and then alternating between as many as four backs before stumbling on Ajayi as a useful starter) but otherwise got the most out of his talents on offense last season.
Is it possible that Gase has the ability to exceed his team's point differential via game management? I would be skeptical. Gase's teams as an offensive coordinator in Denver shot past their Pythagorean expectation twice in two years, but that was also with Peyton Manning at quarterback, and Manning's Colts teams often did the same thing. The Bears actually undershot their Pythagorean expectation by a half-win during Gase's lone season in Chicago.
Furthermore, while there have been plenty of promising young coaches to break out and exceed their Pythagorean expectation as rookies, those same coaches haven't been able to keep it up the following year. There are 12 coaches since 1989 who preceded Gase and posted a win total which was at least 1.5 wins greater than their expected win total. The following year, those coaches produced a combined 108 wins against a Pythagorean expectation of 107.8 wins.
One coach would be familiar to Dolphins fans: Tony Sparano, who led Miami to a shocking division title at 11-5 in 2008 despite posting just 8.8 Pythagorean wins. The following year, the Dolphins fell to 7-9. Gase appears to be a good hire, but he probably doesn't have the ability to defy history. The Dolphins will either improve their underlying performance or see their record suffer for it in 2017.
4. New York Giants (11-5)
Point differential: +26
Pythagorean expectation: 8.8 wins
Record in games decided by seven points or fewer: 8-3 (.727)
Strength of schedule: 0.506 (15th-hardest in NFL)
The numbers suggested the Giants were likely to improve in 2016. Ben McAdoo's first team as head coach was a little better than Tom Coughlin's final squad from 2015, but the results were dramatically different. Both the 2015 and 2016 Giants were 3-2 in games decided by eight or more points, but by flipping its record in close games from 3-8 to 8-3, Big Blue pushed its way back into the postseason.
The biggest reason the Giants turned things around is one you'll rarely hear from a fan: They stayed healthy. It's easy to notice when teams struggle with a season of injuries, but teams who are far healthier than the league average often slip through the cracks. By adjusted games lost, the Giants ranked as the most injured team in the league each of Coughlin's final three seasons at the helm. They ranked as the seventh-healthiest team in football last season.
Specifically, the defensive turnaround was buoyed by a healthy group of stars. The Giants had a hole at free safety heading into the season and struggled with injuries at the position before turning things over to undrafted free agent Andrew Adams, who held his own. Steve Spagnuolo's 10 other projected starters heading into the season missed a total of six games. Rookie cornerback Eli Apple, who moved into the starting lineup in midseason, missed two.
The math depends on who you want to include as a key player, and Landon Collins broke out as a Pro Bowl-caliber safety, but the Giants were built around five key defenders making $6.5 million or more. Those five guys -- Olivier Vernon, Jason Pierre-Paul, Damon Harrison, Janoris Jenkins and Dominique Rodgers-Cromartie -- missed a total of six games. The Giants might very well be healthier than they were in years past with McAdoo taking over for Coughlin, but their defense is unlikely to be as healthy in 2017.
The New York defense improved from 30th to second in both points allowed and DVOA, which is unprecedented in the case of the latter. The previous record was a 27-rank jump, which was pulled off by the 2011 Jaguars. The following year, the Jags fell all the way back to 28th and didn't bound back up toward league average until 2016.
That's an extreme example of what Bill James called the Plexiglass Principle, but teams that suddenly improve like the Giants did often give back some of their gains. Sixteen teams preceding the Giants improved their DVOA rank on defense by 20 spots or more between 1987 and 2015. Those teams declined the next year by an average of just over eight spots in the rankings.
One way that will likely manifest itself is via the red zone. The Giants were the league's best red zone defense a year ago, allowing a league-low 3.6 points per red zone possession. (That figure doesn't include extra points; if it did, the Giants would be closer to 4.0 points per trip but still rank first.) Research that I wrote up last year pointed out that red zone performance is basically random from year to year, although I did notice that the Giants were far better in the red zone in 2015 than they were elsewhere around the field, too. I'm skeptical of that being some sort of real Spagnuolo-driven trend, given the way his defenses have performed in the red zone in the past, as well as the massive personnel turnover for the Giants between those two seasons.
The goal would be for the offense to make up for any decline on the defensive side of the ball, and that's possible, but I'd still be skeptical. Odell Beckham Jr. is incredible, and Sterling Shepard should improve in his second season, but Brandon Marshall might very well be done after declining dramatically in his age-32 season.
Evan Engram could be a fine tight end prospect, but even the league's best tight ends rarely make a serious impact as rookies. Since 2000, first-round tight ends have averaged 29 catches for 317 yards during their freshman campaign. There has been only one tight end in the past 25 years to make the Pro Bowl as a rookie -- Jeremy Shockey in 2002, then of the Giants.
More distressing is the offensive line, which was hardly addressed this offseason and continues to be a major concern. Justin Pugh and Weston Richburg are due to hit unrestricted free agency next offseason, and they're the line's two above-average contributors. Ereck Flowers continues to struggle on the left side, and Jerry Reese's only additions this offseason to help out on the right side were Chargers castoff D.J. Fluker and sixth-round pick Adam Bisnowaty. If the line continues to struggle, the additions at receiver might not matter.
One issue that shouldn't affect things is rest. The Giants face four teams that will be coming off extra rest in 2017, including three teams that will be returning from their bye week, including Andy Reid's famously lethal Chiefs. Per Chase Stuart, the Giants will have 22 fewer days of rest in between games than their opponents this season, the largest gap in the league by a significant margin.
Logically, that seems meaningful, but it doesn't appear to have much of an effect on teams. I calculated relative rest for every team since 1989, and it has no correlation (-0.04) with winning. There have been nine other teams over that time frame with a relative rest difference of 20 days or more, and those nine teams posted a combined record of 76-68. I'd rather be on the team getting extra rest, too, but there's no evidence of it actually affecting teams on the whole.
Comebacks: The Giants were unlucky in some less obvious ways. They were 31st in the Football Outsiders "hidden" special-teams statistics, which measures how opponents perform on special teams on things such as kickoffs and field goals. That number is almost always random from year to year.
McAdoo's team also recovered just 44.2 percent of the fumbles in its games in 2016, which was the eighth-lowest rate in the league and totally random from year to year. It's not crazy to imagine a season in which the Giants are luckier in those elements and the offense improves enough to counteract the defensive decline. It would help to see a disappointing season from the ...
5. Dallas Cowboys (13-3)
Point differential: +115
Pythagorean expectation: 10.9 wins
Record in games decided by seven points or fewer: 7-2 (.778)
Strength of schedule: 0.491 (12th-easiest in NFL)
The Cowboys were the poster boys among teams likely to improve last season, but part of that enthusiasm involved replacing Matt Cassel and Brandon Weeden with Tony Romo. The situation looked less exciting once Romo went down in preseason, only for Dak Prescott to emerge as one of the more unlikely superstars in recent memory. Throw in a dominant running game and the Cowboys rode an irresistible offense to their best season since Bill Parcells was in town.
Prescott and Ezekiel Elliott are back for another go-round, so why should the Cowboys be any worse in 2017? Well, even if we assume that Elliott will avoid a suspension for a domestic abuse accusation, the engine powering the Dallas offense is undergoing repairs. The vaunted Cowboys offensive line is rebuilding. Right tackle Doug Free retired and guard Ronald Leary left in free agency for the Broncos after excelling in La'el Collins's stead last season. Collins will move to right tackle, while 2015 third-round pick Chaz Green seems first in line to start at left guard. The Cowboys are also thinner in the case that anyone gets hurt. The line will not be bad as long as Tyron Smith, Zack Martin and Travis Frederick are still around, but there's going to be an adjustment period.
One of the biggest reasons the Cowboys leaped up the standings in 2016 was their massive improvement in turnover differential. The 2015 Cowboys finished dead last with a turnover margin of minus-22, and we knew this time last year that those teams usually improve significantly the following year. The 2016 Boys were no exception. They improved to a turnover margin of plus-5, a 27-TO swing. Prescott broke Tom Brady's record by starting his career with 176 pass attempts before throwing an interception. An offense with two rookies touching the ball posted the league's fifth-fewest giveaways.
History tells us the sort of leap the Cowboys made almost always gives way to some decline the following season. Teams that improved by 20 or more turnovers in a given season saw their margin decline by more than 11 turnovers the following year. They declined as a group by an average of more than one win. Prescott probably won't post a sub-1 percent interception rate next year. That's reality.
Dallas' defense could help counter that offensive decline by forcing more turnovers, but after jumping from 32nd in takeaways on a per-drive basis to 20th last season, they might be stretched to improve any further. The Cowboys will also have to rebuild their secondary after losing four regulars this offseason. Dallas lost three of their four projected starters from last season, with cornerbacks Morris Claiborne and Brandon Carr joining safety Barry Church in heading out the door. J.J. Wilcox, who started in 2014 and 2015 before giving way to Byron Jones, left for Tampa Bay.
In their stead, the Cowboys have no sure things. They signed Nolan Carroll from the Eagles, but otherwise flooded the position with mid-round draft picks. Jason Garrett and Jerry Jones used their second-, third-, fifth- and sixth-round picks on defensive backs, which will be helpful in the long term but is unlikely to make much of a difference in 2017. Anthony Brown was better than anyone could have expected as a rookie sixth-rounder, and Jones is rounding into an above-average safety, but the Dallas secondary will have growing pains and could be a yearlong problem.
Teams can cover for bad secondaries with an excellent pass rush, and, again, the Cowboys will need to hold on and hope that Rod Marinelli can make magic happen up front. The good news is they finally gave Marinelli a top draft pick, first-rounder Taco Charlton, but the rush will be a work in progress. David Irving, who flashed dominant stretches of play last season, will miss the first four games of the season on a performance-enhancing drug suspension. Damontre Moore is suspended for two games. Randy Gregory is suspended for all of 2017. This defense could be a serious, serious problem.
Comebacks: Even if the Cowboys do decline, their margin for error is higher than everyone else's on this list. Even if their record was a little inflated, their 10.9-win Pythagorean Expectation suggests they were the third-best team in the league last season, behind the Patriots and Atlanta Falcons. DVOA has them just ahead of the Falcons and behind the Pats in second.
The Cowboys could regress toward the mean and still be favored to come away with a playoff spot, if not the NFC East. Barring a serious injury to Prescott, they'll be a good team in 2017.