In preparation for the 2015 NFL season, FiveThirtyEight is running a series of eight division previews, each highlighting the numbers that may influence a team’s performance (including projections and rankings based on ESPN’s preseason Football Power Index). Up today: the NFC East. It may not always be the best — or even the most competitive — division in football, but it’s always the NFL’s most visceral, featuring four teams that aren’t so much rivals as mortal enemies. Beneath the violence, though, the East is also about contrasting levels of continuity, and perhaps never more so than in the wake of the Eagles’ startling offseason makeover.
2014 Record: 10-6 | 2015 Projected Wins: 9.4 | Playoff Odds: 63.6%
Offensive Rank: 7th | Defensive Rank: 16th | Special Teams Rank: 3rd
Coach Chip Kelly is running the Eagles like the Ship of Theseus. If a team is the sum of its parts, and a ton of those parts change, is it still the same team?
During the spring, Kelly jettisoned his starting quarterback, Nick Foles, for a promising QB with balsa-wood knees, Sam Bradford.1 He sent Pro Bowl running back LeSean McCoy to Buffalo in exchange for an injured linebacker. He ditched a whole slew of veteran standbys.2 He guaranteed $21 million to running back DeMarco Murray within days of signing RB Ryan Mathews for $5 million guaranteed. He went on to spend a guaranteed $95.1 million more in free agency. Also, Tim Tebow now plays for this team.
In other words, the overarching theme of the Eagles’ offseason was chaos — far more so than would be expected of a team coming off a 10-6 campaign.
To measure the continuity (or discontinuity) of a team’s roster, we tracked the percentage of its players’ Approximate Value3 (AV) from one season that returned to the team the next year. Since 1970, the average NFL team has kept about 78 percent of its AV from the previous year, with the other 22 percent departing in the league’s cycle of life — players retire, get traded, leave in free agency, etc.
Teams as good as Philly was last season tend to retain more talent than average. Since the Eagles went 10-6 a year ago, we’d have expected them to bring back 82 percent of their AV, which would rank 11th-highest in football this season. But all that chaos means that Philadelphia’s projected continuity percentage4 for 2015 is merely 68 percent, good for fourth-lowest in the league.
Only the San Francisco 49ers had a larger negative disparity between their expected and actual continuity rates, and the Niners might have just suffered through the worst offseason in NFL history. By contrast, the Eagles’ moves were largely by choice: Kelly turned this roster upside down as part of his grand plan to bring a Super Bowl to Philadelphia.
If he succeeds, though, he’ll be bucking four and a half decades of NFL history. Even after controlling for their Simple Rating System (SRS) scores the previous season, teams that had as much roster turnover5 as the Eagles tended to win about a half-game fewer than would be expected if they’d posted an average rate of continuity.
As we said in March, Kelly might be some sort of lunatic savant. But it remains to be seen if his Ship of Theseus is seaworthy enough to get the Eagles back to the playoffs.
FiveThirtyEight is previewing the 2015 NFL season ahead of the first game of the year. Check out our coverage of every division »
2014 Record: 12-4 | 2015 Proj. W: 9.3 | Playoff Odds: 62.2%
Off. Rank: 4th | Def. Rank: 17th | S.T. Rank: 23rd
Cowboys owner/general manager/Senior Senator from Naboo Jerry Jones has a meddler’s reputation for a reason: In a given season over the past quarter-century, Dallas was either breaking in a new quarterback or a new coach 48 percent of the time. But for the last four years, that hasn’t been the case, as coach Jason Garrett and quarterback Tony Romo have brought an uncharacteristic amount of leadership stability to the franchise.
Continuity of that sort matters in the NFL. Among teams since 19906 that had both the same head coach and leading passer as the previous season, about half went to the playoffs. Meanwhile, teams that were breaking in a new coach or a new QB in a given season made the playoffs only 30 percent of the time, and teams that changed both only reached the postseason 22 percent of the time.
|New Coach, Same QB||30.4%|
|New QB, Same Coach||30.8|
|New QB, New Coach||22.1|
|Same QB, Same Coach||49.5|
While some of that is a mix of correlation and causation — good teams, if given the choice, tend not to abandon their coach/QB combinations — more continuity is associated with greater success even after controlling for a team’s existing talent level. A logistic regression on all NFL teams since 1990 shows that an average team7 loses 6 percentage points of playoff odds when it changes head coaches (but keeps the same QB), 11 percentage points when it switches the QB (but keeps the same coach) and 17 percentage points when both are swapped out.
This suggests that constant leadership shake-ups take their toll on a team, with instability at quarterback exacting almost twice the cost (in terms of playoff probability) as a head coach shake-up. And say what you will about Jones as a meddler, but lately he’s erred on the side of ditching coaches, not QBs. While Dallas has gone through three head coaches over the past decade, Romo has been a constant under center throughout (aside from the year he missed 10 games due to injury).
So by resisting the media pressure to churn through new faces at coach and (especially) quarterback, the stability of a fifth straight Garrett-Romo season puts Dallas in a comparatively strong position.
New York Giants
2014 Record: 6-10 | 2015 Proj. W: 7.7 | Playoff Odds: 27.7%
Off. Rank: 11th | Def. Rank: 26th | S.T. Rank: 27th
New York’s obnoxious devotion to its own history is also realized in its player management strategy: The team typically treasures continuity to the point of boredom.8 But consistency at the expense of dynamism has generally paid off for the Giants — two Super Bowls during the Tom Coughlin-Eli Manning years ain’t shabby.
Over the past 25 seasons, the Giants were breaking in a new coach or a new quarterback just 28 percent of the time. Only Green Bay and Houston (a franchise that’s only been around for 13 seasons) have had a lower coach/QB turnover rate.
|NEW COACH||NEW QB||EITHER|
For every 16 games over that span, the Giants also averaged about 8.5 wins, which is pretty typical of a franchise that spends most of its time being neither great nor bad. Coach Tom Coughlin is perpetually described as being on the hot seat, but that might be the opposite of true. Given New York’s resistance to change and Coughlin’s own pattern of solid seasons punctuated by the occasional Super Bowl run, his seat could very well be flame-retardant asbestos, as coarse and ancient as Coughlin himself.
There’s reason to think the Giants will seesaw back toward the realm of decentness in 2015. No team was more severely crippled by injuries in 2014, according to adjusted games lost (AGL), a metric derived by Football Outsiders to account for which teams statistically suffered the most from benched and injured players. Jason Pierre-Paul’s missing finger aside,9 heavily injured teams tend to be healthier in subsequent seasons, because there’s very little relationship between a team’s AGL from one year to the next.
In other words, look for another solid season from an uneventful franchise that works like clockwork. Maybe there are better situations than being a stagnant-by-design team in a division where two teams are on the rise and the fourth can’t fall any further. Then again, if this team is truly like clockwork, it’s destined to win the Super Bowl every four years — and this is that year.
2014 Record: 4-12 | 2015 Proj. W: 5.7 | Playoff Odds: 5.7%
Off. Rank: 27th | Def. Rank: 22nd | S.T. Rank: 12th
Washington is, as always, a disaster. The team’s Elo rating hit its all-time10 nadir in mid-November of last season, and Washington promptly built on that to set new record lows each of the next four weeks. It had the worst passing defense in the league last year by multiple metrics. Former starting quarterback Robert Griffin III is, when healthy, one of the worst in the league. Coach Jay Gruden will sit him in Week 1 in favor of backup Kirk Cousins … who also is one of the worst in the league.
Did we say “disaster”? This is a cordoned-off federal Superfund site. And it all starts with toxic mismanagement.
In a given year since 1990, there was a 1-in-3 chance that Washington was breaking in a new coach, and no quarterback has been the team’s top passer for more than three consecutive seasons since Mark Rypien in 1993. The team’s longest-running coach-QB collaboration was Rypien and Joe Gibbs, which lasted four years.
In the 16 seasons since Daniel Snyder, noted bastion of team-owning excellence, bought the franchise in 1999, Washington has churned through eight head coaches and nine passing leaders. Under Snyder’s watch, there’s been a 69 percent chance that at least one of those two key personnel positions was changed in a given season. Other teams — Cleveland, Detroit and Minnesota — have been slightly more chaotic over the same span, but all it would take for Washington to grab the lead would be Cousins losing his starting job at QB or Gruden getting fired. (Neither of these things is particularly unlikely.)
Washington’s most significant offseason move was hiring a GM, Scot McCloughan, who spent last season living on a farm. But, for once, that’s not the punch line — McCloughan is an acclaimed talent evaluator. And Washington also boasted a slightly above-average level of roster stability this offseason, with 79 percent of the team’s 2014 AV projecting to return to the team in 2015. In a strange twist, the franchise known for its volatility has the highest continuity rate of any team in the NFC East.11 The ongoing preseason RGIII drama suggests that Washington’s days of dysfunction are far from over, but there are at least some signals for hope that the team won’t be defined by tumult and misrule forever.