How The Cowboys Turned Their Season Around


For a 5-5 team, things are looking better than average for the Dallas Cowboys. Their division is a train wreck. The Giants are lilliputian, the Eagles are grounded, and Washington is without its starting quarterback. The Cowboys, somehow, are the ones who seem to have the best shot of making the playoffs.

We’ll know a lot more after Dallas and Washington play in a Thanksgiving showcase on Thursday. Washington is 6-4, but has a far lower Elo than Dallas (1504 vs. 1542), and Washington’s quarterback is now perma-journeyman Colt McCoy after Alex Smith gruesomely broke his leg in his last game. The Cowboys are still only 46 percent likely to make the playoffs, according to FiveThirtyEight’s projections, but a win against Washington would go a long way. That the playoffs are even within sight is a stunning reversal of fortune from earlier this month, when a home loss to the Titans dropped Dallas to 3-5 and had observers calling for head coach Jason Garrett’s head. Making matters worse, star linebacker Sean Lee injured his hamstring again in the Titans loss. The last time that happened, Dallas’s defense went down with him.

Dallas tried to guard against another Lee injury by fortifying their linebacker corps in the 2018 draft. Their first-round pick, Leighton Vander Esch, was expected to be too inexperienced to adapt quickly to the NFL game, but has turned into easily the best linebacker at the top of his draft.

Dallas’s rookie linebacker was a steal

Rookie linebackers drafted in the first round by Pro Football Focus’s grade

Rookie Linebackers Team Draft Position Pro Football Focus Grade
Roquan Smith Bears 8th 62.6 Average
Tremaine Edmunds Bills 16th 49.6 Below average
Leighton Vander Esch Cowboys 19th 84.1 Good
Rashaan Evans Titans 22nd 55.4 Average

Source: Pro Football Focus

Vander Esch has not only replaced Lee, he has arguably surpassed him. Last year, Dallas lost four of their five games without Lee, giving up 13 more points per game 26 and 46 more rushing yards. But this year, with Vander Esch replacing him as starter for five games, Dallas is better — allowing 30 fewer rushing yards per game and 2.4 fewer points. Dallas is now 4-1 when Lee hasn’t played, and Vander Esch single-handedly changed the outcome of Sunday’s game.

Vander Esch is the second Dallas blue-chip defender that general manager and owner’s son Stephen Jones has found relatively cheap in the draft. 2017 Pro Bowl defensive end and sack-master Demarcus Lawrence was grabbed in the second round in 2014.

Lawrence’s ability to generate pressure allows the Cowboys to blitz infrequently, just 20.8 percent of the time, according to ESPN Stats & Information Group. This may limit big defensive plays — the Cowboys have just four interceptions this year — but it seems to prevent big plays by the opponents, too. They have one of the most efficient units in football as measured by yards opposing offenses must gain to score a point.

Much of that stat is driven by red zone success, which is viewed by many as random. But Dallas combines efficiency in the red zone with an ability to stop teams from getting there at all. The Cowboys rank near the top of the league in opponents’ red zone possessions per game.

Dallas’s defense has to be efficient because the offense lacks explosiveness. The offense has put up just 34 plays of 20 or more yards this season, tied for 25th in the league, according to Pro-Football-Reference. Even after the addition of wide receiver Amari Cooper, who cost Dallas a 2019 first-round pick, Dallas’s passing game is stuck mostly in neutral. Prior to Cooper joining the Cowboys in their eighth game, Dallas was 28th in yards per pass play. Since then they are improved slightly, but still rank 23rd.

The most important change to the Cowboys passing offense hasn’t been Cooper but a player who was already on the roster — running back Ezekiel Elliott.

Elliott is a bigger part of the passing game

Ezekiel Elliott’s involvement in the Dallas Cowboys’ passing game, 2016-18

PER GAME
Season Receptions Receiving Yds Targets
2018 4.2 34.1 5.5
2017 2.6 26.9 3.5
2016 2.1 24.2 2.5

Source: ESPN Stats & Information Group

This season, Dallas has figured out that Elliott can be featured as a receiver, too. His role in the passing game27 has more than doubled since his rookie season and is up dramatically from 2017, too. This has been especially the case the past two weeks when Elliott’s running and passing has accounted for 388 of Dallas’s 733 scrimmage yards (53 percent)

Of course, the Cowboys are just one ill-timed stinker away from wrecking their season, too. Using our ELO projections, only one game in Week 12 holds more importance for the playoffs than Washington-Dallas:

The best matchups of Week 12

Week 12 games by ranking of average Elo ratings (using the harmonic mean) plus ranking of total potential swing for the two teams’ playoff chances, according to FiveThirtyEight’s NFL predictions

Playoff % Playoff %
Team A Current Avg. Chg* Team B Current Avg. Chg* Total Change Game Quality
SEA 45.0% +/-22.7 CAR 61.5% +/-21.6 44.3 1564
DAL 46.3 22.4 WSH 54.7 20.1 42.5 1522
MIN 52.3 17.3 GB 16.4 13.7 31.0 1522
TEN 36.2 20.2 HOU 84.1 13.0 33.2 1516
ATL 9.0 6.0 NO 99.8 0.3 6.3 1615
DEN 4.1 4.1 PIT 96.6 3.1 7.3 1548
CHI 87.0 9.5 DET 6.9 6.0 15.5 1519
MIA 12.7 9.8 IND 22.3 9.7 19.5 1460
BAL 37.4 8.4 OAK 0.0 0.0 8.4 1436
NE 96.9 2.9 NYJ 0.2 0.2 3.1 1480
LAC 83.1 5.7 ARI 0.0 0.0 5.7 1476
PHI 20.3 5.3 NYG 0.3 0.5 5.8 1467
CIN 24.5 9.0 CLE 0.3 0.5 9.4 1408
JAX 0.9 1.0 BUF 0.9 0.7 1.6 1432
TB 0.5 0.3 SF 0.0 0.0 0.4 1397

Game quality is the harmonic mean of the Elo ratings for the two teams in a given matchup.

*Average change is weighted by the likelihood of a win or loss. (Ties are excluded.)

Source: ESPN Stats and Information Group

Dallas already lost to Washington in Week 7 on the road, 20-17. And if they somehow lose to McCoy (career 7-18 as a starter including 1-3 for the Redskins when he last started in 2014), Dallas would be two games back and Washington would control the first tiebreaker with only five games remaining. In that improbable scenario, the Cowboys would find themselves suddenly all-but removed from playoff contention and hurtling instead to an offseason of probable tumult and coaching change.

But according to our projections, Dallas has a 65 percent chance of winning on Thursday and avoiding all that.


FiveThirtyEight vs. the readers

Be sure to check out FiveThirtyEight’s Elo ratings using our NFL prediction interactive. It simulates the rest of the season 100,000 times, helping you see how likely each team is to make the playoffs and win the Super Bowl. And if you want to prove you can pick better than Elo, you can also face off against the algorithm in our prediction game. In addition to bragging rights, you’ll be playing for the chance to climb up our giant leaderboard.

Using data from the prediction game, here are the matchups in which Elo made its best — and worst — predictions against the reader picks last week:

Elo’s dumbest (and smartest) picks of Week 11

Average difference between points won by readers and by Elo in Week 11 matchups in FiveThirtyEight’s NFL prediction game

OUR PREDICTION (ELO) READERS’ PREDICTION
PICK WIN PROB. PICK WIN PROB. Result READERS’ NET PTS
WSH 62% HOU 51% HOU 23, WSH 21 +11.9
ARI 68 ARI 65 OAK 23, ARI 21 +2.4
TEN 55 TEN 51 IND 38, TEN 10 +1.9
LAC 80 LAC 78 DEN 23, LAC 22 +1.3
ATL 62 ATL 60 DAL 22, ATL 19 +0.7
PIT 68 PIT 70 PIT 20, JAX 16 -0.5
NO 74 NO 76 NO 48, PHI 7 -1.0
LAR 54 LAR 53 LAR 54, KC 51 -3.7
CHI 54 CHI 52 CHI 25, MIN 20 -3.9
NYG 52 TB 51 NYG 38, TB 35 -4.4
BAL 67 BAL 60 BAL 24, CIN 21 -6.9
CAR 57 CAR 64 DET 20, CAR 19 -9.9
SEA 67 SEA 56 SEA 27, GB 24 -11.2

Home teams are in bold.

The scoring system is nonlinear, so readers’ average points don’t necessarily match the number of points that would be given to the average reader prediction.

Elo beat the field again in Week 11, picking up 23.3 points on the average reader to give the algorithm its 10th win of the season. The readers were right to think Deshaun Watson and the Texans would beat Washington, but they were dealt blows by the Lions (who beat the Panthers) and Seahawks (who beat the Packers). Readers have now lost 37.7 points on average by picking the Panthers this season, which is the second-most of any team behind the Patriots.28

Congrats, however, are in order to Voris Struck, who led all identified users in Week 11 with 93.4 points, and to the wonderfully named Greg Chili Van Hollebeke, who took the season-long lead with 778.4 points. Thanks to all 12,421 of our players — and if you haven’t played yet, you can make picks now and still try your luck against Elo this week.

Additional reporting by Neil Paine.

Check out our latest NFL predictions.