clock menu more-arrow no yes

Filed under:

Real Quarterback Rating Differential: How Can The Cowboys’ Defense Make A Leap Forward In 2017?

New, comments

In our third piece on Real Quarterback Rating Differential, a stat that is second to none in predicting winners and losers in the NFL, we turn to the Cowboys’ defense, which has not ranked well over the last five years. Is there any chance for the Cowboys to change this next season?

NFL: Baltimore Ravens at Dallas Cowboys Matthew Emmons-USA TODAY Sports

In our first article on Real Quarterback Rating Differential, we established that this stat, developed by Cold Hard Football Facts, highly correlates with winning in the NFL. The Real Quarterback Rating, as opposed to the more well-known passer rating, includes all the things a quarterback does with his legs as well as his arm, including rushing yards and rushing TDs, and adds in sacks and fumbles.

The differential ... measures a team’s own quarterback play against how they limit their opponent’s quarterback play. The team that wins this differential has won between 85.5% and 87.5% of the games over the last five years.

Moreover, in the last five years, the ten Super Bowl teams have ranked first or second six times, fourth twice, and twelfth twice. So it pays to finish fourth or higher in this differential.

In our second piece, we looked at Dallas’s 2016 offensive rating, which was third in the NFL, and evaluated whether the Cowboys could expect an improvement next year. The biggest hurdle is that Dallas will be playing a lot more teams that are better at holding down opposing quarterbacks.

In this third article, we’re going to look at Dallas’s defense to see what might be done to improve its 23rd ranking last year. As you can see from the table below, Dallas’s offense has ranked highly twice in the last three years, but its defense continues to rank poorly in this stat.

Team Year ORQR Rank DRQR Rank RQRD Rank
Dallas 2016 97.12 3 85.78 23 11.34 5
Dallas 2015 70.97 31 89.47 22 -18.5 30
Dallas 2014 99.7 2 85.8 23 13.9 5
Dallas 2013 88.14 7 90.88 28 -2.74 20
Dallas 2012 84.14 10 86.35 27 -2.21 16

(ORQR is offensive real quarterback rating, DRQR is the same on defense, and RQRD is the differential.)

Where Are Dallas’s Weaknesses In Defending Quarterbacks?

In looking at the Cowboys’ defense, we are only looking at how they defend opposing quarterbacks and keep those QB ratings down. Here are a few problem areas.

Completion Percentage

This stat is broken out on defensive passer rating, but not given for defensive real quarterback rating because the latter blends in quarterback rushing attempts. But it’s a decent proxy for understanding where Dallas has been weak. In 2016, Dallas allowed opponent quarterbacks to complete 67.14% of passes, 31st in the NFL. In 2015, it was 65.22%, or 25th in rank. In 2015, it was 66.49%, which ranked 30th. The best in the NFL, Denver, allowed a 55.43% completion percentage.

Interceptions and Forced Fumbles

Here’s another area where Dallas has been poor. Nine interceptions in 2016 ranked Dallas tied for 27th. Eight picks in 2015 ranked Dallas tied for 30th. In 2014, Dallas had 18 picks, which tied for 7th. Kansas City, San Diego, and Baltimore led the NFL with 18 picks in 2016.

Quarterback fumbles are added in defensive real quarterback rating, and Dallas did force six of them this year, which is very good. They only forced one in 2015, and two in 2014. Carolina and Oakland forced seven quarterback fumbles last year.

Passing Yards Per Attempt

Dallas did a little better here. but it was still mediocre to poor, ranking 13th in 2016 with 6.91 yards per attempt. Denver was best in the NFL at 5.8 yards per attempt. Dallas was 23rd in 2015, and 20th in 2014, at 7.64 and 7.5 yards per attempt.

Football Outsiders Stats

Football Outsiders creates defensive efficiency ratings, and ranks how teams do against #1, #2, and other wide receivers, tight ends, and running backs (as receivers). They also rank team defenses on the offense’s left, middle, and right, as well as short and deep. How did Dallas do here in 2016?

Dallas ranked 11th against #1 wide receivers, 18th against #2 wide receivers, and 8th against other wide receivers. They were 30th against tight ends, and 18th against running backs as receivers.

On the offense’s left, Dallas ranked 9th, middle 28th, and right 22nd. On deep passes, Dallas was 4th, but 29th on short passes.

This data looks like a classic bend-but-don’t-break defense, with the left side, where Brandon Carr roamed, scoring better than the right side. Had Mo Claiborne stayed healthy, perhaps this ranking would have been more even. In the middle and covering tight ends, Dallas was weakest, which looks like a safety and perhaps a linebacker coverage problem.

Preliminary Conclusion

Looking at this data suggests that Dallas needs to get better players, or see leaps in performance from young players, if it’s going to get much better at holding down opposing quarterbacks. Four of Dallas’s defensive backs are free agents - Brandon Carr, Mo Claiborne, Barry Church, and JJ Wilcox - so there is certainly opportunity for turnover.

A lot of commentary so far this offseason has suggested that Dallas bring back one or more of these defensive backs. Brandon Carr, who never misses a snap, is often mentioned, though he is also contemplating retirement. Barry Church is usually listed as well. Mo Claiborne is good, but too fragile. And J.J. Wilcox hits hard, but has had coverage issues. One wonders if Dallas could do better than all of these players.

To get at the question regarding how Dallas might improve, we are going to look at one team that made a huge leap in Defensive Real Quarterback Rating from 2015 to 2016 - the New York Giants.

The New York Giants Made A Great Leap Forward Last Year; Can Dallas Do The Same Next Year?

To get an idea how a team might jump forward on defense, we looked at the Defensive Quarterback Ratings for all teams in 2015 and 2016. Several teams improved, but the Giants jumped by far the most — from 93.15 in 2015 (28th) to 70.32 in 2016 (2nd). The other teams advancing more than 10 points were Tampa Bay (12.5 point improvement) and Baltimore (10.5 point improvement). Dallas went from 89.47 in 2015 (22nd) to 85.78 in 2016 (23rd).

Looking at the stats above, the Giants reduced opponent’s completion percentage from 66.46% in 2015 to 58.57% in 2016, which ranked third. They reduced QB touchdowns (passing and rushing) from 33 to 15, and overall quarterback turnovers increased from 16 to 19. Passing yards per attempt dropped by almost a yard, from 7.76 to 6.76, sixth in the NFL.

Looking at the Football Outsiders’ stats, the Giants ranked second against #1 wide receivers, fifth against #2s, seventh against other wideouts, third against running backs, but only 26th against tight ends. They were also tenth on the left, second in the middle, and tenth on the right, plus second for deep passes and sixth for short ones. In other words, other than tight ends, they didn’t have any easy weaknesses to exploit.

Most of the difference was personnel driven, as the charts below demonstrate. Here are the defensive backs, by snaps, for 2016 and 2015. The AV column in these charts represents Approximate Value, which is Pro-Football Reference’s attempt to assign a single value number to each NFL player. (Note: all players with fewer than 10% of defensive snaps were not included in these charts.)

Defensive Backs

2016 Player Def Snaps Percent AV 2015 Player Def Snaps Percent AV
21-L.Collins 1105 99.50% 13 21-L.Collins 1093 94.40% 5
20-J.Jenkins 956 86.00% 12 41-D.Rodgers-Cromartie 889 76.80% 9
33-A.Adams 746 67.10% 5 22-B.Meriweather 834 72.00% 4
41-D.Rodgers-Cromartie 733 66.00% 9 20-P.Amukamara 765 66.10% 4
24-E.Apple 701 63.10% 4 43-C.Dahl 429 37.00% 2
31-T.Wade 355 32.00% 2 28-J.Hosley 528 45.60% 2
25-L.Hall 383 34.50% 2 31-T.Wade 529 45.70% 2
29-N.Berhe 164 14.80% 1 38-T.McBride 342 29.50% 2
30-C.Sensabaugh 113 10.20% 1
Totals 49 30

The two biggest differences here are that Landon Collins turned into a monster at safety in his second season, and Janoris Jenkins was a huge upgrade from Prince Amukamara at cornerback. That accounts for 16 of the 19 point gain in Approximate Value from 2015 to 2016.

For contrast, here are Dallas’s 2016 defensive backs.

2016 Player Def Snaps Percent AV
39-B.Carr 1015 96.00% 6
31-B.Jones 985 93.20% 6
30-A.Brown 717 67.80% 4
42-B.Church 675 63.90% 5
32-O.Scandrick 645 61.00% 4
27-J.Wilcox 557 52.70% 2
24-M.Claiborne 406 38.40% 3
38-J.Heath 243 23.00% 1
Total 31

The big difference is that Dallas lacked any high performers, while the Giants had three.

Defensive Linemen

Here is the Giants’ defensive line comparison.

2016 Player Def Snaps Def Snap Pct AV 2015 Player Def Snaps Percent AV
54-O.Vernon 1040 93.60% 15 99-C.Jenkins 733 63.30% 5
95-J.Hankins 764 68.80% 8 72-K.Wynn 579 50.00% 4
90-J.Pierre-Paul 792 71.30% 8 91-R.Ayers 569 49.10% 5
98-D.Harrison 674 60.70% 18 96-J.Bromley 479 41.40% 2
78-R.Okwara 368 33.10% 3 90-J.Pierre-Paul 502 43.40% 4
96-J.Bromley 248 22.30% 2 93-G.Selvie 370 32.00% 2
58-O.Odighizuwa 169 15.20% 1 95-J.Hankins 410 35.40% 3
72-K.Wynn 116 10.40% 1 98-D.Moore 248 21.40% 1
78-M.Kuhn 313 27.00% 3
79-M.Hughes 119 10.30% 0
58-O.Odighizuwa 127 11.00% 0
Totals 56 29

The difference here is more striking than it was in the defensive backfield. Clearly Olivier Vernon and Damon Harrison made a massive difference, as together their AV of 33 outpointed the entire 2015 Giants’ line. They also helped Johnathan Hankins and Jason Pierre-Paul elevate their games.

Here is Dallas’s 2016 line by comparison.

2016 Player Def Snaps Percent AV
96-M.Collins 656 62.10% 7
98-T.Crawford 627 59.30% 7
58-J.Crawford 529 50.00% 6
95-D.Irving 489 46.30% 3
97-T.McClain 471 44.60% 7
93-B.Mayowa 383 36.20% 4
90-D.Lawrence 329 31.10% 2
92-C.Thornton 278 26.30% 2
75-R.Davis 155 14.70% 1
Total 39

Is there any more promise here? Maliek Collins might leap forward in his second year. And one might expect better results from David Irving, DeMarcus Lawrence, Benson Mayowa, and Cedric Thornton. But that might only yield a group of average or slightly above average players, without any real difference makers.

Linebackers

Here is the Giants’ linebacker comparison.

2016 Player Def Snaps Percent AV 2015 Player Def Snaps Percent AV
57-K.Robinson 780 70.20% 4 54-J.Casillas 673 58.10% 4
52-J.Casillas 796 71.60% 8 53-J.Brinkley 420 36.30% 4
59-D.Kennard 533 48.00% 5 59-D.Kennard 487 42.10% 4
91-K.Sheppard 453 40.80% 6 55-J.Thomas 400 34.50% 4
47-U.Unga 433 37.40% 3
52-J.Beason 160 13.80% 1
94-M.Herzlich 132 11.40% 2
Totals 23 22

The Giants’ linebackers rated about the same in both years, though Jonathan Casillas made a big leap forward and Kelvin Shepard was a solid free agent addition.

Here are the Cowboys’ 2016 linebackers.

2016 Player Def Snaps Percent AV
50-S.Lee 977 92.40% 16
59-A.Hitchens 581 55.00% 7
57-D.Wilson 284 26.90% 3
56-J.Durant 282 26.70% 1
Total 27

This is the one area where Dallas outshone the Giants, with Sean Lee demonstrating his outstanding value and All Pro form. If Jaylon Smith can play all out in 2017, this difference should be even more pronounced.

What can we conclude from this comparison?

First, it is possible to improve a club dramatically in free agency, but you may have to spend some money to do it. The Giants added Snacks Harrison, Janoris Jenkins, Olivier Vernon, and Kelvin Sheppard, among others, and all did very well. Compare them to the Cowboys’ budget additions of Cedric Thornton, Benson Mayowa, Ryan Davis, and Justin Durant.

Second, it’s also possible for players to make big leaps forward. Landon Collins made a huge jump from year one to year two. Jonathan Casillas also upped his game, though he’s been in the league for several years.

Third, it would appear that adding a few studs helps elevate the play of teammates, as Johnathan Hankins and Jason Pierre-Paul both doubled their AV numbers after Vernon and Harrison were added.

Fourth, though the stats don’t show it, defenses are interdependent units, with the defensive line helping the secondary by getting pressure on the quarterback, and secondaries helping the front seven by preventing easy pitch-and-catch completions. The Giants made personnel improvements at all three levels, and they worked together to make a huge leap.

What does this mean for the Cowboys going into 2017?

If adding super studs makes a big difference, then Jaylon Smith’s potential return could be a huge boost. That might really help Dallas defend short passes, do better in the middle of the field, and improve tight end coverage.

But Jaylon Smith isn’t going to fix the defensive line. Some improvement could come from what Dallas already has. DeMarcus Lawrence scored a seven AV in 2015, five higher than this year. Dallas’s other linemen might also get better. But there’s no real difference maker on the line yet like there might have been if Randy Gregory hadn’t been suspended and was allowed to develop as a player. Dallas needs one or more of those types of players, either through free agency or the draft.

In the secondary, Dallas could be worse if all four free agents leave or retire. Yet just bringing them back isn’t a recipe for a leap forward. Dallas needs to find an impact cornerback and safety to make a difference, with a keen eye towards increasing interceptions. This would shore up the middle of the field, provide a domino-effect improvement at covering wide receivers, and help lower opposing quarterbacks’ completion rates and touchdowns.

Not every team that’s made a significant improvement in their defensive real quarterback rating has spent money like the Giants did. It is possible that significant improvement could be handled within the Cowboys’ cap. Dallas also doesn’t need to make a 22-point leap on defense like the Giants did to move to the top of the real quarterback rating differential — as the Cowboys finished less than 11 points behind New England for first place.

But Dallas can’t also stand pat or just bring back its free agents if the team wants to get better on defense. That may seem like an obvious conclusion, but it’s helpful to see how it has worked for other teams.

In short, Dallas needs a few more Sean Lees.