There has been some talk, slander, if you will, about the legitimacy of the Dallas Cowboys defense this season. It seems the unit that has collected a league-high 10 interceptions in just five weeks of action is ruffling some feathers around the league.
Some have gone so far as to say the Dallas “Cowboys defense should be no match for Mac Jones” and the Pats this weekend. Headlines such as this is nothing new for the Cowboys after posting some, stop me if you heard this before, historically bad numbers in 2020. Doubters are expected.
But to think an over-drafted rookie is somehow the key to beating the Dallas Cowboys, is bordering on the preposterous. How could Mac Jones do what Justin Herbert (a far more superior QB who’s already approaching “elite status”) could not, is anyone’s guess. But the doubters do bring up some honest concerns regarding the legitimacy of the Dallas Cowboys defense.
Did the Dallas Cowboys defense really improve by leaps and bounds this season or is this Fool’s Gold?
Today we’ll look at the numbers…
The Dallas Cowboys defense is currently ranked 6th in defensive DVOA ratings. That’s a huge improvement over last year were they never climbed out of the 20s. DVOA is a score that compares one unit to the average league unit. It looks at every play, puts it in context of the situation, and scores accordingly. For instance: a four yard gain on 3rd-and-5 is going to score much higher for an offense than an four yard gain on 1st-and-10.
So by this scoring method, Dallas is a top-10 defense because many of the yards against them are either fairly empty yards or when opponents are trying to catch-up.
Looking at the game data, that seems to make sense:
When you remove the fourth quarter, when Dallas has already built a sizable lead, you’ll see an EPA score that’s strikingly similar to Football Outsiders’ DVOA score. With an EPA yielded of -.035 the Dallas Cowboys are sixth in defense.
EPA, or expected points added, is the value gained by each play considering down, distance, and field position. Using historical data it can tell you whether a play increased or decreased your chances of scoring. For instance: a run for three yards on 1st-and 10 from the offense’s 30-yard line is a win for the defense. Because the offense is in worse shape at 2nd-and-7 from the 33 than they were at the original 1st-and-10 spot.
The Dallas Cowboys are technically also ranking 6th if you include all four quarters, but the numbers within the rankings change considerably. All four quarters they give up an average of -.035 per play in EPA with a success rate of 45.2%. But if you remove the fourth quarter like we did in the illustrative tweet above, Dallas averages -.062 in EPA at a success rate against of just 42.7%.
It’s clear teams are piling up the damage when they are in catch-up-mode late in the game, making traditional grading methods like yards against and points given up, a little misleading.
The biggest concern is the Dallas Cowboys’ reliance on turnovers. Because if those dry up there will be a drastic reduction in the scores we’re leaning on. But even though turnovers are relatively unstable, some teams are clearly better at getting them than others.
The Dallas Cowboys have forced two or more turnovers for the past nine games. Their prowess dates back to last season indicating these aren’t just lucky breaks but rather it’s ingrained in their DNA.
The final takeaway is this – be suspicious of this defense. It’s not suddenly elite and it’s leaning heavily on turnovers, a fairly unstable cornerstone stat. But they’ve shown they can create big plays consistently and aren’t just relying on luck. They are deep (over 30 players have played on defense alone this season) and starting-level reinforcements are on the way (DeMarcus Lawrence, Neville Gallimore, Kelvin Joseph).
Given they are so often playing with a lead, and understandably softening up late, a lot of the volume stats should be taken with a grain of salt. DVOA and EPA/SR numbers all indicate the Cowboys are better than outsiders are giving them credit for.
This is a better-than-average defense, plain and simple. And as long as the turnovers keep coming, this is a fringe top-10 defense. It’s shocking when you consider how far they’ve come.