For Dallas Cowboys, quality of opposing QBs could make or break 2018 season

Last year, NFL teams went 100-17 (.855) in games where they scored 30+ points. Teams like the Eagles (9-0), Saints (7-0), Panthers and Vikings (both 5-0), Falcons and Steelers (both 4-0) did not drop a single game when they scored 30+ points.

The Cowboys were 4-2 when scoring 30 points or more, and only the Texans (3-3) lost more games in which they scored 30+ points. If the Cowboys had won every game in which they scored 30+ points, they would have been in the playoffs.

But they didn’t, and they weren’t. In fact, points scored did not correlate at all with wins for the 2018 Cowboys. But points allowed did, as the following table illustrates:

The Cowboys were 4-2 when scoring 30+ points and 5-5 when scoring less than 30 points, an indication that the ability to score did not have as big an impact on the outcome of games as one would expect.

In 2017, the Cowboys W/L record was primarily driven by the defense, not the offense. That may sound like a weird thing to say about a team that has defined itself by its offense for years. But when you look at points allowed, the Cowboys were 0-4 when allowing 30+ points, but 9-3 when allowing less than 30 points. And incidentally, the Cowboys were 9-0 when allowing 20 points or less.

I’ve argued at length and repeatedly that the key to the Cowboys making the playoffs is an improved pass defense, and last season was another textbook example of that.

This next table features last year’s schedule, along with the opposing QB and his passer rating in that game, as well as a W/L indicator. For your convenience, the table is sorted by passer rating in descending order.

If you take a 95 passer rating as the demarcation line between a predicted win (opposing passer rating below 95) and a predicted loss (opposing passer rating of 95 or higher), then the opposing passer rating was a correct predictor in 15 of 16 games, which is a remarkably strong correlation.

Over the full 2017 season, the Cowboys allowed a defensive passer rating of 94.6, the 27th-ranked value in the league, and almost 10 points above the 85.1 league average passer rating. Which means that on average, every QB looked better against the Cowboys than he did for the rest of the season. Broncos QB Trevor Siemian is an obvious case in point. Despite throwing for an atrocious 73.3 passer rating over the entire season, he did manage a game of 113.0 against the Cowboys, his best game of the season in terms of passer rating.

Games like the Broncos game inevitably lead to suggestions that one of the unique qualities of the Cowboys defense is its ability to make ordinary QBs look great. But despite Siemian’s standout performance, that wasn’t really the case last year. The next chart shows which quarterbacks over- or underperformed (as measured by passer rating) against the Cowboys versus their average regular season performance.

Philip Rivers for example had a near-perfect 149.1 passer against the Cowboys last season, 53.1 points above his regular season average of 96.0. And as you look at the other QBs who overperformed against the Cowboys (Siemian, Rodgers, Ryan, Wilson, Palmer), the Cowboys defense last year was less about making average QBs look great, and perhaps more about the inability to stop some of the league’s premier passers.

Interesting data point: In all six division games, the Cowboys defense was able to hold the opposing NFC East QB below his season average passer rating.

But back to our demarcation line at 95.0. The next table is a repeat of the table above, except this time we’re using each QB’s 2017 total regular season passer rating. Like before, the table is sorted by passer rating in descending order.

The season-long passer rating in this table is not quite as accurate a predictor as the in-game passer rating, but still scores impressively as a predictor, getting 13 of 16 games right.

Now that we’ve established the idea that the Cowboys will live and die by their defensive performance, which in turn is likely determined by the quality of the QBs they face, let’s take a gander at the 2018 schedule and the quality of QBs the Cowboys will be facing. Obviously, we have no idea what the 2018 passer ratings of the opposing QBs will look like, but we’ll take their 2017 passer ratings as a proxy (2016 for Andrew Luck).

Based on these 2017 passer ratings, the Cowboys are looking at a tough season. If that 95.0 demarcation line holds true, if all the QBs on the schedule repeat their 2017 performance, and if the Cowboys pass defense turns out to be just as bad as it was last year, then the Cowboys are looking at a 7-9 season in 2018. But there are a lot of ‘ifs’ in this assumption.

The biggest of these “ifs” is probably the QB performance. A lot of the QBs on the list above had a 2017 passer rating that was higher, sometimes quite significantly higher, than their career average passer rating.

Alex Smith, for example, had a career-high passer rating of 104.7 last year in Kansas City. Will he be able to repeat that performance in Washington, or will he revert to his 87.4 career passer rating average? Hard to say, but worth looking at anyway. This next and final table lists the 2018 opposing QBs by their career passer rating.

Going by their career passer ratings, the 2018 schedule is packed with a lot of pretty average QBs, and the Cowboys could be looking at a 13-3 record, and that’s assuming Houston’s Deshaun Watson can repeat his six-game performance from 2017.

Of course, we can’t know for sure what the career trajectories of some of the younger QBs will look like, or what form players like Watson, Wentz, or Luck will be in after their injuries, but you’ve got to like the Cowboys’ chances against a lot of these QBs.

And if you then factor in some kind of improvement for the Cowboys pass defense, you’re looking at a very manageable slate of opposing QBs in 2018.

Then again, if you continue to believe in the Cowboys’ ability to make even ordinary QBs look great, the Cowboys will have plenty of opportunities to do just that next year.