Analytics are all the rage right now, which is why we’re going through each game on the Cowboys’ schedule for the 2021 season and using analytics for each team to make a prediction on who wins. In case you missed previous entries in this series, here is where you can find them:
And now, predictions for weeks six through nine!
Week 6 at New England Patriots
After a three-game stretch at home that helped the Cowboys reach a four-game win streak, they travel up north to Foxborough to take on the Patriots, who themselves are looking to bounce back from a poor season after losing their star quarterback. After Tom Brady left for the Buccaneers, New England signed Cam Newton, who ended up missing a few games with COVID-19, while seven key Patriots players opted out before the season. The result was the Patriots missing the playoffs for the first time since 2008, and their first losing record at 7-9 since 2000.
That prompted the Patriots to spend an uncharacteristic amount of money in free agency to ensure that doesn’t happen again. Newton is back at quarterback, although first-round pick Mac Jones could take over under center at any given point. On offense, New England upgraded their pass-catchers with Nelson Agholor, Kendrick Bourne, Hunter Henry, and Jonnu Smith while beefing up the offensive line with Ted Karras, Trent Brown, and retaining David Andrews. On defense, they added Henry Anderson, Raekwon McMillan, Kyle Van Noy, Matt Judon, and Davon Godchaux while also drafting Alabama defensive lineman Christian Barmore.
All in all, this makes the Patriots very hard to predict. We know that the coaching staff can win big, but the distinct dropoff in talent last year proved how hard it is for coaches to win without great players. They’ve just had a massive influx of talent - Agholor, Henry, and Smith all finished in the top 25 in yards/route run - but how it all fits together will largely dictate the outcome. Quarterback play will too, as 2020 Newton had the biggest dropoff in EPA/play between man and zone coverage; against zone, Newton was almost unstoppable, but he was 31st in EPA/play against man. He was also 31st in EPA/play against single-high looks, which should play well into Dan Quinn’s preferences.
Defensively, the Patriots are very similar to the Giants in terms of scheme, so it helps that the Cowboys face these two in consecutive weeks. One big difference is that while New York varies their coverages a bit more, New England runs almost exclusively man coverage; in fact, the only two teams who used man more last year were led by head coaches from Bill Belichick’s tree (Brian Flores and Matt Patricia). It would make sense for Dallas to incorporate a lot of rub routes to take advantage of that in this game, although you don’t have to go back too far to remember the last time the Cowboys hoped Amari Cooper could top Stephon Gilmore.
Another big factor here is that we don’t really know how much of the Patriots’ poor 2020 season was due to the downgrade at quarterback and how much was due to all of the opt-outs. Either way, New England has more talent on the roster this year, and this game is happening on their turf, where Belichick holds an astounding 135-33 career record. Even last year, New England went 5-3 at home, and odds are likely they’ll be better. This should be a very close game for a few reasons, but it’s hard to bet against the Patriots in Foxborough.
Prediction: Cowboys lose, fall to 4-2.
Week 7 BYE
The analytics are very clear on this one: nothing will happen.
Week 8 at Minnesota Vikings
Some teams and coaches in the NFL struggle when coming out of a bye week, and others thrive. Mike McCarthy has traditionally thrived. In his career, McCarthy is 10-4 coming off the bye week, and his tenth win came last year when the Cowboys snapped a four-game losing streak against the Minnesota Vikings. But things are different this time around. For starters, the game will be in Minnesota, and Andy Dalton won’t be starting under center for Dallas. More importantly, both teams are expected to be considerably better this year.
For the Vikings, last year featured a sneaky good offense - eighth in DVOA, sixth in run DVOA, and Kirk Cousins finished seventh in EPA/play and 18th in QBR - but a very bad defense. In fact, it was the first time since 2014 - head coach Mike Zimmer’s first year in Minnesota - that the Vikings defense finished in the bottom half of the league in DVOA (18th). That was for a few reasons. Minnesota embraced a youth movement at cornerback last year, and that didn’t go so well, and on top of that they were decimated by injuries on the front end.
This year, things should be different. Danielle Hunter, Michael Pierce, and Anthony Barr are all healthy, while young cornerbacks Cameron Dantzler and Mackensie Alexander are now flanked by veterans Patrick Peterson and Bashaud Breeland. Minnesota may not immediately jump back to their perennial top five status they enjoyed from 2017 to 2019, but it stands to reason that this defense will be much better.
Which means the Vikings have to hope their offense doesn’t show any significant regression. Statistically speaking, that seems unlikely. They will once again be led by a new offensive coordinator, with Klint Kubiak taking over play-calling for the first time after his father Gary retired following last year. The scheme will be largely the same, which bodes well for the players’ ability to execute. But that also means continuing to be a run-heavy offense, and the odds aren’t good for the Vikings to repeat their title of leading the NFL in run blocking efficiency last year.
Further to the point is the fact that the offensive line finished 18th in pass block win rate last year, which isn’t terrible but also isn’t good. Factor in how poor Cousins fares against pressure - the quarterback ranked third in EPA/play with no pressure and 13th with pressure - and this becomes a big potential Achilles heel, especially when Minnesota’s most significant upgrade to the line came by way of drafting left tackle Christian Darrisaw and guard Wyatt Davis. That’s not to say those two won’t be good, but it’s a rarity when rookie offensive linemen provide a meaningful upgrade in pass protection right away.
Despite all this, though, the Vikings offense gets a favorable matchup against Dallas at home. Cousins is at his best when he can sit in the pocket and read the defense, and Dan Quinn’s scheme isn’t known for its blitzes. It is, however, known for its heavy use of single-high safety looks, and Cousins finished sixth in EPA/play against that look last year. Much like last year, when the Cowboys allowed 28 points to Minnesota but scored 31, this is setting up to be a game where the Cowboys will have to rely on their offense in order to win. With things being so close, you could flip a coin or just side with home-field advantage. I’ll do the latter.
Prediction: Cowboys lose, fall to 4-3
Week 9 vs Denver Broncos
The Broncos are a very difficult team to read. Led by head coach Vic Fangio, one of the most accomplished defensive minds in the NFL right now, Denver has all the makings to field an elite defense. But in each of Fangio’s two years on the job, that side of the ball has been decimated by injuries. In 2019, they lost five starters for the year and went various stretches without other impact players as well. Last year, Von Miller was lost for the year a week before the regular season began, and Denver went on to lose four more starters throughout the season.
On the offensive side of the ball, it’s been a different kind of tragedy, all surrounding the quarterback position. In 2019, they started the year with Joe Flacco under center; he went 2-6 as a starter before a neck injury ended his season. Then it was Brandon Allen, who went 1-2 and prompted the Broncos to debut rookie quarterback Drew Lock. After he ended the year 4-1 as a starter, Lock was anointed the new starting quarterback and face of the future in Denver.
Then last year happened. Lock regressed in every imaginable way. His completion rate dropped from 64.1% to 57.3%, his touchdown rate dropped from 4.5% to 3.6%, his interception rate climbed from 1.9% to 3.4%, and he won just as many games as he did in 2019 despite playing in eight more games. That’s why Denver opted to trade for Teddy Bridgewater in the offseason, and are now having an open quarterback competition that is likely going to continue into the regular season.
Whoever lines up under center for this game will drastically alter how things go. Just based on last year’s numbers for these two as it relates to Dallas, Bridgewater ranked 12th in EPA/play against single-high safety looks while Lock was 29th; similarly, Bridgewater was 18th when facing pressure, while Lock was 32nd. Also noteworthy is the vast difference in the two quarterback’s play styles: Lock averaged 6.1 completed air yards and 9.1 intended air yards with a 19.6% aggressive throw rate, while Bridgewater averaged 5.1 completed air yards and 7.3 intended air yards with a 12.2% aggression rate. In other words, Lock attacks downfield at an inefficient rate while Bridgewater is very good at dink-and-dunking but not much else.
As for the Broncos defense, it’ll look very similar to the Vikings that Dallas faced the week before. That means a lot of split-safety looks with disguised coverages and not much blitzing. If healthy, this defense has the potential to be great. Miller and Bradley Chubb are a great edge rushing duo, while their secondary consisting of Kyle Fuller, Patrick Surtain II, Bryce Callahan, Justin Simmons, and Kareem Jackson is one of the best units on paper. If Fangio can actually field his starting defense for more than couple of games, this unit could be special.
However, it won’t matter much without good quarterback play. Neither Bridgewater or Lock is an especially appealing option. Bridgewater is a safer choice, but Lock has the juice to make big plays when things go right for him. In this specific matchup, neither one offers a significantly better chance at winning. Dan Quinn’s scheme is built to take advantage of more aggressive/reckless quarterbacks and mitigate the effectiveness of dink-and-dunk types. The trends are favoring Dallas here, although the Broncos defense (if healthy) is good enough to keep them in any game with halfway decent offensive production.
Prediction: Cowboys win, improve to 5-3