SAN FRANCISCO — If there’s one thing years of quarterback data have taught us, it’s that QBs do poorly under pressure.1 In about a quarter of all dropbacks, the QB is under pressure, according to ESPN Stats & Info, and in those cases, their QBR is 7.9.2 This is very bad. It is, to shine a little point of reference, JaMarcus Russell on an off day, and helps reverse engineer a John Madden quote: When a quarterback is being harassed, he doesn’t have much of a chance.3 Great.
As it happens, Cam Newton did not have much of a chance in Super Bowl 50. Stats & Info has him down for 48 dropbacks, of which 21 were pressured. His QBR on those 21 plays was 2.6; for the game, it was 16.9 — just a hair over his 15.6 mark on pressured plays for the season. He was sacked six times, hit 15 times (five on plays that didn’t generate a “pressure”) and blitzed 25 times. He threw an interception, and a few other balls sailed high, and he gave up two fumbles in his own red zone that would result in 15 points for the Broncos. It was, as our MaddenBot might say, a very bad day for Cam Newton.
(Peyton Manning was pressured on 11 of 28 dropbacks, and had a similarly miserable 2.5 QBR on those plays.)
Days like Sunday didn’t happen to Newton very often this season. Through the league championship, he was pressured 179 times in 654 dropbacks, or 27.4 percent. (His QBR on those plays was 15.6.) Going by pressures per dropback, the Panthers played only four top-10 pressure defenses this season. During the regular season, the Texans, Seahawks and Colts each played the Panthers tight and held Cam under 50 QBR (and under 80 in the traditional passer rating), but the Panthers defense and general mediocrity of those teams carried Carolina on through. In the divisional round rematch against the Seahawks, defensive end Cliff Avril — one of Seattle’s best pass rushers — left the game in the second quarter and didn’t return. The Seahawks pressured Cam just twice all game. The Cardinals brought pressure six times in the conference championship, a season-low for them.
|TEAM||DROPBACKS||BLITZES||BLITZ RATE||PRESSURES||PRESSURE RATE|
This may be compelling evidence that Carolina simply stumbled in the big one, having just clubbed the second- and third-best pressure defenses in back-to-back playoff games. But Seattle and Arizona are both hugely different defenses from Denver. The Seahawks are a low-blitz team that relies on a killer secondary. The Cardinals also relied more on coverage before free safety Tyrann Mathieu went down for the year, but they also leaned heavily on blitz packages; their one-on-one pass-rushing was so anemic that 35-year-old Dwight Freeney led the team with eight sacks — and no other Cardinal had more than five.
As you saw Sunday, that is not at all how the Broncos play defense.
To illustrate this, I’ve created a series of charts and graphs. Here’s a simple line chart showing various opponents’ pressure against Newton over the course of the game:
Now here’s a slope chart, with the left column being Newton’s QBR on unmolested attempts, and the right his QBR when pressured:
A histogram showing the distribution of pressure plays by scoring margin:
This one is a stacked bar chart, in two segments:
This is a pie chart:
As the data shows, Newton was well and truly, uh, accounted for.
On the first fumble of the game, the one that set things spinning, Von Miller wheeled around the edge, sized up right tackle Mike Remmers like he was riding down on a quintain, and crashed past him and into Newton, knocking the ball out and into the end zone, all within something like two-and-a-half-Mississippi.
Miller circled back around on Remmers often enough to run up 2.5 sacks, 2 more hits and a half-dozen more near-thumpings. We have stats that can reflect what a player does in a game, but not as much what he does to it. The old Bill James line is that baseball stats can acquire the powers of language, telling a story or calling a game all on their own. Those Broncos numbers are burly enough to carry luggage, but they don’t quite have the powers to tell you how different the game looks when the pocket collapses before the snap finds the center’s ass.
This works the other way around as well. Cam’s numbers aren’t getting any prettier no matter how you slice them. But at least twice Sunday he carried a defender on his back as he wound up and got a throw off toward the sticks — incomplete pass, but not a sack. They were hand-to-God some of the most impressive feats of strength I’ve witnessed. There aren’t stats for that, at least not in the ESPN database, as there aren’t any for something like whispering a third-quarter pass 30 yards in the air, over the shoulder of the defender and into Jerricho Cotchery’s breadbasket at the 5-yard line, only for the defender — a linebacker sprinting 25 yards downfield — to close at the last second and knock it out. It was just Cotchery’s third drop of the season; it also happened to be his third drop of the game. The streaking ‘backer was Miller, naturally, who was named the game’s MVP.
After the game, receiver Devin Funchess was one of the few Panthers to serve up more than platitudes — if just barely — about preparing for Miller and the rest of Denver’s defense. “You can’t do that in practice,” he said, asked about preparing for the rush versus actually having to play against it. “Our guys can’t mimic that.”
It’s true of the stats just as much as it is of the players. Other teams may have pressure numbers that resemble what this Denver defense put together this season, but only because our tools of observation remain imperfect. Near as we can tell, the Broncos are the best defense the Panthers faced in a postseason flush with top units. But for really unusual teams, our best guesses can often be far enough out that we can’t fully appreciate what it is that we’re watching without accepting that greatness affects the game in ways we can’t yet fully capture.
Or, as Madden might tell you just as easily, When you have great players playing great, well, that’s great football.4