- Peter Keating, ESPN Senior Writer
As you may have heard by now, this has been quite a year for quarterback play. And as we have admired, analyzed and argued about Peyton Manning's comeback, Colin Kaepernick's takeover and Griffin-Luck-Wilson's rookie exploits -- and as QBs threw for 118,418 yards in the regular season, the most ever -- there hasn't been much talk among fans or media types about defense. Admit it: When you think about the Broncos possibly meeting the Patriots in the AFC title game, do you envision Manning versus Tom Brady, or Chris Harris versus Devin McCourty?
But now that the calendar has turned, our attention should, too, to the other side of the ball, because the quality of a defense can have a huge impact on an opposing quarterback's play in specific matchups. Indeed, looking at QBR reveals there are three reasons D is critical right now. For one thing, while there's clearly a wide range in performance among QBs who are still alive in the playoffs, from Manning (Opponent-Adjusted Total QBR: 82.5) to Joe Flacco (46.8), the spread among defenses is just as wide. New England's defense has a QBR Allowed of 64.9, while Denver's opponents have posted a QBR of just 26.6. That's a huge variation.
For another, good defenses seem to have the ability to clamp down on opposing QBs at this time of year. In studying playoff games from last season and the first round this year, I have found that quarterbacks have slightly lower QBR scores than I would have predicted, given their ratings for the regular season and the opponents they faced. It's a tiny sample size, but just by way of example, last weekend, Robert Griffin III (who of course was hurt), Andrew Luck and Russell Wilson all performed worse than we would have guessed from their numbers and their opposition, and only Flacco exceeded expectations. This is a subject for further research, but Dean Oliver, ESPN's Director of Production Analytics, has suggested there's reason to prefer smart defenses over quarterbacks in winner-take-all scenarios: it's easier for teams to scout a QB than an entire defensive scheme.
Finally, it's time to look across the line of scrimmage because something amazing is happening in Denver, and Manning, as great as he is, shouldn't suck the oxygen out of the mile-high air all by himself. Look again at that QBR allowed by the Broncos: 26.6. That means Denver made its average opponent, all season long, play like Brandon Weeden.
The Broncos gave up just 200 passing yards per game this season, the third-lowest average in the NFL. And they specialize in making the kind of plays that traditional passer rating doesn't count: Denver tied for the league lead with 52 sacks, and limited opponents to converting just 30.6 percent of third downs, by far the lowest rate in the NFL. With Von Miller and Elvis Dumervil both ranking among the top five players in the league in Sack Damage (explained in more detail here), and with Harris and Tony Carter playing even better than Champ Bailey, Denver's defense is playing at a special level.
With all that in mind, let's take a look at the divisional matchups, focusing on the impact defenses could have on QB play. Thanks to ESPN Stats & Information for data and research assistance.
Ravens at Broncos
Joe Flacco (Opponent-Adjusted QBR: 46.8) vs. Denver defense (Opponent-Adjust QBR Allowed: 26.6)
Peyton Manning (82.5) vs. Baltimore defense (50.1)
Keys: When Ray Lewis was out, Baltimore tackled worse and their blitzing collapsed; the Ravens gave up 5 TDs and got no INTs when sending five or more pass-rushers in their last 10 regular-season games, versus zero TDs and 4 INTs in their first five. Lewis will help Baltimore's pass rush and limit the Broncos' yards after contact. But even he can't do it all. With him on the field, the Ravens gave up 7.7 yards per pass attempt, compared with just 6.6 when he wasn't playing. And over his last 10 games, Manning has completed more than 70 percent of his passes, with a QBR of 93.4, against four or fewer rushers. He's likely to pick the Ravens apart whenever he's not blitzed.