Injuries could derail Denver
— -- For Sunday's AFC Championship Game, the Denver Broncos will be missing two key defenders who played in the 34-31 loss to New England in Week 12. Linebacker Von Miller tore his ACL in Week 16, while cornerback Chris Harris tore his ACL in last week's win over the Chargers. What kind of impact might the loss of each player have in the rematch? How good exactly have the Broncos been with and without Miller, and which defenders might the Broncos be forced to rely upon in Harris' absence?
The answers to those questions may determine which team pushes through to the Super Bowl.
The good news is that by this point of the season the Broncos should be more than used to playing without Miller. Between his six-game suspension to start the year and the torn ACL, Sunday will be the ninth game the Broncos have played without their star third-year pass rusher, which will match the number of games he actually suited up for this season.
That does not mean the Broncos have played well without Miller. While they took down the Chargers, by Football Outsiders' DVOA ratings ( explained here), Denver's defensive DVOA without Miller is 10.8 percent, which would have ranked 27th over the course of the season. Their actual full-season DVOA was minus-0.2 percent, good for 15th, so it is apparent Miller's presence made a significant difference.
The good news is the Broncos still played excellent run defense without Miller. In the games he missed, they actually had a better run defense DVOA (minus-20.6 percent) than they did with him in the lineup (minus-11.4 percent). That included limiting the Patriots in the first meeting to only 116 yards on 29 handoffs, plus forcing fumbles by both Stevan Ridley and LeGarrette Blount that earned them quick trips to the bench. The Broncos' run defense also played well against the Chargers last week, recording a DVOA of minus-31.2 percent.
It is against the pass where the Broncos have struggled without Miller. With him, they had a pass defense DVOA of minus-9.9 percent, which would have ranked sixth for the season. In the games he has not played, their pass defense DVOA is 28.1 percent, worst in the league.
Given that Miller had 18.5 sacks in 2012, the obvious answer is that without him, the pass rush is lacking. A look at sack data, however, suggests this is not the case. The Broncos' Adjusted Sack Rate, which accounts for opportunities and the quality of opposing offenses, was 6.7 percent with Miller in the lineup and 6.3 percent without him. Given that Adjusted Sack Rates around the league ranged from 5.3 percent to 9.5 percent, that is a virtually indistinguishable change.
The Broncos also hurried the opposing passer slightly less frequently with Miller in the lineup, even though they brought five or more rushers more often with him active. Whatever has caused the Denver pass defense to play worse without Miller seems to be completely in the coverage and how opposing teams attack the Broncos, not in a decline from the pass rush.
Wherever and whenever, Chris Harris was a key part of that Broncos coverage. That was especially true in the areas the Patriots attack frequently.
The aerial attack Tom Brady leads is a horizontal stretch passing offense, attacking all three short areas of the field -- left, middle and right -- while rarely throwing downfield. Of Brady's passes this season, just 18 percent were thrown more than 15 yards downfield in any direction. We saw this in the first Broncos-Patriots game. Only eight of Brady's 50 attempts were downfield shots, while he threw at least 12 passes into each short area of the field.
Over the course of the season, Harris was the Broncos' most-targeted defender on short passes. He was very effective there, recording an excellent Success Rate of 62 percent ( defined here) and only allowing 5.3 yards per play in coverage according to Football Outsiders game charting. He was also the Broncos' only defensive back to be active in all three short areas, with at least 16 targets each listed as left, middle and right.
By comparison, Dominique Rodgers-Cromartie, whose absence from the second half in Week 12 seemed to key New England's comeback, performed nearly as well in coverage on short passes (65 percent Success Rate, 5.4 yards per play), but only had three plays in the regular season in coverage in the middle of the field. The short middle area was the province primarily of linebackers Wesley Woodyard (who was good) and Danny Trevathan (who struggled).
That speaks to the broader question of how the Patriots choose to play the Broncos, and vice versa. The Patriots used three or more wide receivers on only 15 of their 72 offensive snaps against the Colts after doing so 55 percent of the time in the regular season. Against the Broncos, the Patriots used three wide receivers 54 percent of the time, although it is worth noting they played three wide receivers only 47 percent of the time when they came back in the second half and overtime.
In the Week 12 matchup, the Broncos defender in coverage most frequently was Woodyard, whom we have listed in coverage on 12 passes. He did well on passes over the middle, but the Patriots took advantage of him on the outside. Following that game, the Broncos made a lineup change. Woodyard played every snap that game but now plays only in the nickel, with Nate Irving and Paris Lenon getting snaps in the base 4-3.
Neither Irving nor Lenon has performed particularly well in coverage by our metrics, albeit in small, unreliable sample sizes. The Patriots could probably have even more success attacking them in the pass game than they did Woodyard in Week 12. Alternatively, the Patriots could play three wide receivers, force the Broncos to play nickel, and perhaps attack Kayvon Webster, whom Brady exploited after Rodgers-Cromartie went out, or Quentin Jammer, who struggled against the Chargers. Either way, the Patriots should be able to find a favorable matchup to exploit through the air, and they will probably need to, given the quality of Denver's run defense even without Miller.