Tom and Mike have climbed to the top of the Hill of Nonsense and unleash another wave of insane wagering on the world.
31 Jan 2014
by Aaron Schatz
By now, you've read plenty of articles around the Internet about how this year's Super Bowl is a historic pairing of great offense and great defense. Football Outsiders stats agree, of course. This is the first Super Bowl to ever match the No. 1 team in offensive DVOA (Denver, the sixth-best offense since 1989) with the No. 1 team in defensive DVOA (Seattle, the seventh-best defense since 1989). It is only the third Super Bowl to match the top two teams in total DVOA, and only the fourth Super Bowl to match the top team from each conference in total DVOA.
For those who may be unfamiliar with the Football Outsiders stats, they are explained at the bottom of the page. Scroll down or click this link. Please remember that all stats represent regular season only, except for weighted DVOA and anything else specifically noted. All game charting for these two teams is now complete; any game charting data that appears with a asterisk appears courtesy of the ESPN Stats & Information Group. This preview has two different week-to-week charts for each team, one for offense and one for defense. Because defensive DVOA is opposite of offensive DVOA, the defensive charts are flipped upside-down; thus, the higher dots still represent better games.
All readers can click here for in-game discussion on our message boards. If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.
The first important thing to note is that like the 2007 New England Patriots, the 2013 Denver Broncos don't quite come into the Super Bowl as the juggernaut that we imagine them to be. Four of Denver's top five single-game offensive DVOA ratings came in the first four weeks of the season. This is not in any way supposed to be an argument that the Broncos aren't a great offense. The Broncos have still gone over 40% DVOA in three of their last seven games, and their only game all year with a negative offensive DVOA came with 20 mile-per-hour winds in New England. If we take out the first four weeks of the season, the Broncos still lead the league in offensive DVOA. But they wouldn't be a historically great offense, just the best one of the current season.
The Seahawks are as historically great on defense as the Broncos are on offense, and unlike the Broncos they've played their best football in recent weeks. Seattle's best single-game DVOA this year was in Week 2 against San Francisco, but their four best games since then have all come since Week 13. And as Evan Silva did a good job of pointing out in this Rotoworld column, one of the few defenses to give Denver problems this season was Jacksonville, where Gus Bradley runs the same system as Seattle with dramatically inferior players.
These two units are spectacular in so many different ways, often in ways that counter each other. For example, the Broncos led the league with 51.5% DVOA in the red zone, including 80.3% DVOA on passes. That's very good. Seattle's defense was even better: the Seahawks led the league with -70.5% DVOA in the red zone, including -104.1% DVOA against red-zone passes. Both figures are the best in DVOA history. (The only teams in the same ballpark: the 1993 Oilers, 1999 Cowboys, and 2003 Dolphins)
The Broncos led the league this year with 51.5% DVOA on third or fourth down. They were great on third down both passing (62.7% DVOA, third) and running (31.6% DVOA, eighth). The Seahawks' defense also led the league on third or fourth down, with -55.5% DVOA. That included an incredible -78.9% DVOA against passes on third or fourth down, again the best figure in DVOA history. The Seahawks' defensive rating on third down was tempered by the fact that they were only average against the run, allowing 7.5% DVOA (17th in the NFL) and allowing 70 percent conversions (24th in the NFL) on short-yardage runs.
The splits suggest that the best opportunity to pass on Seattle may be on first down, not in a "passing situation" like third down. Seattle's defense was seventh in the league against the run on first down, while Denver's offense ranked just 18th running the ball on first down. However, the Seahawks ranked only 12th in DVOA against the pass on first downs this season, before improving to the become the best defense in the league on second and third downs. One reason for this change is that their pass rush didn't bring as much force on first downs, with an Adjusted Sack Rate of just 5.0 percent compared to 9.2 percent on second and third downs.
Manning's quick release and ability to disect defenses before the snap make him very difficult to sack, and Denver led the league with 3.7 percent Adjusted Sack Rate this season. However, Manning will go down against a very good pass rush. Manning did not face any of the top five defenses in Adjusted Sack Rate this season, but with the Super Bowl, he will have faced all six of the defenses ranked between sixth and 11th. Somehow Manning made it out of two games against Kansas City (sixth in ASR) without taking a sack, but he took 13 of his 20 sacks (and his only intentional grounding of the season) in the five games he played against the teams ranked between eighth and 11th: New England, Indianapolis, Oakland, and Baltimore. That leaves us Seattle, which ranked seventh.
Most defenses believe that the best strategy against Manning is not to blitz, but to try to pressure with just the front four and drop as many players into coverage as possible. Believe it or not, this has not really been the best strategy since Manning arrived in Denver. In 2012, Manning was awesome against four-man and five-man pass rushes but actually struggled against big blitzes of six or more. This year, Manning was better against those big blitzes, but not quite as good against a five-man rush, gaining 8.3 yards per pass with three or four pass rushers but 7.2 yards per pass against both five-man and six-man blitzes.* (This is Peyton Manning, so "not quite as good" still means "pretty good.")
Seattle doesn't blitz much, although when they do, they bring it home with force. The Seahawks only sent a big blitz of six or more pass rushers on 4.5 percent of pass plays, but they gave up a measily 3.1 yards per play.* However, the key to pressuring Manning may not be how many men the Seahawks bring, but who they bring.
Looking over numbers compiled by ESPN Stats & Information and augmented by Football Outsiders game charters, Manning displays a clear weakness. Given his legendary ability to read a defense and adjust the play at the line, it's a bit of a shocking one. Manning only averaged 5.1 yards per play during the 2013 regular season when opponents blitzed at least one defensive back, which ranked 20th in the league. This was not a one-year fluke, either; in 2012, Manning averaged 5.4 yards per play on DB blitzes, which ranked 24th.*
(It's worth noting that Manning has been a bit better in the playoffs, completing 9-of-10 passes against DB blitzes for 6.9 yards per pass.)
However, defensive back blitzes would require Seattle to really go away from their usual tendencies. The Seahawks sent a DB blitz on just 4.3 percent of pass plays this season, less often than any defense except San Francisco, and they actually allowed 7.2 yards per pass on these 24 plays.* In no game this season did Seattle send more than three DB blitzes. They have only sent two DB blitzes in the playoffs, both in the fourth quarter against New Orleans; one resulted in Drew Brees overthrowing Jimmy Graham, but the other resulted in Richard Sherman giving up a touchdown to Marques Colston on fourth down.
The cornerbacks are almost always in coverage in part because they are among the best in the league. Here are the cornerback charting stats for Seattle; ranks require a minimum of 42 charted passes, with 85 cornerbacks ranked.
|Seattle Cornerback Charting Stats, 2013|
|Cornerback||Tgts||Yd/Pa||Rk||Suc Rate||Rk||Avg PYD||Avg YAC|
Don't let Sherman's seemingly "good but not great" charting numbers confuse you into thinking he's somehow overrated. Cornerback charting stats have a small sample size, so a couple of bad plays (or plays misattributed by our game charters) can move a guy 10 or 15 places in the rankings. "Overrated" is a Pro Bowl player who ranks in the 70s, not in the 20s. (Brandon Flowers, for example, ranked 69th in both Yards per Pass and Sack Rate this season.) And Sherman may have more targets than any other Seahawks corner, but that's because he's been the one constant starter all season. Sixty-seven targets is not a lot; Sherman is currently tied for 40th among cornerbacks, and in reality would rank lower than that because we're still missing a few halves of charting from teams other than Seattle and Denver.
The Seahawks generally play cornerbacks by side, so Sherman will always be on the offensive left and Byron Maxwell will always be on the right, with Walter Thurmond in the slot. (Browner has been suspended indefinitely for positive drug tests.) Don't get fooled by Thurmond's numbers either, as we've learned not to fall in love with nickelbacks who have one year of spectacular charting stats. His great stats also make sense because he never has to cover deep; that's what happens when you are nickelback on a team that plays a lot of Cover-3. Thurmond was one of only two corners this year with at least 40 charted targets whose average pass in coverage came less than nine yards downfield. (Trumaine Johnson of the Rams was the other at 7.8 yards.) However, even though he was covering short all the time, most of these passes were still on the outside, which is why Thurmond can't be held responsible for the hole Seattle had in the short middle of the field.
Yes, as I pointed out in both the Divisional Round preview and the NFC Conference Championship preview, the Seahawks ranked just 13th in DVOA on passes to the short middle. They were even worse by standard stats: 21st with 7.8 yards allowed per pass, and 26th with a 73 percent catch rate allowed. That means we can expect a lot of crossing routes and pick plays, and a lot of Wes Welker. Thurmond on Welker will be an excellent test of how real those great Thurmond charting stats were. It also means that the Broncos should use Knowshon Moreno in the passing game. Moreno had 22 short middle targets, second on the team behind Welker, and had an excellent year as a receiver, finishing third among running backs in DYAR and fourth in DVOA.
The Seahawks led the league by allowing just 4.1 average yards after the catch, so even when Manning completes passes to open receivers in the middle of the field, they are unlikely to rack up big gains in the open field. It helps to have the best free safety in the game.
A few other interesting notes about this matchup:
An analysis of Seattle's offense against Denver's defense is a bit wrapped up in a question we often ask around here: how recent is "recent?" How much time do you need to judge that a team has improved or declined. Two weeks? A month? Two months?
It's true that Russell Wilson has struggled over the past few weeks; Cian Fahey wrote a scouting report on that here. He's played fast, made some poor decisions, and left plays on the field. He's also been less accurate, particularly on passes down the right sideline.
A statistical look at Russell Wilson's slump will look very different depending on whether or not you consider the slump to have started in Seattle's Week 14 loss to San Francisco, as Wilson actually had a pretty good 25.0% passing DVOA that day. Nonetheless, that's where a lot of people seem to believe it started, and that's the first game that Cian Fahey focused on when charting Wilson's passes, so we'll split things up there. Here are Wilson's stats before and after Week 14.
|Russell Wilson, Weeks 1-13 vs. Weeks 14-20|
|Weeks||Games||DVOA||DYAR/G||Yd/Pa||C%||TD/G||INT/G||Sacks/IG per G||Avg Pass D Faced|
As you can see, Wilson's slump is real, but it also isn't quite as bad as it looks from standard stats because a) five of his last six opponents, including the Giants (!), ranked in the top 10 in DVOA pass defense this season and b) at least he isn't suddenly giving the ball away to the other team with a rise in interceptions. He hasn't been getting away with near-interceptions, either; we've only logged three dropped interceptions for Wilson this year, with only one since Week 14 and none in the two playoff games.
(OK, do you want specifics? The three plays were Desmond Trufant in Week 10, Corey White in Week 13, and Patrick Peterson in Week 16. The first two passes are actually listed with the intended receiver "defensing" the near-pick, rather than being outright drops. Peyton Manning also had only three dropped interceptions this season; two in the regular season and one by San Diego in the playoffs.)
Wilson's slump has really only been short-term, but the idea that Denver's defense is peaking right now is based on an even shorter term. It's based on just two games. In fact, if you believe in Football Outsiders numbers, it's really based on just one-and-a-half games, as the Denver defense only controlled the San Diego offense during the first half of the Divisional Round game. Yes, much of Philip Rivers' production came in the fourth quarter, but that wouldn't have really been garbage time if the San Diego defense could have gotten the ball back for Rivers to try a game-tying drive. Those yards given up matter, and they tell us something about the Denver defense. (For example, they tell us that Quentin Jammer has very little left, which the Chargers already knew for themselves.)
At Football Outsiders, we generally believe in looking at the longer-term rather than the short term. Even weighted DVOA, our rating which drops the strength of older games to get a better idea of how good we can expect teams to be now, still gives either 100 percent or 95 percent strength to the last eight weeks of games. So despite Wilson's recent struggles, a look at his entire season is probably a better guide to how well we can expect him to play on Sunday, and the same goes for the Denver defense.
In fact, a more important guide to how this game might go is not how Denver has played in one or two recent games but how much talent they do or do not have on the field. All four of the AFC division champions this year were devasted by injuries and for Denver, those injuries have mostly come on the defensive side of the ball. Over the last few weeks the Broncos have lost five members of their original -- well, original "if not for suspensions" -- starting defense: Derek Wolfe, Kevin Vickerson, Von Miller, Rahim Moore, and finally Chris Harris. The Seahawks' offense, on the other hand, is healthier than it has been all year. Percy Harvin will be on the field for only this third game this season, although he's always an X-factor and nobody ever knows how many snaps he can stay on the field for. Seattle's offensive line is also entirely healthy, and a healthy Russell Okung means that the line is at least one-fifth not horrible.
The injury to Harris leaves Denver with only one trustworthy cornerback, Dominique Rodgers-Cromartie, plus a lot of weakness and a big veteran question mark in Champ Bailey. Here are the Denver charting stats from the regular season:
|Denver Cornerback Charting Stats, 2013|
|Cornerback||Tgts||Yd/Pa||Rk||Suc Rate||Rk||Avg PYD||Avg YAC|
DRC's average of 7.3 yards after catch allowed was third highest among ranked corners, but that average is boosted by two huge plays early in the season where the bigger problem wasn't DRC's coverage but rather blown tackles by Duke Ihenacho: a 73-yard touchdown by Denarius Moore in Week 3 with 52 YAC -- Ihenacho literally collided with DRC to ruin that play -- and a 79-yard completion to Dez Bryant in Week 6 featuring 46 YAC. Without those two plays, DRC drops to just 3.9 average YAC allowed, and without the yards after catch on those two plays, he would drop to 6.1 yards allowed per pass.
The Broncos tended to move DRC around all year to cover the opponent's best (and more importantly, fastest) receiver, so you have to figure that defensive coordinator Jack Del Rio will have him on Percy Harvin whenever Harvin comes into the game. Because otherwise, oy.
It's great to see future Hall of Famer Champ Bailey finally appearing in a Super Bowl, but he had to hobble his way here. A foot injury kept him off the field for all but three regular-season games this season. He played a little bit as the fourth corner against San Diego, but the Harris injury thrust him into the starting lineup against New England. Bailey primarily stayed in short zones against the Patriots. We only charted two passes thrown at Bailey in that game, but that's because there were a number of plays where he passed a receiver on to a safety downfield, and a few plays where Bailey was sort of close but we marked "hole in zone." The one time Bailey was caught one-on-one deep, Aaron Dobson was wide open, but Tom Brady overthrew him on a deep play-action bomb.
Rookie Kayvon Webster was Denver's nickelback for most of the season, but broke his thumb in Week 15. He missed Weeks 16 and 17 and played just one defensive snap against San Diego, then 11 against New England. The only pass we charted against him in the AFC Championship was a screen. If Webster isn't healthy enough to play regular snaps as the nickelback, that leaves Quentin Jammer and Tony Carter. Jammer got lit up during San Diego's fourth-quarter comeback, giving up a 30-yard conversion on third-and-4, a 16-yard touchdown on third-and-4, and a 49-yard conversion on fourth-and-5. The Broncos basically pulled him off the field in the AFC Championship; he played only nine snaps, all on special teams. They replaced him with Tony Carter, who was very good in 2012 but had problems this season. Carter had been a healthy scratch against San Diego but played 33 defensive snaps against New England. Our charting has him giving up four completions on seven passes, all of which moved the chains.
The Broncos got away with this two weeks ago when Tom Brady overshot a couple of receivers open deep. They're going to have to hope that Russell Wilson continues his inaccuracy of recent weeks and has the same problem. It's hard to imagine the Broncos putting DRC on deep-threat fourth receiver Jermaine Kearse, but if I was a Broncos fan, seeing any of these other guys on Kearse would give me a heart attack.
As Vince Verhei wrote for ESPN Insider this week, the strongest element that tied together Denver's worst performances this season was poor pass coverage, particularly on deep throws (defined as passes that travelled 16 or more yards past the line of scrimmage). Overall, the Broncos had an average DVOA against deep throws, allowing a better-than-average 35 percent catch rate but getting too few interceptions and allowing too many third-down conversions. This is a big reason why the Broncos were 29th in defense on third down despite ranking fifth against the run. They were dead last in third-down defense against the pass (40.0% DVOA). That third-down defense has been better recently, but again, "recent improvement" only qualifies if you think two weeks is enough sample size. Denver's third-down defense was actually at its worst in the final six weeks of the regular season, allowing 60.0% DVOA on passes. Denver opponents had an average conversion rate on third and fourth down, but they hit a number of huge plays, making Denver one of just three defenses to give up more than seven yards per pass on third down. The Broncos also haven't gotten a turnover on third down since Week 8.
However, to really pick on the Denver defense, Wilson is going to need to go somewhere he has barely gone this season: the deep middle of the field. The safeties may be a bigger weakness than the non-DRC cornerbacks, especially with Rahim Moore out, and the Broncos allowed 13.7 yards per pass on deep middle passes this season compared to 10.5 yards per pass on deep passes to the outside. As we've noted numerous times during the postseason, Seattle somehow made it through the entire season throwing just two deep middle passes, neither of which was complete -- although they did throw a third against San Francisco in the NFC Championship and Wilson connected with Doug Baldwin for a huge 51-yard gain.
Deep passes are just one half of Seattle's surprisingly old-school offense, which features deep passes and a big dose of Marshawn Lynch. Seattle and San Francisco were the only two teams in the league to run more than they passed the ball this year. Like the Broncos on offense, the Seahawks are also better running the ball to the outside rather than the inside, despite Marshawn Lynch's tackle-breaking abilities. That's the proper strategy against the Denver defense, which ranked No. 1 in the NFL in Adjusted Line Yards against runs up the middle but was just 19th against runs marked left end and 26th against runs marked right end. The Broncos also allowed 4.8 yards per carry against 40 zone read runs this season, which ranked 13th in the NFL. (Half of those runs belonged to the Eagles way back in Week 4; the rest of the year, the Broncos allowed 4.2 yards per carry on zone reads by Kansas City, Oakland, and Washington.)
Seattle will want to run a lot on first down to get Wilson into good down-and-distance situations to break out the deep pass on second down. Denver's run defense ranked 15th on first down this year, but fifth on both second and third down. The Seahawks' running game, on the other hand, got progressively weaker by down, ranking fifth in DVOA on first down but 12th on second down and 17th on third or fourth down. And that ranking of 17th on third or fourth down is a bit deceiving, as many of the positive plays were actually Russell Wilson scrambles, not designed runs. When the Seahawks needed just one or two yards, they struggled to get it. Even with Mr. Beast Mode at running back, the Seahawks were dead last converting short-yardage runs, at just 49 percent. That's just one of the many ways in which the offensive line was the clear weakness of this team. Russell Wilson gets sacked a lot, last in the league with 9.6 percent Adjusted Sack Rate, and there really was no improvement near the end of the season as the original starting offensive linemen returned from injury. The Seahawks were also second in the league in offensive penalties, behind only Oakland, with 22 offensive holding calls and 21 false starts over the course of the regular season.
One way the Broncos improved their pass defense this year was by blitzing. Denver allowed 6.7 yards per pass with four pass rushers, but 6.1 yards per pass with five and 5.3 yards per pass with six or more.* The problem with this is that Russell Wilson absolutely killed the blitz in 2013. Wilson gained 6.7 yards per pass against four pass rushers, but 7.9 against five and 8.2 against six or more.* He also had a league-high 9.8 yards per pass on plays where opponents blitzed a defensive back.* As Vince Verhei pointed out earlier this week on ESPN Insider, when Wilson had trouble against teams like Houston, Arizona, and St. Louis, the pressure generally came from just the front four, and the biggest problems were caused by edge rushers such as Whitney Mercilus and Robert Quinn. That makes Shaun Phillips and Robert Ayers very important players in this game.
Against San Francisco in the NFC Championship, the Seahawks finally covered for the weakness of their offensive line by bringing in Alvin Bailey as a sixth lineman. They used six linemen on 18 plays after doing so by our count only 10 times in the entire regular season. Phillips and Ayers aren't quite Aldon Smith and Ahmad Brooks, but Bailey could play a big role if the Broncos are getting to Wilson early.
Unfortunately, if you don't take Wilson down once you pressure him, you are in serious trouble. Wilson averaged 7.6 yards per play when forced outside the pocket, including sacks and scrambles.* That was second in the NFL behind Ben Roethlisberger, and actually slightly higher than the 7.2 yards per play that Wilson averaged from inside the pocket. The Denver defense was pretty good against quarterbacks outside the pocket, allowing 4.6 yards per play (tenth in the NFL), but of course that number isn't adjusted for opponent and includes a lot of Terrelle Pryor (27 of the 103 out of pocket plays we have listed against Denver) as well as quarterbacks who aren't particularly known for their mobility such as Philip Rivers, Eli Manning, and Joe Flacco.
Other interesting notes about this matchup:
This may be the biggest mismatch in the game, and the one that nobody seems to be talking about. On the surface, it doesn't look like there's a big difference between these two teams, until you consider the way Denver's altitude gives the Broncos' kicking stats a huge boost in home games. Over the past two regular seasons, for example, Matt Prater kicked 95 touchbacks at home (an 86 percent rate) and just 51 on the road (a 54 percent rate). As for Prater's field-goal numbers, there's no doubt that he had a fabulous year in 2013. He not only broke the NFL record with a 64-yard field goal, but also went 9-for-9 on all attempts of 45 or more yards. Even after we adjust for weather and altitude, Prater comes out as the third-most valuable kicker in the league on field goals. However, like most field-goal kickers, Prater has been inconsistent from season to season, and he doesn't have a history of making (or even attempting) long field goals in cold weather. In 2011 and 2012, Prater was just 7-for-14 on attempts of 45 or more yards. Over that entire three-year period, Prater only attempted three field goals of 45 or more yards on the road with a temperature below 60 degrees, hitting one (a 47-yarder with the roof open in Indianapolis this season) and missing two (in Kansas City and Baltimore last season). In none of those games was the temperature below 50 degrees.
The Broncos do have value on returns, at least. We know Trindon Holliday can break a big play, and while the Broncos finally got tired of playing fumble roulette with him on punt returns, he's still dangerous on kickoffs. Eric Decker has taken over punt returns and while it's hard to say whether he's particularly good or bad, he did have a 47-yard punt return in the Divisional Round game against San Diego. Denver's top gunners are Jacob Tamme (nine special teams tackles), David Bruton (seven), and Nate Irving (seven).
Seattle, on the other hand, was in the top 10 in four of the five areas of special teams we measure, with excellent years from kicker Steven Hauschka and punt returner Golden Tate. Although Jon Ryan wasn't particularly good on gross punt distance, the Seattle punt coverage team was excellent; they allowed only four returns over eight yards all season, although three of those were in Week 17 against St. Louis. The one weak spot for Seattle was on kick returns, with a variety of players doing very little, but a healthy Harvin gives them a much stronger weapon to set field position after Denver scores. Seattle's top gunners are Jeremy Lane (12 special teams tackles) and Heath Farwell (10).
The current Vegas line on this game is Denver -2.5. That line properly suggests that this Super Bowl presents an extremely close matchup, but the favorite is just plain wrong. You can't blame the oddsmakers for that, though; they actually opened with the Seahawks as slight favorite, but a lot of money quickly came in on Denver and moved the line. This is one case in which our numbers suggest that the oddsmakers knew better than the public.
A great defense does not necessarily neutralize a great offense, because in general, the best offenses are better than the best defenses. Nonetheless, if the Seattle defense plays at the level it has maintained all season and the Denver offense plays at the level it has shown since Week 5, these two units do basically cancel each other out. Peyton Manning will get some points, he'll move the ball, but he won't do as well as usual and he won't dominate the game. That leaves things up to the other two matchups, and the Seahawks have the edge in both. Their offense is better than Denver's defense, and healthier as well; one strong game against New England doesn't mean the Broncos suddenly won't miss Chris Harris. Special teams should also give the Seahawks a field-position advantage, although that's harder to trust, since special teams can turn the game with a single play and even the best coverage team will give up a big play every so often.
Could this be wrong? Of course. Other than the usual random fluctuation, there are two reasons to believe that Seattle's higher DVOA rating is illusionary. First, if Russell Wilson's struggles from the last few weeks continue. Second, if Seattle's huge home-field advantage means that we are overestimating how good the Seahawks are away from Qwest Field with a neutral crowd.
My guess is that neither of those will be a big issue. Russell Wilson won't be Peyton Manning, but he will play better than he did against New Orleans three weeks ago. The Seahawks won't be quite as dominant as they are at home, but they were the best defense in the league on the road as well, and Denver doesn't get a home-field advantage either. I doubt this game will be like Super Bowl XL where the stands seem to be 80 percent stuffed with fans of the other team.
The Seahawks have now led the league in DVOA for two years in a row. They were better than Denver this year, they didn't decline in the second half of the season, and they are healthier than their opponents. I picked Seattle to beat Denver in the Super Bowl before this season started and I'm sticking with that pick. I'm not looking forward to having to spend the next few months constantly writing about how losing a second close Super Bowl shouldn't damage Peyton Manning's legacy, but I expect I will have to.
DVOA (Defense-adjusted Value Over Average) breaks down each play of the season and compares it to the NFL average based on situation and opponent. You'll find it explained further here. Since DVOA measures ability to score, a negative DVOA indicates a better defense and worse offense, and a positive DVOA indicates a better offense and worse defense.
SPECIAL TEAMS numbers are different; they represent value in points of extra field position gained compared to NFL average. Field goal rating represents points scored compared to average kicker at same distances. All special teams numbers are adjusted by weather and altitude; the total is then translated into DVOA so it can be compared to offense and defense. Those numbers are explained here.
Each team is listed with DVOA for offense and defense, total along with rush and pass, and rank among the 32 teams in parentheses. (If the DVOA values are difficult to understand, it is easy to just look at the ranks.) We also list red zone DVOA and WEIGHTED DVOA (WEI DVOA), which is based on a formula which drops the value of games early in the season to get a better idea of how teams are playing now (explained here).
Each team also gets a chart showing their performance this year, game-by-game, according to total DVOA. There are separate charts for offense and defense. In addition to a line showing each game, another line shows the team's trend for the season, using a rolling average of the last five games.
All third-down stats include the occasional play on fourth down as well.
77 comments, Last at 04 Feb 2014, 11:24am by The Ancient Mariner