Writers of Pro Football Prospectus 2008

Most Recent FO Features

HarvinPer09.jpg

» Impact of the NFL's Kickoff Rule Change

After three NFL seasons of kicking off from the 35-yard line, what has been the impact on touchbacks, returns, field position, scoring and injuries? Also, is this rule responsible for a record number of big comebacks?

31 Jan 2013

Super Bowl XLVII Preview

by Aaron Schatz

So, here we are again, with the "mediocre team gets hot in the playoffs" conundrum. On offense and defense, the Baltimore Ravens were essentially just an average team during the regular season. They were very inconsistent on offense (30th in variance), but the very good games and very bad games averaged out at average. It took a superb performance on special teams to pull them to a 10-6 record.

Then the postseason came, and the Ravens played three straight good games to advance to the Super Bowl. Is this improvement, or did the inconsistency roulette wheel just happen to land on black for three straight weeks? A dominating win like the AFC Championship victory over New England makes it seem like the Ravens are "better than ever," but the Ravens played other games that good during the regular season. The AFC Championship game (70% DVOA) doesn't seem out of place when you compare it to Week 1's win over Cincinnati (44-13, 78% DVOA); Week 3's "not really as close as the final score indicated" win over New England (31-30, 71% DVOA); or Week 10's dismantling of the Oakland Raiders (55-20, 68% DVOA).

Of course, if you look at it in a slightly different way, the Ravens are definitely streaking. Baltimore has eight games this year with single-game DVOA over 20%, but four of those games came in the past five weeks (including the 33-14 win over the Giants in Week 16).

On the other side of the ball, we have a team that has been consistently good all year long. San Francisco has ranked in the top four of total DVOA every week this season except after Week 1 (they were 12th) and after Week 10 (they fell to sixth after the tie game against the Rams). They've declined a bit on defense since Justin Smith was injured midway through the Week 15 game against New England, but that's generally been offset by an improvement in passing efficiency on offense.

So, will Sunday be a good day for the Ravens, or a bad day? Or does their recent improvement suggest that the inconsistency is less of an issue, and they have simply put it into a higher gear for the playoffs? We'll look at the matchups between the two teams, but in particular, we want to look at where Baltimore showed weakness during the regular season, and whether those weaknesses have continued to be a problem in the postseason. We also want to look at where the 49ers offense has specifically improved since Colin Kaepernick replaced Alex Smith as the starting quarterback, and where the defense may be having trouble without Justin Smith at full power.

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.

Ravens on Offense
BAL OFF SF DEF
DVOA 3.0% (13) -14.3% (2)
WEI DVOA 8.3% (11) -10.6% (6)
PASS 9.6% (15) -10.3% (6)
RUSH 7.5% (7) -19.7% (2)
RED ZONE -6.1% (19) -4.1% (11)
49ers on Offense
SF OFF BAL DEF
DVOA 17.0% (5) 2.2% (19)
WEI DVOA 21.7% (2) -3.0% (12)
PASS 33.2% (5) 3.4% (13)
RUSH 12.6% (3) 0.9% (26)
RED ZONE 7.0% (11) -10.4% (8)
Special Teams
BAL SF
DVOA 9.0% (1) -1.5% (20)
BAL kickoff +12.4 (3) -0.6 (19)
SF kickoff +13.3 (3) -2.6 (23)
BAL punts +7.4 (10) +0.3 (15)
SF punts +2.5 (9) +13.4 (5)
FG/XP +9.4 (3) -17.8 (32)

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.

WHEN THE RAVENS HAVE THE BALL

There are a lot of storylines going into this game, and some of those storylines even have to do with what's actually going on down on the field. We all know that Baltimore is coming off two upset wins and goes into the Super Bowl with the "momentum." We know that Colin Kaepernick has been super dangerous and the 49ers offense looks great. We know that David Akers can't seem to find the goalposts with a map and a spotlight. With all of these storylines, the struggles of the 49ers defense have gone a bit under the radar. San Francisco played its worst defensive game of the season against Atlanta in the NFC Championship, and it was the fourth straight game where the 49ers' defense has been below-average according to DVOA. (Yes, by "below average" I actually mean a positive rating, and yes, giving up 13 points to the Arizona Cardinals actually qualifies as below average after opponent adjustments.)

Where's the decline coming from? Let's start by looking at the big picture, splitting things right in the middle of the regular season instead of when Smith gets injured in Week 15. Looking before and after the 49ers' bye week, the pass defense didn't declined much, as an overall increase in net yards allowed per pass has been offset by more interceptions and more success preventing third-down conversions. However, the run defense has been worse, although even the San Francisco run defense after the bye week is a little bit better than the NFL average.

San Francisco Defense, Before and After Bye Week
Weeks Run DVOA Yd/Att Pass DVOA Net Y/P INT/G
Weeks 1-8 -29.7% 3.86 -6.7% 5.17 0.6
Weeks 10-20 -8.4% 4.06 -4.3% 5.99 1.0

Now, let's look at the smaller picture. A view of just Weeks 16-20 shows that since Justin Smith's injury, the 49ers are clearly struggling against both the run and the pass. Smith came back to play in the postseason, but he's still out there with a triceps that's 50 percent torn. With the injury, Smith's gone from the second-best 3-4 defensive end in the league to a good, but not great, defensive end. He's having a hard time holding the left guard so that Aldon Smith can rush untouched on a stunt/twist, and he's not blowing up plays in the run game like before. Both pass defense and run defense have declined for the 49ers on all three downs.

San Francisco Defense, Before and After Justin Smith Injury
Weeks Run DVOA Yd/Att Pass DVOA Net Y/P INT/G
Weeks 1-15 -22.8% 3.79 -13.2% 5.24 0.8
Weeks 16-20 4.2% 4.67 22.5% 7.19 1.0

However, the Smith injury may not be the only reason why the 49ers are struggling on defense right now. The 49ers seem to be having trouble sacking the quarterback (only six in the past four games, including zero for Aldon Smith) but they aren't really any different when it comes to pressuring the quarterback. Before the Smith injury, game charting from ESPN Stats & Information says the 49ers had the opposing quarterback under pressure on 12.3 percent of pass plays.* In the past four games, that number is 12.6 percent, which really isn't any different. But the performance of the pass defense has changed both with and without pressure. With pressure (including sacks except for coverage sacks), the 49ers have gone from allowing 1.3 yards per pass to 6.4. Without heavy pressure, they've gone from allowing 5.8 yards per pass to 7.3.

By the way, including both the regular season and the postseason, Joe Flacco had 4.8 net yards per play with heavy pressure (which is pretty good, considering it includes sacks) compared to 6.9 net yards per play without heavy pressure.*

One of the biggest strengths of the San Francisco defense has been the ability to stop opponents from establishing a good down-and-distance situation with gains on first down. The 49ers were the best defense in the league on first downs, and even as their defense has declined over the past month, they have remained better on first downs than they have on second or third downs. During the regular season, the Ravens ranked 14th on first downs, and this is one split that hasn't improved in the postseason. The Ravens' passing game has improved on first down, but the running game has actually declined.

The running game has improved on second and third down, though, and therefore the Ravens' ground game has reversed the decline that plagued it in the second half of the season. Here's a look at some of Baltimore's run stats in the first half and second half of the year, and then in the postseason:

Ravens Running Game, 2012
Weeks DVOA Rk ALY Rk Stuff% Rk
Weeks 1-7 22.5% 2 4.67 2 12% 1
Weeks 9-17 -0.7% 15 4.15 13 19% 15
Weeks 18-20 35/4% -- 4.44 -- 13% --

Overall, the Ravens are better running up the middle or to the right than they are running left, while the 49ers defense has been stronger against runs outside than it has against runs inside. However, it will be interesting to see how often Ray Rice can find his way past Patrick Willis and NaVorro Bowman when he's running up the middle. The 49ers aren't strong against the run because they stuff runners at the line, but because they don't let anyone get away for a big long gain. The 49ers led the league in Second Level yards allowed per carry, and were second in Open Field yards per carry (those stats are explained here). A number of the matchups in this game have changed since the 49ers and Ravens played each other last season, but the Ravens offensive line and 49ers defense are both pretty much the same up the middle except for Ravens rookie Kelechi Osemele at left guard. Rice had just 59 yards on 21 carries in that game. His three runs longer than four yards were left tackle or right tackle, but he had just 2.3 yards per carry going up the middle. You also have to remember that Rice is more shifty than he is strong. Thanks to our partnership with ESPN Stats & Information this season, we have numbers for yards after contact, and Rice averaged only 1.3 yards after contact, fifth worst among backs with at least 200 carries. The 49ers defense allowed a league-low average of 1.2 yards after contact.*

Of course, the inability to get yards after contact doesn't stop you from getting just one or two yards when that's all you need, and the Ravens are much better off running rather than passing on third-and-short. During the regular season, the Ravens converted two-thirds of the time when they ran the ball with 1-2 yards to go, but just one-third of the time if they passed. That split has continued in the playoffs, although with absurdly small sample size (five out of six runs converted, and zero of two passes). The 49ers were a very strong defense on third-and-short, fourth in the NFL, but they were much better against the pass (-31.2% DVOA) than against the run (14.7% DVOA) in those situations.

Baltimore's passing numbers on third-and-short contrast strongly with those when it is third-and-long. Relatively, the Ravens are much better in the latter situation: fourth in DVOA. As I pointed out a couple of weeks ago in the AFC Wild Card preview this doesn't mean the Ravens are actually better off in third-and-long. The Ravens convert 30 percent of the time on third or fourth down with 7+ yards to go. They rank only 28th in DVOA on third-and-short, but even still, convert those situations 55 percent of the time. Still, that strong 49ers defense on first downs doesn't put the Ravens in quite as big a hole as it would other teams.

One of Joe Flacco's weaknesses that we've been pointing out each week in our playoff previews is his difficulty throwing to the left side of the field. This is one weakness that hasn't seen improvement in the postseason. Here's the table we've been running of Joe Flacco passes by direction, but we'll add on the playoffs as well:

Joe Flacco Passes by Direction, 2012 (includes DPI)
  Regular Season Playoffs
Direction Passes C% Yd/Play DVOA Passes C% Yd/Play DVOA
left 183 49% 5.1 -22.1% 27 41% 4.6 -14.1%
middle 116 69% 9.7 57.1% 23 55% 7.0 58.9%
right 241 63% 7.8 30.1% 30 64% 13.3 123.6%

This is going to make things much easier on Tarell Brown, who is usually at right cornerback for the 49ers. The three 49ers cornerbacks come out relatively similar in the charting stats, with Carlos Rogers doing the best. That's a bit of a surprise, because Rogers is the guy who moves inside to play the slot against three-wide sets, and the 49ers were much stronger against starting receivers than they were against "other receivers." Rogers allowed more yards per pass on passes between the numbers, but so do most cornerbacks. These numbers for the 49ers corners include the postseason, with average yards through the air and average YAC:

San Francisco CB Charting Stats, 2012 (includes playoffs)
Player Passes Yds Suc% PYD YAC
Tarell Brown 108 7.3 47% 14.1 2.7
Carlos Rogers 86 6.0 52% 11.0 2.5
Chris Culliver 85 7.6 49% 13.5 2.5

We know that the Ravens like to throw deep, and the Falcons certainly had some success with deep passes in the NFC Championship. During the regular season, however, San Francisco had the league's best DVOA against passes that went over 15 yards through the air.

Although Willis and Bowman's talents in pass coverage make the 49ers a better-than-average team covering tight ends, but they did (relatively) worse when two tight ends were on the field for a pass play, giving up 5.4 yards per pass (15th in the NFL) against 12 personnel. The Ravens ran 12 personnel only 15 percent of the time this season,* even though they have two good tight ends in Dennis Pitta and Ed Dickson. They may want to run it more often in the Super Bowl.

Andy Benoit's Film Room preview suggested that the Ravens should feed Ray Rice with screen passes to beat the 49ers' man coverage. The 49ers' numbers against running back screens this year are interesting. They forced a gain of less than five yards more than half the time, but had a very poor overall average of 7.9 yards allowed because they allowed a couple of huge plays like a 50-yard gain by Joique Bell in Week 2 and a 22-yard gain by Steven Jackson in Week 13. Meanwhile, the Ravens weren't particularly good with screens this year, just 3.8 yards per pass.* Dumpoff passes where Rice can use his elusiveness to turn nothing into something seem to be a better place to use his receiving skills than planned screen passes.

One last note: both the Ravens offense and the 49ers defense got relatively weaker as they got near the goal line, but the splits are interesting here. The Ravens' offense was 28th in red-zone passing DVOA, but fourth running the ball. The 49ers' defense was 26th against the pass in the red zone, but second against the run. So, strength there matches strength, and weakness matches weakness.

WHEN THE 49ERS HAVE THE BALL

Ironically, the Ravens finally made it back to the Super Bowl in a year where their long streak of defensive dominance ended. For 12 straight seasons, from 1999 through 2011, the Baltimore Ravens were in the top six for defensive DVOA. This year, they fell to 19th. Injuries alone can't explain that kind of a drop, but they certainly are one of the main culprits. Pretty much every important player on defense suffered some kind of injury and either missed time or was forced to play at less than 100 percent. We're not done with the AGL numbers quite yet, but we did Baltimore and they had 47.5 AGL on defense, which would have ranked 28th in 2011.

Surprisingly perhaps, DVOA suggests that the Ravens defense was very consistent even with guys going in and out of the lineup all year. They were as consistent on defense as they were inconsistent on offense, and only Detroit had a smaller week-to-week variance.

Once we adjust for opponent strength, we see that a defense that was consistently mediocre defense in the regular season has now played its two best games of the year against Denver and New England (and they were good against Indianapolis too). Correlation isn't causation, but it seems fair to suggest that this improvement is tied to the fact that the Ravens defense is healthier now than it has been since very early in the season. Lardarius Webb and Jameel McClain are on injured reserve and still missing, but otherwise, nobody is listed as worse than Probable on Baltimore's injury report.

The improved health on defense makes it difficult to use any past performance as a guide to how well the Ravens might be able slow down San Francisco's pistol read-option plays. Baltimore faced a similar offense once this year, against Washington in Week 14. They mostly contained Robert Griffin (7 carries, 34 yards) but because they always had to have a defender assigned in case Griffin kept the ball, they couldn't stop Alfred Morris (23 carries, 129 yards). However, that team didn't have Ray Lewis, Terrell Suggs, or Dannell Ellerbe on the field. They also lost Jameel McClain in the third quarter to a spinal contusion, which left them playing Brendon Ayanbadejo and Josh Bynes at inside linebacker.

In an article at Grantland today, Chris Brown suggests that we can still learn from that game how the Ravens will defend the 49ers, even if we can't learn how well. The Ravens don't have the fastest defense, but they do have a very experienced and smart defense. Against the read-option, you're either going to get fooled by the quarterback or you'll be stuck trying to stop the running back with one less defender than usual. Brown believes that the Ravens will try to make up for this by bringing an eighth guy into the box, which seems like a wise decision if stats from the rest of the season are any guide. Pretty much every defense gets better against the run with more men in the box, but the Ravens got comparitively better.* With six in the box, they ranked 15th (4.5 yards per carry). With seven in the box, they ranked 12th (4.1 yards per carry). And with eight in the box, they ranked sixth (2.4 yards per carry). The 49ers faced an eight-man box more often than any other team, and averaged just 3.8 yards per carry, although that's still a little bit above the NFL average. With the standard seven-man box, the 49ers had 5.8 yards per carry. And Frank Gore in particular was good when running against a six-man box, with 6.6 yards per carry, second among backs with at least 100 carries.*

We don't know how they'll do against read option plays, but the Ravens have certainly been better against standard running plays in the postseason. The Ravens' run defense was a miserable 26th in DVOA during the regular season. Things didn't seem so bad on the surface -- 4.10 yards per carry by running backs was still 13th in the league -- because the Ravens were allowing consistent gains but not a lot of big breakaway highlight runs. They stuffed runners only 14 percent of the time, last in the league, and they allowed opposing running backs a Success Rate of 52 percent, also dead last. The Ravens also couldn't stop the run in the most important situations. The Ravens ranked 24th in DVOA run defense in the red zone. The Ravens ranked 28th in DVOA run defense on third down. If we include quarterback sneaks as well as running back carries, the Ravens allowed opponents to convert 76 percent of Power runs, which was 29th in the league.

In the playoffs, the Ravens are allowing just 3.94 yards per carry to running backs. They have improved from 23.7% DVOA to 3.0% DVOA on third down, and from 10.0% DVOA to -57.3% DVOA in red zone, although of course now we're getting into small sample size theater. (They haven't allowed a conversion on nine red-zone carries, but then again, how much do we learn from just nine carries?)

The problem with all this running, and with moving Bernard Pollard down into the box to help against read-option runs, is that it sets things up for the play-action pass. The 49ers ran play action on 26.8 percent of pass plays (sixth in the league) and were very good at it: 8.7 yards per pass with play action compared to 6.4 yards per pass without play action, the fourth-largest gap in the league.* As for the Ravens, wow, were they bad against the play-action pass this season. Their gap in yards allowed between play-action and non-play action was the largest in the league: 5.4 yards per pass without play action, but 8.6 yards per pass with play action. Is this one of the weaknesses that the Ravens have shored up in the postseason? Well, sort of. The Ravens have allowed only 6.6 yards per pass on play-action passes over the last three games. However, their defensive success rate hasn't really changed; actually it has declined slightly, from 51 percent to 49 percent. So the Ravens haven't allowed big 40-yard gains in the postseason, but opponents are still finding it fairly easy to get first downs with play-action passes. (If you are wondering about DVOA, because of the way the spreadsheets are set up I don't have DVOA in the spreadsheet with charting data until I do a bunch of work in the offseason, sorry.)

Based on regular-season numbers, the 49ers have a huge advantage on first and second downs, while the Ravens have the advantage on third downs. However, if we only look at the 49ers with Kaepernick at quarterback and the Ravens in the playoffs, those advantages all seem to go away.

For example, the 49ers on first downs were the best offense in the league, both overall and specifically on passes. The Ravens were 12th against the run on first down, but 21st against the pass. However, this is one of the places where there's been great improvement over the past three games. Baltimore's defensive DVOA against first-down passes has gone from 15.0% in the regular season to -10.9% in the playoffs, and they've gone from an average of 6.7 yards allowed to 5.5.

Second down was also a huge problem for the Ravens during the regular season. They had 16.6% defensive DVOA on second down, which ranked 29th in the league. In the postseason, however, the Ravens have -25.0% defensive DVOA, which would have ranked third during the regular season. The 49ers offense was 10th on second down during the regular season.

The Ravens also were terrible against third-down runs during the regular season. That's where their inability to stuff runners at the line and prevent steady, moderate gains really became a problem. However, they were excellent against passes on third downs, and they've been even better in the postseason, allowing a completion rate below 50 percent with four sacks and four takeaways (two interceptions, two fumbles). The Ravens defense was at its best on third downs when opponents needed to make up a lot of ground, with the best defense in the league on third-and-long...

... but except for the institution of the pistol and the read-option, the biggest difference between the 49ers offense with Alex Smith at quarterback and the 49ers offense with Colin Kaepernick has been passing on third downs, especially third-and-long.

San Francisco ranked only 18th passing on third down during the regular season. But let's split it between games started by Smith and those started by Kaepernick, including the playoffs:

San Francisco Passing on Third/Fourth Down in 2012 (includes playoffs)
Weeks Plays DVOA NetY/P Avg to go C% Suc% Sacks INT PYD
Weeks 1-10 73 -24.6% 4.92 7.9 66.1% 32% 11 3 7.5
Weeks 11-20 76 50.2% 7.33 8.2 56.3% 38% 5 1 8.5

Alex Smith was Captain Checkdown here, whereas Kaepernick has been at least trying to get first downs -- and succeeding more often because of it.

(A table that compared the players rather than weeks would show a smaller gap, because Kaepernick played a couple of third downs before his first start and converted only once in seven tries.)

Third down is probably when the Ravens are most likely to bring the heat, but we would suggest a five-man blitz instead of a six-man big blitz. Including the postseason, Kaepernick gained 8.8 yards per pass play with four pass rushers, 6.6 yards per pass play with five, and 7.6 yards per pass play with six or more. And the Ravens defense this year allowed 6.6 yards per play with four pass rushers, 5.3 yards per play with five, and 6.3 yards per play with six.*

If you look at our numbers for defense vs. types of receivers, you can see that Baltimore has struggled all year against the other team's starting wide receivers. This is in large part because the Ravens' best cornerback this year has been the guy in the slot, former special teams specialist Corey Graham. Our numbers give Graham a 62 percent Success Rate, 11th in the NFL among corners with at least 35 charted targets. He's allowed 5.8 yards per pass, which ranks 16th. On the outside, the Ravens have Cary Williams on the offensive left, and over the last few weeks Jimmy Smith gave way to Chykie Brown on the offensive right. In our previous playoff previews, I've noted each week that Williams had horrendous charting numbers this year: Williams has 43 percent success rate (82nd) and 7.8 yards per pass (59th). Well, whether it is improved performance by Williams, or improved performance by the pass rush in front of him, or just random chance, Williams has been much, much better over the last three weeks. In the playoffs, he has allowed just 4.0 yards per pass with a 55 percent Success Rate.

Chykie Brown also has better charting stats in the playoffs, but he didn't have too many charted targets before the playoffs, so it makes more sense to just present his stats in total: 40 charted passes, 8.0 yards allowed per pass, and a 53 percent Success Rate. He gives up some big passes deep, while Williams' problem was more giving up steady mid-range gains in front of him. Receivers tend to move around depending on where the strong side of the formation is, but that seems to suggest that Michael Crabtree will have more success against Williams while Randy Moss would be better against Brown. If you haven't looked at our wide receiver stats this year, you might be surprised at how well both Crabtree and Moss did. The 49ers throw to Crabtree a lot; only five other teams threw more often to their No. 1 receivers. And while Moss was used more often as a decoy, even after the injury to Mario Manningham, he played well when he was thrown the ball, with 11.8% DVOA.

As for the third receiver... well, the 49ers don't use three receivers as much as most other teams, especially since Manningham was injured. Their preferred personnel package is to go with two tight ends, Vernon Davis and Delanie Walker. The Ravens overall did well defending tight ends this year, but they had serious problems with two-tight end packages. Remember, again, that their best cornerback this year was their nickelback. If the 49ers keep coming out with only two wide receivers, will the Ravens play nickel? Do they think that Corey Graham can cover Vernon Davis? If they come out with just two cornerbacks, does Graham move outside and replace Brown? We have Graham covering Heath Miller a little bit in Week 13, but otherwise the Ravens didn't use him to cover tight ends.

This isn't quite the same as predicting a big game from Davis and/or Walker, because it's not that tight ends have made particularly big plays when opponents put two of them on the field. The whole thing just seems to leave the Ravens defense sort of discombobulated, leading to big plays from receivers and running backs. The 49ers gained only 5.2 yards per play this year with 11 personnel, 27th in the NFL, while the Ravens defense allowed 4.8 yards per play, fourth in the NFL. Switch that to 12 personnel, however, and now the 49ers are leading the league with 7.1 yards per play, while the Ravens are allowing 5.9 yards per play, which ranks 29th. That includes allowing 8.1 yards per pass against 12 personnel, although this is another weakness the Ravens have fixed in the postseason. Over the last three games, they've allowed just 4.9 yards per play against 12 personnel, and just 5.6 yards per play on passes.*

We've written in the postseason about how the return of Ray Lewis has helped the Ravens' run defense but hurt them in covering running backs, but this won't be a big deal because the 49ers don't target their running backs much in the passing game.

One other split that deserves a mention is the Ravens' performance against shotgun compared to non-shotgun plays. I don't think this really has much bearing on the Super Bowl, because defending Colin Kaepernick in the pistol isn't quite the same as defending Tom Brady or Peyton Manning out of shotgun. The split is more weird and noteworthy than it is important for this game. Almost all defenses are better against plays with the quarterback under center than they are against plays with the quarterback in shotgun. Not the Ravens, though. They had the league's biggest gap the other way. Against shotgun, the Ravens allowed -10.0% DVOA (fifth) and 5.2 yards per play (second). Against non-shotgun, the Ravens allowed 9.1% DVOA (30th) and 5.3 yards per play (25th). This probably ties into the issue of Graham being their best corner this year, since shotgun sets are often those with three and four wide receivers, but there's more to it than that and I'm not sure what the "more to it" is. The 49ers actually went the other way, one of the few offenses that was worse from shotgun than it was otherwise. However, that split goes away if you only look at the games with Kaepernick starting and include the playoffs.

SPECIAL TEAMS

Special teams present an interesting dichotomy. On one hand, this is clearly Baltimore's biggest advantage in 2012 performance. On the other hand, special teams are the most inconsistent part of football. Baltimore's strong special teams performance from the regular season didn't stop them from having an abysmal performance covering kicks and punts against Denver in the Divisional round. It's also interesting to note that the two teams were completely switched in 2011. San Francisco had the second-best special teams in the league a year ago, better in every way except punting (where Andy Lee didn't decline in 2012). Baltimore was 30th in special teams, far worse in every single aspect of special teams. If we want to look at the longer view here, the two teams seem a lot more evenly matched.

For both teams, the biggest difference bween 2011 and 2012 is at kicker. Baltimore brought in rookie Justin Tucker, who was probably the best all-around kicker in the league this season. He led the league in gross kickoff value -- the only kicker in the league whose average kickoff went over 70 yards -- and was third in weather-adjusted field goal value, behind only Sebastian Janikowski and fellow rookie Blair Walsh. For San Francisco, of course, veteran David Akers went from good to horrendous on field goals. (He also went from good to average on kickoffs.) At this point, Akers' struggles on field goals have lasted so long that it's pretty hard to write it off as simply the usual inconsistency of field-goal performance. By FO metrics, the 2012 49ers were the fourth-worst team we've tracked on field goals, ahead of only the 1999 Bears, 2003 Jaguars, and 1993 Patriots. Akers was nice enough to then blow a field goal in the NFC Championship to remind us all of how much he was sucking.

Obviously, there's never been a kicker struggling this badly going into the Super Bowl, and 49ers fans are understandably terrified of sending David Akers out there with the game on the line. But there's a surprising historical precedent if Akers comes through when we least expect it. In 2003, Adam Vinatieri was worth -13.6 points in FO numbers after hitting just 25 of 34 field goals during the regular season. He missed two field goals in the first half of the Super Bowl that year, one wide right from just 31 yards and one blocked from 36 yards. Then he hit the 41-yard field goal that won the game with time expiring anyway. Other than Akers and Vinatieri, only two kickers appeared in the Super Bowl since 1991 after a season worth less than -5.0 points by our field-goal metrics: Scott Norwood (1991 Bills, -7.9 points, hit a 21-yarder in the Super Bowl) and Jason Elam (1997 Broncos, -6.9 points, hit a 51-yarder in the Super Bowl).

One other thing to watch for: a fake, especially from the Ravens. Baltimore ran two fake punts and a fake field goal this year, converting all three times. And you may remember that back in Week 10, the Rams clearly saw something in the 49ers punt coverage that had Johnny Hekker audibling to a fake punt not just once but twice -- and converting both times. If the Ravens see the same thing, it may be Sam Koch passing time.

OUTLOOK

So we go back to the beginning, and the question about teams that improve in the playoffs. If were looking at this game solely through the prism of the regular season, the 49ers would be clear favorites. The changes in these two defenses over the last few weeks close that gap. The Ravens are healthier than they've been all season. The 49ers aren't exactly riddled with injuries, but the one injury to Justin Smith has definitely had an effect. Defensively, these teams now seem even. That makes this game very hard to predict, because both teams have been inconsistent on offense, and while the Ravens have a clear advantage on special teams, you can't trust special teams to win you games because those game-changing plays are fairly rare.

However, although both offenses are inconsistent, since Colin Kaepernick became the quarterback, the 49ers certainly have had more good days than they have bad days. (This is particularly true if you consider that they only had negative offensive DVOA in Week 15 because Kaepernick couldn't hold on to the snap in the Foxborough rain, and there won't be rain in the Superdome). San Francisco's offense matches up well against Baltimore by putting an emphasis on the run and using two tight ends to get Corey Graham off the field. They're going to show Baltimore an offensive scheme that the Ravens haven't seen very much of this year -- and no matter how much film they study on it, you know that Jim Harbaugh has one or two things in the game plan that nobody is expecting, things we haven't seen all year or have only seen very rarely, like when he pulled out the inverted veer last week for a 15-yard LaMichael James touchdown. After all the late nights talking shop on phone calls, and sharing ideas in the offseason, can John Harbaugh sniff out whatever that unexpected wrinkle is? My guess is no, and so the 49ers' ability to be flexible and do unexpected things on offense is why I think they are slight favorites to win Super Bowl XLVII.


STATS EXPLAINED

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.

Unless noted, third-down stats include the occasional play on fourth down as well.

Posted by: Aaron Schatz on 31 Jan 2013

21 comments, Last at 10 Feb 2013, 6:22pm by Foolio

Comments

1
by Tom.k (not verified) :: Fri, 02/01/2013 - 1:40am

I never comment on articles but this is an outstanding article. Best super bowl breakdown I have read ever. Great job.

14
by AJ A (not verified) :: Sun, 02/03/2013 - 4:04am

I completely agree.

17
by Macheath. :: Sun, 02/03/2013 - 5:15pm

Without a doubt.

2
by Kal :: Fri, 02/01/2013 - 3:22am

So this sort of begs the question - what is more likely to be the case?

is it more likely that the 49ers are going to play defense like they've been all season or like they have in the playoffs? Is it more likely the Ravens are going to play that way as well? Have you done any regression analysis on prior seasons to see if there's any correlations there?

3
by MC2 :: Fri, 02/01/2013 - 6:59am

Perhaps the Ravens are better against the shotgun because of their problems with play action, which is obviously much more prevalent with the QB under center.

9
by Reinhard (not verified) :: Fri, 02/01/2013 - 2:17pm

That's what I was thinking.
Ray Lewis and Ellerbe are big guys, and they like stuffing the run. But their change of direction is weak... and against the niners they can't sit back against the run action or they'll get trucked.

4
by MC2 :: Fri, 02/01/2013 - 7:01am

Also, I think this sentence is a typo:

That split has continued in the playoffs, although with absurdly small sample size (five out of six passes converted, and zero of two runs).

"Passes" and "runs" should probably be reversed.

7
by Aaron Schatz :: Fri, 02/01/2013 - 1:57pm

Edited.

5
by CBPodge :: Fri, 02/01/2013 - 8:20am

Is there much of a difference to the Ravens' DVOA as a whole when Ray Lewis is playing and when he isn't? Not just to the defense for his performance, but for the whole team? It would be very interesting to see whether the intangible of just having Lewis active has led to the rest of the team overall performing better?

6
by wangdon (not verified) :: Fri, 02/01/2013 - 1:31pm

Best Super Bowl preview I've ever read. One quibble. The Ravens don't have two good tight ends. Ed Dickson is pretty lousy.

8
by Hurt Bones :: Fri, 02/01/2013 - 2:02pm

I think "two good tight ends" is a fair statement. Pitta is certainly better than Dickson, but there are teams in this league that would like Dickson as their starting tight end, and a whole lot of teams who would like him as their no. 2. Not everyone can have Gronk and Hernandez.

10
by Kal :: Fri, 02/01/2013 - 2:45pm

Dickson is also excellent at run blocking; Pitta is not.

11
by Gimlet (not verified) :: Fri, 02/01/2013 - 8:38pm

How did the Seahawks manage to completely contain Kaepernick and the SF offense in Week 16? Was that just the SEA homefield advantage, or was it scheming? Does BAL have the personnel to mimic the SEA plan?

12
by Kal :: Sat, 02/02/2013 - 2:21am

Basically earl Thomas. A way to beat the read option is to bring down both safeties; this leaves you vulnerable to the pass. Seattle only brought one down but Thomas range is so large he could get there fast enough to threaten plays.

The other thing Seattle did that Minnesota also did was simply get an early lead. Gore only carried it 6 times. The 49ers couldn't sustain drives, didn't have any big plays until much later on the game and were basically forced to play catch up as soon as that fg was blocked.

13
by robbbbbb (not verified) :: Sat, 02/02/2013 - 2:16pm

"70% of the universe is covered by dark matter. The rest is covered by ET."

Earl Thomas is the straw that stirs the Seattle defense. He's the one irreplaceable player they have. He's got so much range, and is so talented in so many other ways that he allows the Seahawks defense to do things that other teams can't. He can play centerfield, he can run safety blitzes, and he can come into the box and play the run.

But his main strength is his ability to play the deep defender in cover-1 and allow everyone else on defense to play their responsibilities. So the Seahawks can bring Kam Chancellor into the box and still play press-man on the outside.

15
by armchair journe... :: Sun, 02/03/2013 - 8:12am

Mr. Norwood no doubt appreciates you only mentioning the one he made.

//AJMQB

16
by Aaron Brooks Go... :: Sun, 02/03/2013 - 4:37pm

You're thinking 1990.

1991 was the Redskins 37-24 win, where they lead 24-0 early in the 3rd. It was never close.

19
by armchair journe... :: Sun, 02/03/2013 - 10:46pm

ah. you're right. '91 bowl, but '90 season.

//AJMQB

18
by Foolio (not verified) :: Sun, 02/03/2013 - 9:37pm

Wrong again.

20
by RickD :: Mon, 02/04/2013 - 2:10am

*chortle*

You just don't "get it", do you?

21
by Foolio (not verified) :: Sun, 02/10/2013 - 6:22pm

You're the one that doesn't get it, chortler. Enjoy that Kool-Aid.