Football Outsiders
Innovative Statistics, Intelligent Analysis

Super Bowl XLIII Preview

by Aaron Schatz, with sidebars by Bill Barnwell

This Sunday night in Tampa, the Pittsburgh Steelers will attempt to proclaim total dominance over the NFC West. If the Steelers win Super Bowl XLIII, they will have won three Super Bowl titles over three different NFC West teams: the 1979 Los Angeles Rams, the 2005 Seattle Seahawks, and the 2008 Arizona Cardinals. By doing so, they will take the all-time franchise lead with six Super Bowl titles, breaking a tie with the other NFC West team, the San Francisco 49ers.

Pittsburgh has been one of the favorites for Super Bowl XLIII all season, but a month ago the idea that the Steelers would face the Arizona Cardinals seemed completely ridiculous. As we all know, that's no longer the case. Over the past three games, the Cardinals have seemed like a completely different team compared to the Cardinals that slept through half the season after clinching the NFC West sometime around Rosh Hashanah. Are they really a different team? Let's accept the premise that trends and splits from the entire regular season won't show us how the "real Cardinals" might do against Pittsburgh on Sunday. In that case, we need to answer two questions:

1) Which aspects of Arizona's postseason performance represent significant improvement, and which aspects just show the Cardinals going back to their performance level from the first half of the season, before they clinched their division?

2) What are Arizona's current strengths and weaknesses, if we expect their improved postseason play to continue into the Super Bowl, and how do these match up against the strengths and weaknesses of the Pittsburgh Steelers?

Those are the two questions we're going to try to answer in the rest of this Super Bowl preview.

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. This year, we're going back to the old school for our in-game discussions, so please feel free to use this thread to discuss the Super Bowl before, during, and after the game itself.

If you have FO Premium, you can click here to see all the matchup of DVOA splits for this game.

In an attempt to answer our questions about the Arizona Cardinals, I split the season into four parts:

  • Weeks 1-6: This is when Arizona was playing very well. The Cardinals went into their bye week at 4-2 and sixth overall in the Football Outsiders DVOA ratings.
  • Weeks 8-12: Arizona cooled down a bit, losing close games to good teams (Panthers, Giants) and winning close games against bad teams (49ers, Seahawks).
  • Weeks 13-16: The Cardinals collapse, getting stomped on by Philadelphia, Minnesota, and New England while beating the hapless St. Louis Rams.
  • Weeks 17-20: Arizona gets its act together in the final regular-season game against Seattle, then runs through the NFC playoffs.

The DVOA ratings below are listed with what the rank would be among the 32 teams if this had been Arizona's rating for the entire regular season. I've also taken those week-to-week DVOA charts for Arizona and separated passing and rushing, so you can get a better idea of how the Cardinals played in each facet of the game during different parts of the season. I'm not running the basic stats for the Steelers, because by now we're all pretty familiar with their basic strengths. Unlike Arizona, Pittsburgh has been basically the same team for the entire season. (The Pittsburgh graphs also represent total offense and defense, not split into passing and rushing.)

WHEN THE CARDINALS HAVE THE BALL

Arizona
Pass Offense
Rank Arizona
Run Offense
Rank
Weeks 1-6 30.2% 5 0.0% 18
Weeks 8-12 44.7% 1 -29.0% 32
Weeks 13-16 -12.3% 25 -24.3% 32
Weeks 17-20 50.4% 1 2.3% 14

The Cardinals have won three postseason games in three different ways. They beat Atlanta by playing a strong all-around game. The Carolina game was all about the pass defense, especially with Jake Delhomme throwing a pick roughly once every 15 seconds. As you will see below, the pass defense was actually below-average against Philadelphia, but Kurt Warner and the receivers had their best day of the entire season.

When we look at the development of the Arizona offense over the course of five months, it seems pretty clear that the passing game has never really suffered an extended slump. (This should really surprise nobody.) Arizona's passing DVOA is poor during the Week 13-16 period, but even that is almost entirely based on the snow game in Foxboro where Warner completed just six of 18 pass attempts for a total of 30 yards. (To quote a message I got this week from FO programmer Patrick Laverty: "I gotta think one of the keys to the game for Arizona is to not play in the snow in New England.") It's pretty safe to say that the passing game we're seeing in the playoffs represents "the real Cardinals."

The biggest weapon in that passing game is, of course, Larry Fitzgerald, currently enjoying one of the finest postseason performances by any wide receiver in NFL history. Most of Fitzgerald's highlight plays have been deep bombs, and nobody hit the bomb like the Arizona Cardinals this season. Arizona averaged a league-leading 17.2 yards per pass on deep passes, those that went over 15 yards through the air (not counting passes intentionally thrown away). However, the Pittsburgh defense was the league's best against deep passes, allowing just 8.5 yards per pass. The Steelers were especially good against passes to the deep left, which is where cornerback Ike Taylor is usually stationed. Only six of 25 deep left passes were completed during the regular season by Pittsburgh opponents, for an average of just 6.8 yards. That's generally the side where you will find Larry Fitzgerald. However, Fitzgerald also runs a lot of routes that have him crossing into the deep middle of the field, and Ike Taylor isn't so great covering those types of routes. We have Taylor listed as the defender on six deep middle passes -- including two by San Diego in the Divisional round -- and five of those six were complete for a total of 127 yards. (Note: Obviously, we're talking about a really small sample size here, so judge things accordingly.)

If we want to look at short and deep passes combined, we can see that Pittsburgh generally did a pretty good job of containing opposing number-one receivers, ranking seventh in the NFL in DVOA. Only one of the Steelers' three losses was notable for a strong performance by the other team's best receiver -- when Justin Gage caught five passes for 104 yards in Week 16. DeSean Jackson had just 40 yards when the Eagles beat Pittsburgh in Week 3, and Plaxico Burress had only 15 yards when the Giants beat Pittsburgh in Week 8.

The Steelers also do an excellent job of neutralizing slot receivers, which means they should be able to contain whoever is in the slot, Anquan Boldin or Steve Breaston. Pittsburgh ranked fourth in DVOA against "other wide receivers." According to our game charting, Pittsburgh nickel back William Gay led all NFL cornerbacks (starters or reserves) by allowing just 3.3 yards per pass in coverage, and his 71 percent Success Rate was close to the top of the league. If the Steelers decide to use safety Troy Polamalu more in coverage instead of keeping him in the box, well, it just so happens that Polamalu's game charting stats include 3.7 yards allowed per pass and a 72 percent Success Rate, better in both stats than any other safety listed in coverage on at least 20 charted passes.

The best way to shut down Fitzgerald, Boldin, and Breaston is to make sure that Kurt Warner can't actually get them the ball, and the Steelers are in excellent position to do that. The Cardinals had superb pass protection this season, and Warner actually averaged more yards per pass against five pass rushers (8.2) than he did against four pass rushers (7.1). That's important since Pittsburgh sends five pass rushers on 31 percent of passes, more often than any team except Dallas, but almost never sends more than five. However, the issue here is not how many pass rushers but where they are coming from. Arizona struggled against the zone blitz, with Warner averaging only 4.0 yards per pass during the regular season on plays our charters marked as zone blitzes. The issue is probably not Warner but rather the ability of the offensive linemen to identify where the rush is coming from. Let's be honest, none of these guys are known for being among the NFL's best. Pittsburgh's pass rush doesn't qualify as zone blitz under FO charting standards, because the three linemen rarely drop into coverage, but it works off the same principles -- you know one or two other guys are coming, but you never know where from. When Philadelphia started sending guys from every direction after halftime of the NFC Championship game, the line struggled to protect Warner. The Steelers should be able to harass him if they keep constantly switching the direction where the pressure comes from.

What about the ground game? Well, in reality, the Cardinals haven't been running the ball in the playoffs as well as conventional wisdom seems to believe. Their good running games came against two poor run defenses in Atlanta and Carolina, and while Edgerrin James ran well in the first half of the NFC Championship, the Eagles run defense clamped down in the second half (except when they let Tim Hightower get around them on a fateful fourth-and-1).

Arizona benched Edgerrin James after his dismal seven-carry, 17-yard performance against Carolina in Week 8, but replacement Hightower was actually much worse over the course of the season. Edgerrin James finished with exactly 0.0% rushing DVOA, 21st out of 49 running backs with at least 100 carries. Hightower had -20.6% DVOA, the second-worst figure among backs with at least 100 carries. Take only the games with Edge as the starter, and the Cardinals basically were an average running team, and that's what they've been in the playoffs now that Edge is starting again.

The best way for Arizona to try to gain yardage on the ground would be to run outside on the left. Left end runs would match the strength of the Arizona running game against one of the few weaknesses of the Pittsburgh defense. The three Arizona running backs combined to average 5.1 yards per carry on runs to the outside (left end or right end) but only 3.3 yards per carry on runs listed as middle, tackle, or guard. Yes, most teams gain more yardage on outside runs than they do on inside runs, but the gap for Arizona is triple what the gap is for the average offense. This split has continued into the playoffs, where the Cardinals' backs have gained 120 yards on 20 outside carries (6.0 yards per carry) and 220 yards on 72 other carries (3.1 yards per carry).

As for the Steelers, they ranked seventh overall in Adjusted Line Yards during the regular season, but just 21st against runs around left end and 29th against runs behind left tackle. So far, however, they haven't had a problem in the playoffs, allowing just 10 yards on six carries listed as left end or left tackle.

The other good move for Arizona might be the draw play. It's a favorite strategy of Cardinals offensive coordinator Todd Haley, and our game charters marked Arizona with more draw plays than any team except New England. The Steelers defense, meanwhile, allowed a surprising eight yards per carry on regular-season draws, more than any defense except Kansas City.

On the other hand, there are reasons to believe that draw plays won't do any better than other runs against the Steelers. They've tightened up on draws in the postseason, allowing San Diego and Baltimore just 11 yards on six draws. And while the Cardinals like to run the draw, they aren't particularly good at it. The Cardinals averaged just 3.9 yards on draws during the regular season, much less than the NFL average of 5.5 yards. Even in the playoffs, the Cardinals have averaged 4.1 yards on draws.

A better way for Arizona to take advantage of Pittsburgh's pass rush tendencies might be the screen pass. 50 percent of Arizona's running back screens this season met our definition of success (based on down and distance), one of the best figures in the NFL. Remember, the touchdown that sent them to the Super Bowl was a screen pass to Tim Hightower. While the Steelers were overall the best defense in the league against running backs in the passing game, they were only average against running back screens.

BILL BARNWELL on STOPPING LARRY FITZGERALD

Trying to stop Larry Fitzgerald is a classic case of picking your poison.

Take trying to jam him at the line with a cornerback, which the Eagles have been criticized for not doing. There are two things wrong with that strategy. First, the Cardinals don't run a whole lot of timing routes like a West Coast offense does, so you're not really throwing off their timing unless they're running a pick play. Second, you need a cornerback who is actually capable of successfully jamming the 225-pound Fitzgerald at the line; neither Sheldon Brown (200 pounds) or Asante Samuel (185 pounds) had any prayer of consistently doing that. Brown jammed him on the first play of the game, but Philly stayed off him until late in the third quarter; he was jammed only four times the entire game. The Steelers' corners are all around 190 pounds, so don't expect any of them to be able to jam Fitzgerald consistently.

In addition, Fitzgerald is fantastic at getting through the jam; he has the upper-body strength to simply toss the hands of defenders aside, throwing them off balance and allowing him a free release. If the cornerback doesn't get his hands on Fitzgerald immediately, the former Pittsburgh star is also great at feigning an outside release with his first jab step and then cutting back inside for another free release. If he does go outside, most cornerbacks have to respect the possibility of a deep pass being lofted to him before the safety can get over to help, which opens up a back shoulder fade pattern that the Cardinals love to throw.

If you give Fitzgerald a deep cushion, you're creating a whole new set of problems. Allowing him a free release gives the Cardinals the option of throwing him a quick hitch or a quick slant, both of which have been successful plays for Fitzgerald this year in a limited sample.

Jamming him's only an option if he actually lines up on the line of scrimmage -- while he mostly serves as the Cardinals' split end (or "X" receiver), he'll also move to different sides of the field and even occasionally line up in the slot. One of the Cardinals' core offensive plays involves lining up Fitzgerald in the slot next to Anquan Boldin (if only Todd Haley had told him this!), from which they can run a variety of different pattern combinations. The most common one involves Boldin running a deep post or deep curl pattern, designed to bust open zone coverage while keeping corners off of Fitzgerald, who gets a free release and runs a drag (basically an in) designed to get him the ball with momentum and free space. The Steelers will know this, so expect them to try and use Troy Polamalu as a physical deterrent when the Cardinals send Fitzgerald over the middle.

The solution, of course, is to beat up the guy trying to give you poison. The Eagles' best success against Fitzgerald didn't come through a player or coverage combination; it came from attacking Kurt Warner and preventing him from getting a clear, consistent shot at getting Fitzgerald the ball. The Eagles had five hurries against the Cardinals, all of which came between 1:27 of the second quarter and 4:01 of the third quarter. During that period, Fitzgerald had only one catch for 14 yards, a dumpoff against a Prevent defense that was the lone Cardinals pass attempt without a cornerback matched up against Fitzgerald before the snap. Otherwise, he was kept totally in check. Taylor will likely follow Fitzgerald around the field on Sunday, but it won't really matter; the Steelers' best shot at keeping him quiet will be to cut off his supply.

WHEN THE STEELERS HAVE THE BALL

Arizona
Pass Defense
Rank Arizona
Run Defense
Rank
Weeks 1-6 13.2% 18 -8.5% 8
Weeks 8-12 2.8% 11 -15.4% 6
Weeks 13-16 53.0% 32 21.4% 30
Weeks 17-20  -24.0% 2 -17.9% 4

Like the pass offense, the Arizona run defense has been strong for almost the entire season. The exceptions: two early games, and then those three crushing losses near the end of the season. It is pretty easy to look at this chart and say that Weeks 13-16 did not show the "real" Arizona run defense. Teams like Philadelphia, Minnesota, and New England got out to early leads against the Cardinals and then proceeded to run the ball down their throat, up the gut, and in directions towards other body parts. Sometimes Adrian Wilson was penetrating into the backfield, but more often the defense looked like 11 guys trying to make a play, not a team flowing to the ball as a unit. Gaps were abandoned in an attempt to make big plays that didn't work out, while normally sure-handed tacklers like middle linebacker Gerald Hayes seemed to miss arm tackle after arm tackle when they weren't already blocked. The Cardinals switch back and forth between 3-4 and 4-3 alignments, and in the postseason they've used the 3-4 more frequently. This lets the linebackers get off the line to make more plays, plus the overall tackling has improved.

The real mystery of the Arizona playoff surge is the improved pass defense. If we look at the season split into four segments, the pass defense from Week 17 to Week 20 really stands out. However, if we look at Arizona's week-to-week performance, the postseason just looks like a continuation of the defense's general inconsistency. For most of the year, the Cardinals pass defense was up and down; the exception was five straight terrible games between Week 12 and Week 16. The game against Carolina was the pass defense's best of the season, but the Jake Delhomme turnover festival wasn't that much different from Week 1's J.T. O'Sullivan turnover festival (three fumbles and an interception). More importantly, the Arizona pass defense actually has a DVOA rating above 0% in two of the past four games. (Remember, this means the Cardinals gave up more pass offense than the league average baseline in those games.)

It's hard to look over trends with a sample size of just one or two games, so we combine those two strong performances with the two slightly below-average games against Seattle and Philadelphia. That's still the Cardinals' best pass defense over any four-week period this season. Are they doing anything differently?

Statistically, much of the improvement of the Arizona pass defense comes in the red zone, where they ranked 22nd in DVOA during the regular season. Their numbers for the postseason are excellent (-53.5% DVOA) but that's entirely because of two red-zone interceptions thrown by Jake Delhomme. Donovan McNabb and Matt Ryan still combined to complete 9 out of 11 passes in the red zone. Three of those passes were touchdowns, and three more gave the offense first-and-goal. Changing our period of consideration from the playoffs to "Weeks 17-20" does even more to show how fluky the Carolina game was. Include Week 17, and now the pass defense in the red zone goes from -53.5% DVOA to -8.6% DVOA, only slightly better than average. In the final game of the regular season, Seneca Wallace was 4-for-6 in the red zone with a three-yard touchdown, a five-yard pass that set up that touchdown, and two other conversions that created first-and-goal.

An improved pass rush is certainly part of Arizona's recent defensive surge. The Cardinals had a high Adjusted Sack rate early in the year, then dropped off, but their ASR in the playoffs resembles that from the first six weeks of the season. However, that doesn't necessarily mean they're getting more pressure on the quarterback. Here's a table showing the Arizona pass rush in each segment of the season, with three stats: Adjusted Sack Rate, quarterback hits per pass play, and hurries per pass play. The "pass plays" counted for these last two stats include scrambles but not sacks.


Pass Pressure by Arizona Defense
Adjusted
Sack Rate
QB Hits
per Pass
QB Hurries
per Pass
Weeks 1-6 8.8% 13.6% 17.9%
Weeks 7-12 2.8% 8.9% 14.8%
Weeks 13-16 5.0% 7.2% 12.3%
Weeks 17-20 6.5%* 5.9% 11.8%
*8.3% ASR if we only include the playoffs, not Week 17 vs. Seattle.

The Cardinals were only 23rd in Adjusted Sack Rate during the season, but they were sixth in total quarterback hits (not including sacks). At first, I thought this was partly Arizona official scorer home cooking, but going back and watching some of the games where Arizona had a ton of hits, these hits were for real. So Arizona's pass rush was putting pressure on quarterbacks all season long, even if they didn't necessarily have the sacks in the middle of the year. The flip side of that is the fact that the Cardinals aren't really getting to the quarterback much more than they did during the regular season -- they've just been a little better at taking the quarterback down before he was able to pass the ball. Adjusted Sack Rate is up, but hits and hurries are down.

Even if it has not improved quite as much as people think, the Arizona pass rush seems primed to take advantage of one of Pittsburgh's biggest weaknesses: Ben Roethlisberger's proclivity for getting sacked. Part of the issue is a mediocre offensive line, but there's also Roethlisberger's own desire to scramble and keep plays alive. Then again, Roethlisberger's ability to throw well on the run often makes the decision to hold onto the ball too long look a lot less stupid.

However, just as Arizona's pass rush has changed over the course of the season, so too has Roethlisberger's ability to avoid pressure. In previous playoff previews, I had noted that Roethlisberger took fewer sacks in the second half of the season. It's actually easy to mark the change with Pittsburgh's Week 6 bye, and there actually were two personnel changes that could help explained the improved pass protection. Right guard Kendall Simmons was injured early in the season, and second-year guard Darnell Stapleton replaced him in the lineup as of Week 5, the week before the bye. (Stapleton, who went undrafted out of Rutgers, is yet another example of Pittsburgh's excellent ability to spot talent among rookie free agents.) Then, the week after the bye, veteran left tackle Marvel Smith finally gave up trying to play with a painful back injury, and was replaced with healthy backup Max Starks. Roethlisberger didn't exactly turn into Peyton Manning when it comes to avoiding pressure, but you can see a clear difference in the numbers:


Pass Pressure on Pittsburgh Quarterbacks
Adjusted
Sack Rate
QB Hits
per Pass
QB Hurries
per Pass
Weeks 1-5 13.2% 8.5% 19.1%
Weeks 7-17 7.7% 6.0% 14.6%

There are a few other splits to look at, but they don't tell us much about how Arizona has changed in the last few weeks. The Cardinals' pass defense has improved in pretty much every down-and-distance situation. They've improved against passes to the deep left and short right, which doesn't tell us much because, based on our game charting data, Roderick Hood and Dominique Rodgers-Cromartie don't play specific sides of the field -- both cornerbacks are listed in coverage on passes of every length and direction. (Below, Bill Barnwell delves deeper into the issues surrounding Hood and Rodgers-Cromartie.)

One thing Arizona did relatively well during the season -- defend passes to running backs -- has actually become a problem in the postseason. The Cardinals allowed 125 yards on 15 passes to running backs over the past three games. This is one reason why Karlos Dansby told the NFL Network that the Cardinals are more worried about Mewelde Moore than Willie Parker. Pittsburgh threw only 15 percent of its passes to running backs, less than any team except Denver, but if they throw, they're going to throw to Moore.


Pittsburgh Running Backs as Receivers (includes postseason)
Player DVOA Passes Yards Catch
Rate
First
Downs
21-M.Moore 4.6% 57 341 75% 17
39-W.Parker -81.9% 14 13 36% 0
38-C.Davis -43.7% 10 54 90% 2

If Arizona can hold Pittsburgh to runs and shorter passes on first and second down, it could have a big advantage on third-and-short. During the regular season, the Steelers were horrible in short yardage, and the Cardinals defense was excellent -- except during the late-season period when they were getting slapped around. Pittsburgh ranked last among all 32 teams in offensive DVOA on third and fourth down with 1-3 yards to go. The Arizona defense ranked 18th in the league for the full season, but seventh if we remove Weeks 13-16.

What about the playoffs? Nothing has changed for the Steelers, who converted only three of seven chances on third-and-short, and blew both shots at fourth-and-short (including the fake punt against San Diego). However, this is one place where Arizona's struggles from Weeks 13-16 have continued into the postseason. From Week 17 through the playoffs, Arizona opponents have converted 8-of-11 third-and-short opportunities, plus all three fourth-and-short opportunities.

BILL BARNWELL on THE ARIZONA CORNERBACKS

Roderick Hood and Dominique Rodgers-Cromartie will be the starting cornerbacks for the Cardinals on Sunday. If the Steelers win, they are the most likely Cardinals to play the role of scapegoat.

Hood is a player who's been tricky for us to analyze in the past because his 2007 Game Charting numbers pegged him as one of the league's best cornerbacks. This came despite the fact that he wasn't regarded as a great corner, a data point confirmed by the fact that threw the ball at him 113 times, the sixth-most of any corner.

Having just completed his 16th game, we can say that Hood's numbers took a step down this year. He's been targeted 114 times, but he's averaging 7.5 yards per pass attempt as opposed to 6.3 yards per game a year ago. His success rate has fallen from 56 percent to 52 percent in the process. He was in coverage on a team-high nine touchdowns, including three touchdowns against the Vikings in the Cardinals' humiliating Week 15 loss at home.

The most obvious sign of how other teams regard him, though, is not just how often they target him, but instead how and why they do it. When you watch teams play the Cardinals, they try and maneuver their personnel around to isolate receivers against Hood and then throw at him. The Eagles did this last week with a couple of guys; first, it was Greg Lewis, who they first had run a go pattern that would have been a touchdown if Donovan McNabb hadn't underthrown his pass. The Eagles also ran a quick hitch to Lewis with Hood in coverage, a play stopped by Antrel Rolle. They later got their first touchdown when they motioned out Brent Celek, of all people, one-on-one against Hood and ran a quick slant.

Hood is simply not regarded around the league as a premium corner; in particular, he's known as a poor tackler. The FOX coverage noted before their game against the Vikings that Minnesota head coach Brad Childress -- a member of the Eagles staff when Hood was on Philadelphia -- had every intention of targeting Hood as frequently as possible during the game, both in the air and on run plays. His only run tackle in the game came 32 yards downfield, and Hood was successfully sealed off on the outside by such blocking luminaries as Bobby Wade and Bernard Berrian.

Teams will go after Hood because rookie Dominique Rodgers-Cromartie has shown some signs of brilliance across from him, but the hype about DRC ignores the still-significant weak points in his game. As a pass defender, the elite speed that was his calling card at the combine is still there, and he has good ball skills, but he still has the natural issues with double moves and bigger receivers that all rookie corners have.

Primarily, his problems have to do with his tackling and his play against the run. At the moment, Rodgers-Cromartie's play in run support varies from indifferent to mediocre. Assuming that he actually wants to contribute in this facet of the game, it's clear that he doesn't have confidence in his ability to tackle running backs and take on blockers at the NFL level. Watching him on film throughout the year, he'll almost exclusively go after running backs with ankle tackles and half-hearted dives. He frequently gets frozen when confronted with larger blockers like tight ends and pulling guards, which creates a mismatch on the outside if he ends up against even a gimpy Hines Ward.

Based on our game charting stats, Hood is the guy to target when you want a big play, and Rodgers-Cromartie is a better target if you just need to move the chains. DRC only gave up 5.8 yards per pass this year, and while we have Hood listed as giving up seven plays over 30 yards (including the postseason), DRC has just one, the DeSean Jackson touchdown in the NFC Championship game. On the other hand, DRC has given up 14 first downs or touchdowns on passes that gained less than 10 total yards, while Hood has given up only seven.

The ideal way to attack the poor tackling of the Arizona cornerbacks is the screen pass, a play which the Steelers don't run particularly well. They've only run 32 screens this year, with success rates of only 21 percent on screen passes to running backs and 28 percent to wide receivers. (The league average success rate for those plays are 40 percent and 50 percent, respectively.) Meanwhile, offenses had a success rate of 50 percent on running back screens and 47 percent on wide receiver screens against the Cardinals, along with a 67 percent success rate on plays our charters marked as quick hitches (versus a league average of 43 percent).

SPECIAL TEAMS

Special Teams
PIT ARI
DVOA -1.1% (23) -3.1% (28)
PIT kickoff 8.1 (5) -1.8 (18)
ARI kickoff -10.3 (30) -7.7 (28)
PIT punts 4.4 (13) -6.5 (28)
ARI punts -9.4 (31) -6.6 (26)
FG/XP 0.9 (16) 4.2 (11)

There's a lot of talk about each team's strengths, but one of the most intriguing matchups in the Super Bowl comes where each team is weakest: special teams. When Santonio Holmes returned a punt for six against San Diego three weeks ago, it was Pittsburgh's first special teams return touchdown of the year. Combining kickoff and punt returns, our numbers estimate that poor returns cost the Steelers 19.7 points worth of field position compared to a team with average returns. However, the Arizona coverage teams were also the worst in the league, allowing a combined 22 points worth of field position compared to the NFL average.

Pittsburgh has one particular strength on special teams: kickoffs. Jeff Reed was an average kicker, but the Steelers had excellent kickoff coverage. Pittsburgh was the only team not to allow a kickoff return longer than 40 yards this year.

OUTLOOK

Here we are again, back in the same situation we've been in twice before. This is our sixth Super Bowl preview, and the third time where the Super Bowl matches two teams with dramatically different regular-season performance. I'm left writing the same conclusion I wrote the first two times. AFC team was clearly better than NFC team all year. Two or three strong postseason performances just don't give us enough sample size to be able to say for sure that NFC team will continue its improved play in the Super Bowl. However, we also can't say we have objective evidence that NFC team hasn't actually improved since the regular season. If NFC team plays like it did during most of the year, this will be a blowout. If NFC team plays like it has over the last couple weeks, this will be close. Nonetheless, AFC team has played better than NFC team even if we only look at the playoffs and ignore the regular season entirely. Even the most optimistic view of NFC team's improvement still leaves AFC team as the slight favorite, yet probability is not a guarantee. If NFC team wins, they will claim that "nobody gave us a chance," but only a total idiot would claim NFC team has absolutely no chance to win this game.

Perhaps the Arizona Cardinals will become legendary, the ultimate underdog that finally turned around a franchise-long history of losing. Larry Fitzgerald grew up in Minnesota, so I'm sure he wouldn't mind playing the role of Kirby Puckett, leading an inferior team to victory on the ultimate stage. More likely, Super Bowl XLIII will go down as yet another glorious moment in the long, storied history of the modern NFL's most successful franchise, and the Arizona Cardinals will go down as just another mediocre opponent eventually wiped from the earth by the Terrible Towel of Fate. Hey, nobody remembers the 1979 Rams either.


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.

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). These numbers are all regular season only, except for WEIGHTED DVOA which includes the playoffs.

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 also gets two charts showing their performance this year, game-by-game, according to offensive and defensive DVOA The Pittsburgh charts also include a third-order polynomial trendline showing the development of their performance during the season. The Arizona charts are split into pass and rush, and the trendline is left out to make them easier to read.

Comments

267 comments, Last at 02 Feb 2009, 2:26pm

1 Re: Super Bowl XLIII Preview

"...beating the hapless Rams." Did we watch the same team. The Rams were excellent team this year, they just did not get the good bounces that the Cardinals did. The Rams will return next year to prove that they are the NFC's best in the NFC West.

13 Re: Super Bowl XLIII Preview

really? the rams were excellent this year?!? i guess all those games i watched and they not only lost but got douched completely were just a series of bad bounces.

wow... you, sir, are a moron. cut your losses and pick a new team.

2 Re: Super Bowl XLIII Preview

This is our sixth Super Bowl preview, and the third time where the Super Bowl matches two teams with dramatically different regular-season performance. I'm left writing the same conclusion I wrote the first two times.

For the handy reference of readers, I believe these are the other two times:
http://footballoutsiders.com/game-previews/2008/super-bowl-xlii-preview

http://footballoutsiders.com/dvoa-ratings/2004/super-bowl-xxxviii-preview

7 Re: Super Bowl XLIII Preview

Really? I was thinking 2007 was one, when the Colts D went from turnstile to formidable in the playoffs--just add Bob Sanders!

12 Re: Super Bowl XLIII Preview

No, the cut-and-paste preview refers to the NFC team being the one that showed much improvement in the playoffs. That means it refers to the 2003 Panthers, 2007 Giants, and 2008 Cardinals.

3 Re: Super Bowl XLIII Preview

When you say, 'The Steelers were especially good against passes to the deep left'. Do you mean the qb's left or the defense's left because it's a little confusing. Fitzgerald usually lines up at flanker ie on Warner's right.

6 Re: Super Bowl XLIII Preview

This year’s Steeler compares favorably to two Super Bowl teams in recent memory, the ’95 Cowboys and the ’03 Patriots. Both teams had already won at least one Super Bowl. Like the ’95 Cowboys, the Steelers didn’t have to face the two most dangerous opponents on the road to the Big Game (for Pitt it’s Indy and the Titans, for the Cowboys in ’95 it was Detroit and San Francisco). Dallas just absolutely owned Green Bay in the mid 90’s. Like the 2003 Patriots, the Steelers seemed to win a handful of games decisively, yet by small margins. Thirteen of their sixteen pre-Super Bowl wins were by twelve points or less. What was common about both Super Bowls 30 and 38? They were both surprisingly in-play at the end and were decided because the underdog made a boneheaded, inept mistake (O’Donnell’s brutal pick and Kasay’s kick out of bounds). Expect the same thing on Sunday, Arizona will have a shot in the end until a dumb mistake dooms them. That’s why seven points is too scary a line. But my prediction is simple, at 10:10 pm Sunday night the game will still be in doubt until a crippling mistake wins it for loser Steeler fans.

PS: are steeler fans whining about the amount of strip clubs in Tampa like they whined about Gate D at the Meadowlands?

21 Re: Super Bowl XLIII Preview

This is similar to my opinion:
Arizona passing game keeps them right there. Then in the 4th, tied/behind by 3/etc. Warner has a turnover, probably caused by pressure due to bad blocking. Steelers bleed the clock and score, and that's all she wrote. It is the classic case of "regression to the mean."
The same scenario nearly happened last year, except A. Samuel couldn't come up with Eli's bad pass, and then Tyree made the insane catch. Since probability indicates that "Jetspete" 's scenario is more likely, that's probably how it ends up. As every FO writer and reader knows, last year's SB is an EXTREME outlier. (And I say that having rooted for the G-men during the game!)

26 Re: Super Bowl XLIII Preview

Doesn't "compares favorably to (or with)" mean "is superior to"? Are you saying these Steelers are better than the 03 Pats? Hmmm.

Agree with you about those "dangerous" 95 Lions; they were only 51-7 behind midway through the third quarter of the wildcard game. Dallas dodged a bullet there.

And how can "loser fans" win? Does that make Jets supporters winner fans who lose?

199 Re: Super Bowl XLIII Preview

A few points

- the '95 Lions won their last seven games to make the playoffs. They were the quintessential team nobody wanted to face, their putrid performance in philly not withstanding.

- I'm not saying pitt was better than the '03 patriots. I'm just saying that in 2003 the patriots won a lot of games by small margins, but if you watch those games it always looked as if they were in complete control. the margins rarely told the whole story of the difference between the pats and their opponent

- Steeler fans are losers because they whined to the new york times about Gate D. plain and simple. they root for a winning franchise, but that does not change the fact they are sissies who were sore losers that day last November.

8 Re: Super Bowl XLIII Preview

Why is everyone ignoring Anquan Boldin? This is ridiculous. I know that Fitzgerald is having an insane playoff run but that doesn't mean we should ignore everything that's come prior to that.
Fitzgerald Reg. season stats (per game): 6.0 catches, 89.4 yds, 0.75 TD
Boldin Reg. season stats (per game): 7.4 catches, 86.5 yds, 0.92 TD

And that's with Boldin having to recover from a horrific injury as he went... over their respective careers Boldin has more catches per game and more rec. yards per game. Like I said, Fitzgerald had been outstanding in the last two games, but Boldin is an excellent player as well... I'm not surprised that the mainstream media has latched onto Fitz as the Second Coming, but I'm shocked to see FO mention him only once in passing in this preview.

19 Re: Super Bowl XLIII Preview

That was my big question with this preview as well. Ok, the Steelers might be one of the best teams in the league at shutting down #3 WRs, but Q is arguably still Arizona's best WR. I'm having a hard time thinking of two WRs who are better with the ball in their hands, and you can't write that of quite so easily.

9 Re: Super Bowl XLIII Preview

Aaron - Thanks for providing a thorough, rational and neutral preview. Your recent missives (mostly in the Audibles and Simmons' podcast) were rather bitter sounding, like the Cardinals had been kicking your dog. The above analysis is what I love about FO.

Bill - Props to you as well, I am looking forward to see how the passing matchups play out for both teams.

Go Cards! (I am a diehard Broncos fan that lives in the desert. I still have a hard time believing that I enjoyed the Cardinal's season more than the Broncos!)

14 Re: Super Bowl XLIII Preview

I agree. I know I was critical of Aaron's performance on the Simmons podcast, but this preview was very, very insightful and professional. I'm actually really looking forward to the game.

15 Re: Super Bowl XLIII Preview

I took Aaron and others' writing to be more about stats in general than the Cardinals kicking their dog. FO does a great job, but there are limits to the predictive value of past performance and even in our ability to understand past performance. Are the Cards in the throes of random variation or have they improved? How much can we say about this with such a limited sample size?

These questions are wicked hard, and the Cards brought them to the forefront. So I took the last few postseason writeups more as "grrr stats" than "grrr Cards."

17 Re: Super Bowl XLIII Preview

Reporter: "Will the Steelers legacy be tainted by the fact that their two most recent Super Bowl victories came against teams from the perennially weak NFC West?"

Matt Hasselbeck: "Good grief! C'mon Snoopy, let's get out of here."

20 Re: Super Bowl XLIII Preview

Actually, there's another team in that all-time Super Bowl win tie: Dallas. Go Cardinals!

-D.

25 Re: Super Bowl XLIII Preview

While it's interesting that the Steelers gave up an average of 8 yards per carry on draw plays, this is one situation where FO's advanced stats would have been a lot more useful than conventional stats. If most of these draw plays were on third and >10, then so what if the offense averaged 8 yards per play; conversely, if they were largely in situations with less than 8 yards to go, it's significant.

I'd be more interested in seeing what the opponents' success rate was on draw plays. The Premium stats show that PIT had a DVOA of -19.5% (ranked third) on 3rd/4th down rushing plays, so I'd guess that even giving up 8 YPC on draws, there were a lot of 11-yard gains on 3-18 in there.

28 Re: Super Bowl XLIII Preview

The 8 yard average on draw plays is a little skewed as they got burned against the Patriots and Cowboys (back to back weeks) on draws that went for long yardage. Especially in the New England game (towards the end of the 1st half)....they got hit for 2 draws for (I think) 40-50 yards with their Dime D on the field which didn't have more than two D-Lineman. They got gashed on draws by Kevin Faulk and Tashard Choice in back to back weeks but I only remember it being an issue in those 2 games (I watched all of them). Actually in the NE game they shut the same play down a couple of times in the second half for little or no gain.

29 Sincere question about pass defense strategy

I came up with this when the Colts played the Pats last year in the (cue Wagnerian music) GAME OF THE CENTURY (at least until the next GOTC). I wondered if they would be better off putting a fast 225 lb LB across from Moss on the LOS to jam him. The logic went: Colts have small DBs and Randy is not small. The only way to stop him, aside from mobbing Brady, is to stop him before he gets going. And if the jam turns into a pancake, so much the better. (forget about Welker for the time being)

Could this work tomorrow? Could Pittsburgh put someone larger on Fitz only for the initial bump? Let a DB take him the rest of the way, and the LB then drops into short coverage in the flat, or rushes to the ball if it's a run...?

IMO it sends a great message--"we will hammer you guys." And might make AZ burn a TO right away (always a bonus). They might change their game plan (another bonus). It can be a fake, too, after the first slam. But it's worth trying, no?

...frustrated D coordinator.....

30 Re: Super Bowl XLIII Preview

These teams met last year in week four. Pittsburgh came in undefeated, and lost. This was when the Steelers were still a part of the 'best dvoa ever' watch, I believe. Polamalu and Boldin missed the game, and one of the Steelers d-linemen was out. As far as Xs and Os, I think this actually comes out as a fairly intriguing matchup. Pittsburgh's offense has looked downright pedestrian at times this year, but a lot of that has to do with just how brutal their schedule has been. They played just about every elite defense aside from their own. 12-4 against that difficult a schedule is very impressive. On the other hand, this Cards passing attack is on the verge of 'historically great'. I mean, Warner had a shot at Marino's passsing yards record, and in the playoffs Boldin and Breaston have been afterthoughts. I think its 60/40 that Pittsburgh wins, but it should be a fun game. I didn't see much of the Cardinals in the regular season, but, in the playoffs, both teams have been exceptionally well-coached. I'm far more impressed by Tomlin than I ever was with Cowher

/Cowboys fan, these two teams ruined our season

31 Re: Super Bowl XLIII Preview

It seems to me that the "playoff lottery" angle - the idea that regular season performance is somehow becoming less predictive off post-season success based on the exploits of the Colts, Giants and Cardinals - is far broader than the evidence warrants. Those three teams have something in common: during the regular season, all three were elite at one of the two most important things a football team does - throing the football (Colts, Cardinals)and rushing the passer (Giants). When you're really good at something this vital, it doesn't necessarily take an outlandish up-tick in other areas to turn you into an elite team. It's not unlike an intra-season version of the Colts' "spend all our money on an elite offense and rely on the general randomness of defensive performance to give us some great years" principal.

32 Re: Super Bowl XLIII Preview

I agree, and I hope we see that continue on Sunday. At the same time, New England last year and Pittsburgh this year also excell at one of those elements, too.

38 Re: Super Bowl XLIII Preview

I don't think anything has changed in the past few years. Playoff history is littered with wtf outcomes. If the Steelers win, the narrative will be 'they ran one of the toughest gauntlets in nfl history and came away champions'. If the Cardinals win, it'll be another crazy outcome in a decade full of them. Since John Elway retired, the only two 'chalk' Super Bowls were Oakland-Tampa Bay and New England-Philly, and they were both dull.
In the podcast, Aaron lamented about how if every year was a Villanova-Georgetown, it would stop being special. In the 80s, almost every year was Villanova-Georgetown, and it was the best decade the tournament ever had. If you aren't from the northeast, NC State-Houston was as big of an upset, and Kansas-Oklahoma and Louisville-Duke were close. I think things not going according to form is the most likely outcome of any tournament, and is part of their appeal. The alternative would be 'predictable' after all

39 UK TV Options

For those in the UK the choices are the BBC with a non-descript but non-irritating host plus Rod Woodson, the peerless Mike Carlson and no ad breaks or Sky with Kevin Cadle, Jerry Rice, the Buffoon Halling and five million commercials. Much as I liked Jerry Rice as a player, in what kind of world does anyone choose to watch Sky? Even allowing for the fact that Sky is in HD doesn't come close to compensating for the overwhelming downside of Halling.

41 Re: UK TV Options

In reply to by Paul-London UK

Two questions/comments:

1) You get THREE different choices with three different crews with whom to watch the game? I know nothing about British TV, but is that typical?

2) In the non-commercial broadcast, what do they show when the commercials are on?

42 Re: UK TV Options

Two choices....the BBC which is terrestrial TV and Sky which is via satellite dish or cable. Sky is payable on monthly subscription, the cost dependent on the package of channels chosen. Occasionally Sky and one of the terestrial channels will show the same event iut rarely.

When the commercials are on, the BBC team will analyse plays or discuss tactics etc.

172 Re: UK TV Options

In reply to by Paul-London UK

REALLY impressed with the BBC's coverage of the game - comparable even to my beloved Channel 5. (You may notice the Mike Carlson correlation/possibly even causation - I actually voted for him as my "local" writer in the FO awards, as he is excellent both on TV and on the 'net.)

Of course, it greatly aids the BBC's cause that in Kevin Cadle their only rival (SkySports) has the single most irritating studio anchor in history. Learn to talk, man! I'm also completely fed up returning from an advertising break to hear him tell viewers the wrong score, sometimes even getting the teams wrong! Nick Halling gets more annoying by the day too, Sky's set looks horrid, and most of their other regulars are annoying to the point where I struggle not to throw my remote at the TV. On BBC, Rod Woodson talks a decent game and apart from referring to Roethlisberger's overruled touchdown as a try (the rugby equivalent) the anchor's been actually quite decent. I've enjoyed this Superbowl more than any other since 2005, and I think the best TV team I've seen has had a lot to do with that.

40 Re: Super Bowl XLIII Preview

Ed Werder has now referred to Dick LeBeau's 50 years in the league as "half a decade" multiple times on the air.

44 Re: Super Bowl XLIII Preview

Nice move by the NFL to invite the Flight 1549 entire crew along for VIP treatment.

And I'm with Paul on going for BBC coverage in the UK (personally I prefer to pay for NFL field pass radio coverage, watch the Monday/Sunday night games on terrestrial TV and forget about paying Sky anything...)

45 Re: Super Bowl XLIII Preview

I cannot believe NBC is not streaming this, it seems so shortsighted. I know it drove me to using illegal streaming for the first time ever tonight. That I cannot watch the game on my computer for free with the commercials is just absurd.

46 Re: Super Bowl XLIII Preview

Here we go!!! Finally. Pitt's first drive is impressive. From this distance-- even after the blown first play at the 1, they should score 6.

47 Re: Super Bowl XLIII Preview

Ugh. Arizona had that play stopped and couldn't do it. Is it legal for the OL to more or less grab the QB and pull him in?

Oh-- and looking at the review, that may have been the right call. He may very well have been down before the TD!

48 Re: Super Bowl XLIII Preview

With that defense, I totally disagree with Madden here. Go for it. Even if you don't score, the odds are very likely that you'll get the ball back without AZ scoring.