Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2017 Defensive Personnel Analysis

Defenses have taken a wide variety of responses to the rise of 11 personnel. Is any one system better than another? And how has the rise of the "moneybacker" changed defensive philosophy?

20 Oct 2015

Any Given Sunday: Steelers Over Cardinals

by Andrew Healy

Will the real Arizona Cardinals please stand up? For two-thirds of the season, we have seen the juggernaut Cardinals. Arizona has four of the 13 performances this season where a team had a single-game DVOA over 60.0%. Cincinnati has two such dominant wins; nobody else has more than one. As weird as it is to think of the Bengals dominating teams, it's even more out of character for the Cardinals. From the year they moved to Phoenix in 1988 through 2014, the Winter Olympics happened more often than Arizona blowout wins. The Cardinals won a total of six games by 25 points or more during those 27 seasons. Through Week 6 this year, the Cardinals already have three such victories.

But for the second time this season, the Cardinals followed a blowout win with a loss, continuing a long pattern for the franchise. After Sunday's loss, the Arizona Cardinals are a remarkable 1-7 following wins by 25 points or more since 1988. More relevant for the present edition of the team, Bruce Arians' Cardinals fell to 1-3 in the games following scoreboard blowouts.

We know from previous history that the Cardinals' very strong performance across four blowout wins and two relatively close losses matters the most for predicting the rest of the season. But Carson Palmer made some throws that might worry a Cardinals fan going forward. We'll get to that and to the surprisingly promising situation in Pittsburgh sans Roethlisberger after breaking down the advanced stats from Sunday.

By the Numbers

Going into the game, the Cardinals narrowly trailed the Patriots for the top spot in our DVOA rankings. With the Steelers not as good as their DVOA given that they were starting Michael Vick at quarterback, ever-savvy bettors backed the Cardinals, pushing the line up a couple points to 5.5 before kickoff.

The Matchup
PIT DVOA ARI DVOA Score Line PIT Pre-game
Win Prob
18.5 (5th) 51.5% (2nd) 25-13 PIT +5.5 34.1%

Despite losing to an inferior opponent, the Cardinals actually played Pittsburgh close on both offense and defense according to DVOA. They lost the DVOA matchup mostly by getting dominated on special teams. Without opponent adjustments, the Cardinals actually won on both offense and defense, but not by enough to offset the disadvantage on special teams.

The final score is misleading concerning the flow of the game. The Cardinals outgained the Steelers 469-310, but lost on a series of mistakes. Arizona committed the game's only three turnovers. The Cardinals also shot themselves in the foot with a personal foul by offensive tackle Bobby Massie on their first drive, then a dubious offensive pass interference wiped out a Michael Floyd touchdown just before the half. On the day, the Cardinals racked up 111 yards in penalties, more than twice as many as Pittsburgh.

DVOA (Opponent adjustments included)
ARI 15.5% -5.7% -9.9% 11.3%
PIT 21.3% -3.1% 16.7% 41.2%

VOA (No opponent adjustments)
ARI 13.0% 0.8% -9.9% 2.3%
PIT 5.5% 13.2% 16.7% 9.0%

While the Cardinals were better by DVOA than they were two weeks ago against the Rams (-7.9%), our opponent adjustments likely give the defense too much credit since they got to face Vick and third-stringer Landry Jones, who came into the game after Vick entered the concussion protocol. The Cardinals also played most of the game against a Steelers' line that was missing left tackle Kelvin Beachum, who tore his ACL on Sunday.

The Ups and Downs of Carson Palmer

Carson Palmer had an uneven day. While his numbers (16.1% DVOA) came out OK, he made a series of very poor throws and one inexplicable decision with the game on the line. Here is a sampling of Palmer's mistakes:

Carson Palmer's Biggest Mistakes
Situation Quarter Time Score Description
Third-and-5 on PIT 48 2 14:18 7-0 ARI Poor decision to throw deep into double coverage
First-and-10 on PIT 14 2 1:51 7-3 ARI Misses wide-open John Brown on post corner; ARI settles for FG
First-and-10 on PIT 20 4 2:25 18-13 PIT Throws interception to single-high safety in deep middle

The last throw is the worst of the bunch. Pittsburgh lined up with Mike Mitchell as the obvious safety in the middle of the field. Carson Palmer looked for John Brown on a post not far from where Mitchell was standing at the snap. It was a no-hope throw that was deservedly picked. On commentary, Troy Aikman theorized that Palmer made an awful decision and just missed Mitchell before the snap. I doubt that. Instead, I'm guessing that Palmer saw Mitchell, intended to throw it over the top, and didn't get enough on the throw after getting some late pressure from behind.

Even if that's right, it's a questionable decision with a small margin for error. The Abolitionist had a huge game (10 catches, 196 yards, the second-most yards for any receiver this season) and inflated Palmer's stats by making a couple of great catches on contested balls, but Palmer has to go somewhere else with that ball. There was no reason to take that risk, particularly on first down.

But there's little reason to think of Palmer's failures against the Steelers as reflecting some larger pattern of failing in the biggest moments of games. In the fourth quarter when the game is within one score, Palmer has a DVOA of 41.7%, just above his overall DVOA of 36.6%. A series of mistakes cost the Cardinals a game they should have won on Sunday, but the most important concern for Palmer going forward is not his performance but keeping him healthy.

A New Defense in Pittsburgh

Coming into the season, the Steelers' defense was the primary obstacle threatening to keep them out of Super Bowl contention. Through six weeks, the defense has been surprisingly competent. A year after ranking 30th in defensive DVOA, the Steelers are 14th with a DVOA just above average at -0.6%. Since opening night against the Patriots, the Steelers have been consistently solid, too. Four out of five times, they have posted a negative DVOA, and the one positive number (Week 4 against the Ravens) came in at 3.3%.

After fielding the third-oldest defense in football by snap-weighted age in 2014, the Steelers have not gotten much younger in the front seven, but they have found youth in the secondary. 25-year-old Antwon Blake has started all six games at corner. After an injury to aging Will Allen against the Chargers, 25-year-old Robert Golden made his first NFL start against the Cardinals, showing the kind of burst at safety that Troy Polamalu could no longer provide last season.

But despite an old and struggling defense, the Steelers did not make wholesale personnel upgrades in the offseason. The biggest change involved replacing a coaching legend at defensive coordinator. Dick LeBeau had been the Steelers' defensive coordinator for 11 seasons before Kevin Butler replaced him for 2015. Whether Butler deserves credit for the Steelers' defensive rebound or whether it mostly just reflects regression towards mediocrity, this defense is -- at worst -- good enough for a team with a top-five offense once Roethlisberger returns.

Creative Coaching

When the Steelers took the lead on Jones' first touchdown pass to Martavis Bryant, they went up 12-10 with 8:45 to go in the third quarter. The conventional wisdom says to kick the extra point to go up three, but Mike Tomlin instead continued his own trend and went for two.

It's possible that this is the wrong call as Scott Kacsmar and Vince Verhei argued in Audibles this week. I think it's a very close call with many factors to consider. Against Tomlin's decision, there is the margin in the game; the fact that his quarterback had just thrown the first two passes of his career; and that the third-ranked defense was lining up to stop them and had stopped third-and-1 runs for losses already. OK, that's a pretty strong case against going for two, and probably most would agree with that point of view.

But there are some other factors working for Tomlin's call. As an underdog, taking risks generally is a good idea. The Steelers were on their fourth kicker. Chris Boswell had looked good on previous kicks in that game, but he was still largely an unknown. More than half the third quarter remained, reducing the importance of going up three compared to later in the game. Finally, their great running back could make a planned pass more likely to succeed.

On balance, I think I lean towards going for the extra point on further thought. But I don't think it's obvious and it's possible that the culture of aggression that Mike Tomlin is establishing could pay dividends later on when the Steelers attempt a more important two-point conversion when most teams would choose to kick.

The Keep Looking at Wins Stat of the Week

In 2013, our predictions for college quarterbacks called Landry Jones "The Asterisk, Part II." While the projections were only for quarterbacks taken in the first three rounds (hence the asterisk), the model at the time based on Jones' college stats and experience gave him a predicted NFL performance better than all the more highly-touted quarterbacks in his class.

Fast-forward to Sunday. Jones comes in for Michael Vick and proceeds to play like a quarterback who should have gone higher in the draft.

Steelers Quarterbacks Against the Cardinals
Player Comp Att Yds TD INT DVOA
Landry Jones 8 12 168 2 0 136.6%
Michael Vick 3 8 6 0 0 -97.4%

Beyond the numbers (which were inflated by big-time YAC from Martavis Bryant on the touchdown that clinched the game), Jones made some impressive throws. Even more than the two touchdowns, his dropped-into-the-bucket throw over the linebacker on a 22-yard throw to Bryant showed both great recognition and accuracy, with Arizona playing man on Antonio Brown and zone elsewhere.

Jones looked good and it's logical that he should jump over Vick on the depth chart, but let's stay calm here. These were just 12 throws made to one of the league's best sets of receivers. Moreover, our updated system QBASE likes Jones significantly less than the 2013 model. It gives Jones about a 60 percent chance of being a bust and ranks him a little higher than Jameis Winston if we extrapolate the projection a bit to allow for a fourth-round pick. QBASE is less bullish on Jones because it incorporates the scouting information through his draft position and because it corrects for his teammates at Oklahoma.

So Jones has some potential, much more than a typical fourth-round pick, but he is not likely to be close to as good as the Asterisk, Part I: Russell Wilson. Wilson's college resume was so off-the-charts good that even his third-round draft position couldn't stop him from being among QBASE's top five prospects of the last 20 years.

Posted by: Andrew Healy on 20 Oct 2015

2 comments, Last at 21 Oct 2015, 6:01am by Topas


by RickD :: Tue, 10/20/2015 - 7:57pm

Seems weird that last year the Cardinals' DVOA trailed far behind their W-L and this season the opposite seems to be happening.

by Topas :: Wed, 10/21/2015 - 6:01am

That reminds me of the times when the Eagles had a middling to good record while being first in DVOA almost over years...