Grantland logo

Arizona Has a Numbers Problem

The Cardinals have their signal-caller back and a way better than average coach, but can that be enough to buck the stats that say the postseason party is over?

On the surface, one of the last words you would use to describe the 2014 Arizona Cardinals is “lucky.” Teams that lose their top two quarterbacks and end up forced to use one of the worst starting quarterbacks in modern playoff history for their first trip to the postseason in five years aren’t lucky, not in the purest sense of the word. You can argue that health is a skill Carson Palmer doesn’t possess, but it’s natural to feel like the Cardinals were a very good team that ended up getting a raw deal by the end of last season.

I don’t disagree with that reading, but a closer look also suggests that Arizona might have been lucky to end up in a spot where it had a playoff game to contest at all. The Cardinals were by no means a bad football team in 2014, but a variety of factors indicate that they punched above their weight class. The numbers might seem to say that Arizona should decline this year, but are there enough mitigating factors to suggest that the Cardinals should defy history in 2015?

Better to Be Lucky Than Good

The 2014 Cardinals were an 11-5 team that outscored their opposition by a mere 11 points. There have been 88 other 11-5 teams since the league went to the 16-game schedule, and 86 of those 88 teams posted a better point differential than these Cardinals. Head coach Bruce Arians is very familiar with one of those two other teams, since he was at the helm for most of its stunning season as the interim head coach of the 2012 Colts. They kept their 11-win stretch up despite a variety of numerical indicators suggesting they would decline. The other example, the 2004 Falcons, dropped off from 11-5 to 8-8 the following year. Different paths!

The 11-point differential suggests the Cardinals were basically about average. The Pythagorean expectation has them playing at the level of an 8.3-win team last season. That gap between their actual win total and their expected win total, 2.7 wins, is the largest in football. No other team had a gap of two wins or more, with the Bengals in second at 1.9 wins.

Those 2.7 wins might not seem like a lot, but it’s just about the maximum of how a team can outperform its level of play over a 16-game schedule. Since 1990, only 10 teams have posted a larger difference between their Pythagorean expected win total and their actual win total than the 2014 Cardinals, with just 34 teams even managing to hit the two-win mark. It’s a meaningful, significant distance.

If you read the Buccaneers article on Monday, you will not be surprised to hear that teams that grossly outperform their point differential do not often repeat their feat the following year. Indeed, that fits this group, too. While those 2012 Colts stand out as a defiant exception to the rule, the majority of these teams — 22 of the other 33, excluding the Cardinals — declined the following season. Six maintained their previous record, while just five actually improved the following year. The average team from the group declined by 2.5 wins.

What I found really interesting about those teams is that virtually all of the difference in their year-to-year record change came with their performance in one-score games. Their performance in games that were decided by eight points or more was markedly consistent; they won 58.6 percent of those games during the year in which they defied the Pythagorean gods and then won 57.0 percent of those games the subsequent season. In the close ones, though, things were totally different:

After going 195-55 in one-score games the first year, our interlopers bounced all the way back below .500 with a 102-121 mark the following year. It was almost the entirety of the reason they dropped off.

The Cardinals have to be concerned about such a decline. They went 4-1 in one-touchdown games last year, with their only loss coming against the 49ers in Week 17, which was a meaningless game for Arizona and a farewell game for Jim Harbaugh and about half of the San Francisco roster. Wouldn’t you know that the 16 other teams that went 4-1 in one-score games since 1990 went a combined 49-48 in those same games the following year? If the Cardinals are .500 in those games this year, that alone would account for an almost surefire decline.

And that 4-1 mark also really understates how frequently Arizona won close games, because there were a number of tilts that went the Cardinals’ way with late scores. That includes:

Week 2 against the Giants, when Rashad Jennings fumbled the ball away totally untouched in the red zone late in the fourth quarter before the Cardinals kicked a field goal with 1:17 left to go up 25-14.

Week 3 against the 49ers, when the Cardinals couldn’t run enough time off the clock with kneeldowns and kicked a field goal to go up 23-14 with 34 seconds left.

Week 6 against Washington, when Rashad Johnson picked off a Kirk Cousins pass with 28 seconds left and could have fallen down to seal the win, but instead returned it for a pick-six to go up 30-20.

The Cards also had an eight-point win over Detroit and another field goal in the final minute to give them an 11-point win over Oakland. And none of their four actual close wins were the sort of fake close games that include a late score from the other team; Arizona beat the Chargers by one on a late Palmer touchdown pass, had to stop the Eagles three times from the 16-yard line to win by four, got a bizarre Travis Kelce fumble in the red zone and a late stop to hold off the Chiefs by three, and stopped the Rams on the Arizona 43-yard line inside of two minutes to win by six.1


1.

St. Louis technically got another drive after that one, but with 18 seconds and 86 yards to go.

One better way to put Arizona’s performance in context is DVOA, which adjusts every single play for various contexts and evaluates how Arizona performed over each snap of the 2014 season. In DVOA’s estimation, the Cardinals weren’t very pretty last season, even if their record was. They were 6.4 percentage points below average last season, leaving them stuck between the 6-10 Giants and the 7-9 Browns in 22nd place.

What about those sorts of squads, teams that win a bunch of games with a middling DVOA? Do they keep it up the following year? Football Outsiders only has DVOA available going back through the 1989 campaign, but that leaves us with 12 teams that posted 10 or more wins with a sub-zero DVOA before the Cardinals pulled off the feat last season. They didn’t fare as well during the subsequent campaign:

Those teams declined by an average of 2.6 wins the following year. By the numbers, the Cardinals fit into a group of teams that dramatically outperformed their level of play, and those teams very frequently decline. Now, we have to figure out whether Arizona might be an exception to the rule.

Cardinals vs. Cardinals

As far as I can tell, there are two strong arguments against the Cardinals regressing back toward the mean, with the usual debates about personnel improvements thrown in as a third reason. Let’s run through those arguments and see how much water they hold.

Argument 1: They won’t have to turn to Ryan Lindley in 2015.

Having seen Ryan Lindley play at the end of last year, I have to admit this is a very compelling argument. The Cardinals were 6-0 with starting quarterback Carson Palmer at the helm, and while Drew Stanton wasn’t quite as effective, Arizona was in the middle of beating the Rams to get to 11-3 when Stanton went down with sprained ligaments in his knee. The Cardinals had averaged 20.5 points per game through that Rams win, but then scored an average of 13 points across their three subsequent losses, including the playoffs, to end the season with Lindley under center.

There’s little reason to fault the Cardinals for having to turn to Lindley; there isn’t a team in football that is going to look good when it’s stuck going to its third-string quarterback, and the chances that a team will ever need to get past its primary backup for meaningful, regular snaps simply aren’t very high. You can maybe wonder why the Cards took Logan Thomas if they weren’t going to give him a shot ahead of Lindley, but the fourth-rounder was a project that Arians was hoping to mold for the post-Palmer future. He wasn’t remotely ready in 2014.

Lindley is no longer on the roster; it wouldn’t be a surprise if his last NFL pass came in the playoff loss to the Panthers. And Arizona shouldn’t need to turn to somebody off the street again in 2015. Palmer’s recovery from his second torn ACL has gone well, and he should be ready to suit up for the Cardinals in Week 1 against the new-look defense of the Saints. Stanton, too, is back on the roster and should be ready if Palmer can’t shake his injury woes.

There’s a difference between healthier and healthy, though, and I’d be worried about Palmer. He’s now a 35-year-old quarterback coming off multiple ACL tears, and while he recovered from the catastrophic knee injury he suffered in the playoffs as a member of the Bengals, changes to his mechanics ended up causing him elbow troubles. (Granted, mechanics is something it seems like Arians could help with.) Palmer was excellent in his six-game sample last season, but it was a step well ahead of how he had played in 2013 with the Cardinals, let alone the previous two seasons in Oakland. It’s extremely unlikely, for one, that he’ll post a 1.3 percent interception rate again.

That’s an even bigger problem if the Cardinals have to turn to Stanton. The former Lions backup wasn’t a Lindley-esque problem during his eight starts, but the numbers weren’t great. He completed just 55.0 percent of his passes, and while that’s not quite as bad in the sort of downfield attack Arians runs, Stanton was careless with the football. He threw five picks on 240 attempts, which produced a very reasonable interception rate of 2.1 percent, but Stanton was incredibly lucky with some of the mistakes he made.

Despite only throwing about a half-season’s worth of passes, Stanton had a whopping eight dropped interceptions in 2014, which was tied for second in the league behind Brian Hoyer’s nine. Stanton had thrown nine picks on his previous 187 attempts for a 4.8 percent interception rate, and while that’s a small sample, that ridiculous dropped interception rate seems to suggest he would be an interception machine if the Cardinals put him back into the starting role for any serious stretch of time in 2015.

And to get back to the bigger picture, while the Cards were unquestionably worse with Lindley playing quarterback, they were still exceedingly lucky with both Palmer and Stanton at the helm. No matter how you split their totals, the Cardinals won more games than their point differentials would have expected, regardless of who was under center:2


2.

I’m counting all points scored/allowed in each game for the quarterback who started here.

If the Lindley era hadn’t happened and the season ended after 14 games, the Cardinals would have been an 11-3 team that we would have expected to win … 8.3 games. Instead, they ended up as an 11-5 team that we would have expected to win 8.3 games. We would be thinking of them as an even bigger outlier if it weren’t for the Lindley-led crash landing to the end of the season.

Argument 2: Bruce Arians is an incredible head coach.

Again, I can’t really argue with this one on its face. For a guy who was unceremoniously dumped into retirement by the Steelers three and a half years ago, Arians has been downright incredible since. He did brilliant work in Indianapolis as the interim head coach while Chuck Pagano dealt with leukemia, and after taking over a 5-11 team that had collapsed under Ken Whisenhunt in 2012, Arians has led a Cardinals team that doesn’t always look like much on paper to a 21-11 record over the past two seasons. He’s proven the naysayers wrong and deserves to be considered among football’s best head coaches.

If there were any factors that would allow a team to win a disproportionately high percentage of its close games year after year, you would think that a great quarterback and a brilliant head coach would be at the very top of the list. Given how poorly some coaches manage close situations at the end of each half, competency in tight situations can itself be incredibly valuable.

It’s worth noting, then, that Arians has managed to begin his coaching career by beating his team’s Pythagorean expectation in three consecutive seasons, including two significant outliers. If we give Arians credit for the entirety of the 2012 season, his aforementioned Colts went 11-5 while being outscored by 30 points, typically the point differential of a 7.2-win team. That’s a massive 3.8-win difference. In 2013, the Cardinals went 10-6 and played like a 9.5-win team, adding another half-win to Arians’s tab. Throw in the 2.7-win gap from last season and Arians’s teams have outplayed their expected win total by seven wins over the past three seasons.

Seven! That is, to the best of my knowledge, unprecedented. I don’t have point differential numbers for individual coaches going back through league history, but given how unlikely a coach is to move on after an 11-5 season like Arians did in Indy, we can come pretty close to finding Arians comparables by looking at how teams performed over three-year stretches. Has anybody outperformed their expected win total over a three-year run more than Arians’s teams have over the past three seasons?

The answer is no, and it’s not especially close. Since the AFL-NFL merger in 1970, just three teams managed to outperform their Pythagorean expectation by more than six wins in a three-year stretch, and none was particularly near Arians’s seven-win figure. Only 10 teams managed to make it past five wins over a three-year stretch. While it’s not a large enough sample from which to really draw any conclusions, let’s also take note of what they did the following year:

Excluding the 2015 Colts, those teams beat their Pythagorean expectation the following year by an average of 0.5 wins; it’s nowhere near what they had done over the previous three seasons, but still above the simple expectation of the mean. And the most notable loser among that group is the 2011 Colts, who had one obvious major change impact their situation, given that they spent the entire season without Peyton Manning at quarterback.

The Manning thing made me wonder; I don’t think anybody is or was suggesting that Jim Caldwell is a great coach, but if you pretend that he wasn’t around for that Manning-less season in 2011, his teams have also greatly exceeded expectations. Caldwell’s Colts outperformed their point differential by 4.1 wins between 2009 and 2010, and in his first year with his new charges in Detroit, the Lions outperformed their point differential by 1.8 wins. For three of his four seasons, Caldwell looked like a stat-annihilating monster.

I’m not saying that to suggest Caldwell has this skill; I’m pointing it out to illuminate how this seeming-skill disappeared entirely once Caldwell lost his starting quarterback, given that the Colts were promptly 1.2 wins under their expected win total based on point differential when Manning missed all of 2011. It makes me think this is more of a quarterback-based skill than a coaching-specific asset. The existence of the 2012-2014 Indianapolis Colts, who have continued to exceed their Pythagorean expectation in 2013 (1.6 wins) and 2014 (1.0 wins) without Arians, also makes me lean in that direction.

It’s difficult for me to say Arians plays little or no role in his team exceeding its expected win total from its point differential. I wouldn’t be surprised at all if it meant something over the long haul, if we could somehow simulate this season 1,000 times with and without him around. In terms of the impact he can offer, though, it’s virtually impossible to keep producing the kind of enormous differences between expected and actual win totals he’s managed in his first three seasons. In the long run, even if he does offer a concrete improvement, Arians’s ability to beat the point differential is surely going to look more like the half-win gap the Cardinals enjoyed in 2013 than the crooked numbers hung up by the Colts in 2012 and the Cards in 2014.

Argument 3: The other stuff.

I’m less enthused about the other soft factors making a meaningful difference in Arizona’s chances of ducking history. I like what they did this offseason in bringing in former 49ers guard Mike Iupati and taking a chance on a pair of talented-but-injury-prone Falcons defenders in linebacker Sean Weatherspoon and defensive tackle Corey Peters, but salary-cap issues led them to lose some key contributors. Massive nose tackle Dan Williams and resurgent cornerback Antonio Cromartie both left for greener pastures in free agency, and while Peters may help chip in for Williams, general manager Steve Keim elected to stay in-house to replace Cromartie. Jerraud Powers will become a starting corner, and the Cards will hope that Tyrann Mathieu will lock down the slot corner job while looking healthier than he did in his return from a torn ACL last season.

Normally, I would say the Cardinals’ schedule should be friendlier, but I can’t even give them that semblance of hope. Arizona had the league’s second-toughest schedule last season, but that’s nothing new. Caught in the wars of the NFC West, Arizona had the league’s fourth-toughest schedule in 2013 and the league’s toughest slate in 2012.

As much as they deserve a friendly schedule, one isn’t coming in 2015. Chase Stuart’s breakdown of Vegas point spreads suggests Arizona is projected to face the league’s toughest schedule in 2015. The NFC West faces a pair of difficult divisions in the NFC North and the AFC North, and since the Cardinals finished second in the West last year, they also run up against the Eagles and Saints. And that’s before considering their six games against the NFC West.

There are other reasons to figure that the 2015 Cardinals might not be as good as the 2014 edition. Arizona lost gifted defensive coordinator Todd Bowles to a head-coaching job with the Jets, and while replacement James Bettcher has spent the past two seasons on Bowles’s staff as the team’s outside linebackers coach, there’s no guarantee he’ll be as good as Bowles was at optimizing his defenses for the unique skills of Arizona’s many tweeners. The pass rush remains a major question mark, as the Cardinals attempted to find talent by taking a low-risk flier on LaMarr Woodley, who hasn’t been a serious threat since 2011. And while there’s plenty of young talent on defense, Arizona’s two most important players on offense will turn 32 (Larry Fitzgerald) and 36 (Palmer) during 2015.

And yet, I can’t fault Cardinals fans for believing. Arians has worked wonders over the past couple of seasons, often with a limited, injury-hit group of players. Arizona should be healthier in 2015, and if the combination of Andre Ellington and third-rounder David Johnson can get going behind a much-improved offensive line, the Cards could actually have a running game to complement that downfield passing attack. It’s not hard to imagine the Cardinals being good. It’s just hard to imagine them being as lucky.