A look at fourth-down decision making in 2014 highlights Sean Payton, Marc Trestman, and... wait, this can't be right... Jim Caldwell? Plus: Chip Kelly may be less aggressive than you think he is.
17 Aug 2011
Guest Column by Chris Berney
This winter, I had a great idea. I decided I was smarter than Cam Cameron. My great idea stemmed from one basic theory I couldn’t get out of my head: Cameron was running an Air Coryell offense with players who were not suited for it, and he was failing at it.
However, the data proved me wrong.
Like many Ravens fans, I saw another frustrating offensive season from the team. Baltimore held the lead at some point in the fourth quarter of every regular season game, only to relinquish that lead in a staggering nine of them, many times because they couldn't move the ball in the fourth quarter to keep the defense off the field. The offense was once again in the bottom half of the league in many categories and wasn’t top ten in anything but interceptions allowed, rushing attempts and turnovers allowed.
The Air Coryell system that Cameron employs relies on two things. A power running game that picks up tough yards through the middle of the field, and speedy downfield receivers that are used to strike deep without warning. The Ravens have neither of these assets, which is why I believed Air Coryell was the cause of the offensive issues.
Ray Rice is not a power runner. At 5-foot-9, he’s a shifty back who forces tacklers to miss and is a stellar receiver. Chuck Muncie, who was the Chargers’ primary back under Coryell, was a 6-foot-3, 225-pound pile-pusher who was used as a receiving back half as often as Rice. The receivers weren't deep threats. Between 2007 and 2009, teams threw passes of 16 or more yards (through the air) to Anquan Boldin only 14 percent of the time and T.J. Houshmandzadeh 18 percent of the time. Derrick Mason has been known as a possession receiver specializing in short and intermediate routes with both the Titans and Ravens. Yet in 2010, Boldin and Houshmandzadeh had 28 and 36 percent of their targets go 16 or more yards, respectively. In fact, nearly one quarter of all Joe Flacco passes went 16 or more yards.
The problem is, while that sound like a nice theory, the data doesn’t support that the Ravens were ineffective at throwing the ball downfield. Flacco, when throwing the deep ball, had an 84.2 quarterback rating, primarily due to a relatively normal 36 percent completion percentage. Flacco’s rate of first downs/touchdowns per pass was 34 percent on passes of 16 or more yards to Boldin, Mason and Houshmandzadeh. It was 40 percent for passes of 15 or fewer yards. Here's the real kicker; looking at the DVOA numbers, the Ravens were significantly better when throwing the ball 26 or more yards. Their DVOA when throwing the ball that deep (which they did almost 40 times) was 67.5%, versus the league average of 22.6%.
Additionally, it doesn’t look like they were trying to force Rice into a power middle running role or misusing the run in general. Only 60 percent of his rushes were behind an offensive guard or center.
LeSean McCoy –- another relatively small scat-back who is also used regularly as a receiver –- rushed to the center of the line 58 percent of the time, so the Ravens weren't pounding Rice in true Air Coryell fashion. The Ravens also ran the ball only 52 percent of the time in the fourth quarter. This seems to be an absurd number for a 12-4 team that held a fourth-quarter lead in every game, but the Patriots, Steelers and Falcons – who finished with records equal to or better than Baltimore – ran only 50 percent of the time in the fourth quarter.
With that, my great idea turned into a bad one. If Cameron wasn’t really misusing their players by running an offensive system that couldn't mesh with their talents, where was the problem? The two main culprits seem to be the poor play of the offensive line and the ineffective running game.
The offensive line was a big problem last year. Flacco was sacked 40 times, second highest in the NFL, with the seventh highest sack rate of quarterbacks with at least 14 sacks. Football Outsiders ranks their Adjusted Sack Rate 25th in the league, and while they finished in the top 10 of Adjusted Line Yards last year, they were just 24th place in success rate on power running situations. Michael Oher was injured most of the season and played poorly at left tackle, Matt Birk had an awful year at center, and Chris Chester was a backup forced into a starting role because Marshal Yanda was playing out of position at right tackle. The cascade effect of losing Jared Gaither ran deep for the Ravens last season.
The running game was also ineffective, and the Ravens treated it as such, rushing on only 46 percent of their plays. Rice finished 19th out of the 46 running backs with 100 or more carries in DVOA, and had a career low 4.0 yards per attempt. The Ravens finished 29th in Second-Level Yards and 27th in Open-Field Yards. That points to Rice also deserving a share of the blame, although Willis McGahee (32nd in DVOA out of the same 46) "helped" by tying his career low with 3.8 yards per attempt.
Could the problem be the play calling? It’s possible, but I can’t find the data to support it. The NFL doesn’t run on data alone, of course, but Cameron’s going to have ten years of coaching experience in the NFL after this season and I’ll still have zero. So until they pay me to watch the games at home ala Tom Moore, the data is what I’ve got, and the data says Cameron’s system wasn’t the problem.
Chris is a Richmond resident with strong family ties to the Baltimore area. An analyst by trade, he writes columns for a Ravens fan-site and his football blog in his free time. He grew up watching Orioles and Colts games, and has been a football fan since elementary school. As a Ravens season ticket holder, he attends as many games as possible.
Football Outsiders is always accepting guest columns that have a unique perspective on either the NFL or college football. Send your ideas or samples to mailbag-at-footballoutsiders.com
63 comments, Last at 23 Aug 2011, 8:43am by PTORaven