by Andrew Healy
Has a football team's season ever turned on a single last name on both sides of the ball as much as the 2014 Ravens? Almost 20 percent of the Ravens' opening day lineup played with "Smith" on the back of their jersey, a last name coincidence topped in awesomeness perhaps only by the '70s Rams Youngblood duo. While the Ravens' Smiths don't include a guy who played a Super Bowl on a freaking broken leg, they do include a middle linebacker (Daryl Smith), a wide receiver (Torrey Smith), and the two Smiths that may have held the key to last Sunday's game, as well as the rest of the Ravens' season. The offensive key is the other receiver (Steve), and the defensive key is the player (Jimmy) who finally developed into a shutdown corner this season before getting injured.
At first glance, it looks like Jimmy is the pivotal Smith. When Jimmy went out early in Week 8 against the Bengals, the Ravens were the second-best team in football (by DVOA). They had allowed just 14.9 points per game in the first seven weeks, and no more than 23 points in any one game. In the five games played mostly or entirely without Jimmy Smith, the Ravens have allowed 27.6 points per game, and at least 27 points in four of the five.
The secondary appears to be the primary cause of this decline. It leaps off the film from the Chargers game. The Ravens' front four actually played well. Elvis Dumervil, despite not getting a sack, had two quarterback hits and was in Philip Rivers' face all game long. Overall, the Ravens' rush consistently got home and Rivers rarely threw from a clean pocket. Rivers nevertheless put up big numbers due to holes in the secondary. That secondary was not only missing Smith, but also the injured Asa Jackson (five games, four starts), previously-cut Chykie Brown (seven games, one start), and previously-cut Dominique Franks (four games, one start). Playing at corner for the Ravens were the largely-ineffective-this-season Lardarius Webb and two players rushed into emergency service after Brown and Franks were cut: Danny Gorrer and converted safety/special teamer Anthony Levine.
Against the Chargers, the patchwork secondary did not hold up well. Receivers ran free all day, none more so than on the first drive of the third quarter, when the Chargers faced third-and-10 from their own 30. Malcom Floyd ran a deep post with Gorrer in coverage. As Floyd made the cut to the post, Gorrer actually appeared to be in pretty good position to defend it.
But then Floyd exploded past Gorrer like Facebook shooting by Myspace. With Gorrer diagnosing the route pretty well, Floyd still ended up with huge separation when the ball arrived.
(Note that this a good example of football speed not always being best measured by a player's 40 time. Despite being an undrafted free agent, Gorrer has elite speed. He ran a 4.40 40-yard dash at the combine, slightly faster than Floyd's 4.44. But in the perhaps more informative three-cone drill, Floyd's time (6.66) beats Gorrer's (6.90) by a decent margin. Floyd also has 5 inches of height on Gorrer.)
With two-thirds of the Ravens' secondary consisting of recently-installed undrafted free agents, Rivers had receivers in favorable matchups seemingly all day. On Sunday, the Ravens lost despite a pretty good day of offense according to DVOA. The Chargers' offense, feasting on the Ravens' secondary, had an even better day.
So it certainly seems that Jimmy is the key Smith to the Ravens' recent struggles, right? Well, not so fast. Sunday's game isn't really representative of how things have been going overall since the injury to Jimmy Smith. Look at the Ravens' DVOA breakdown before and after the Smith injury.
|Before Smith Injury||5-2||14.9||-2.7%||21.0%|
|After Smith Injury||2-3||27.6||-0.8%||-1.8%|
Despite the big jump in points per game, the Ravens' average defensive DVOA has gotten only a little worse, increasing from -2.7% to -0.8%. The Ravens are still good up front and that was on display even on a day where the pass defense mostly failed. The defense's decline might be a little bit more than DVOA shows -- even if DVOA is the best available metric, we don't claim it's perfect -- but it's certainly a lot less than the increase in points per game would suggest.
[ad placeholder 3]
Even if the defense hasn't fallen as far as it seems at first glance, the Ravens still may be in trouble going forward. Their offense has declined markedly over the last five games, falling from the second-best unit in the league (21.0%) to below-average (-1.8%). That decline in performance coincides with a dropoff in production from the league's best quote machine. Over the first seven games of the season, Steve Smith averaged 91.4 yards receiving per game. Since then, Smith's yardage has dropped off a cliff, to just 35.8 yards over the last five games and to just one catch for 2 yards on Sunday. Included in his other three targets was a terrible drop on a corner route early in the fourth quarter. The 2 yards represented Smith's lowest total since he was shut out in a game eight years ago when Chris Weinke was his quarterback and the Panthers had seven pass attempts, the second-fewest pass attempts in thirty years.
While Smith's performance on Sunday was historically unusual, his late-season swoon has become the norm. Until his age-30 season, Smith was excellent despite poor-to-mediocre quarterback play throughout the season. Since his age-30 season, Smith has continued to be very good early in the season, but has dropped off in production later in the year. In fact, in the first half of the season, Smith has averaged slightly more yardage than he did in his younger years.
|Young Steve Smith vs. Old Steve Smith|
(Age > 30)
Since turning 31, his seasons go like another Smith's career, starting at the heights of Fresh Prince and Independence Day before heading down into Hitch. We can only hope Sunday wasn't Steve's After Earth moment. Over the last five seasons, Smith averages 23.4 fewer yards per game in the second half of the season than the first half, a drop of more than 30 percent. His targets, his catch rate, and his yards per target are all markedly lower later in the year.
A couple of graphs show how different Old Steve Smith is from Young Steve Smith in production across the season. Young Steve Smith had basically constant production from Week 1 to Week 16, as shown by the regression line. For fellow stats nerds, the curve is from a kernel-weighted local polynomial regression, which is just a nice way to capture Smith's trend over the course of the season. Young Steve Smith had a flat curve, and so basically constant production.
[ad placeholder 4]
In contrast, Old Steve Smith has a steady decline from Week 1 through Week 13 or so, with his production leveling off about 40 percent lower than where it was at the start of the season. So Smith has had a pattern of late-season decline for some time now. While Smith is an unusual character and seemingly the exact wrong person about whom we should make predictions, the odds are against early-season Smith returning this year. And that's a pretty big problem for the Ravens' offense as they make their playoff push. Smith's swoon coincides closely with the Ravens' offensive struggles.
The Chargers story can't quite keep up with the Smiths. San Diego's offensive play is likely to remain at a high level the rest of the season, as they try to edge into the playoffs with an incredibly difficult schedule staring them in the face. The Chargers' offensive DVOA takes a big hit from the debacle in Miami, but San Diego is actually one of only two teams in football that has an offensive DVOA higher than 15.0% in at least ten games this season (Green Bay is the other). We saw the real San Diego offense on Sunday. Unfortunately for the Ravens, we might have seen the real Ravens' secondary and their real set of receivers, too. Jimmy Smith isn't walking through that door, and Young Steve Smith probably isn't either.
(For more on the play-by-play of San Diego's big fourth-quarter comeback, check out today's edition of Clutch Encounters.)