This week's DVOA commentary is all about worsts. Come find out where Washington stands among the worst special teams in DVOA history, whether San Diego has the biggest gap between offense and defense, and whether Baltimore or Jacksonville has the worst running game we've ever tracked.
17 Feb 2009
by Vince Verhei
A timeline of the month of February for the Arizona Cardinals:
February 1: Cardinals lose classic Super Bowl to Steelers. Optimism for 2009 high.
February 6: Cardinals offensive coordinator Todd Haley hired as head coach of Kansas City Chiefs. Optimism for 2009 dims.
February 7: Cardinals defensive coordinator Clancy Pendergast fired. Optimism for 2009 detectable only by powerful microscopes.
The exact role and influence of offensive and defensive coordinators in the NFL is hard to define and even harder to measure. Their responsibilities and duties change from team to team and from season to season. Stability, however, can only help a team. Will the loss of top coaches on both sides of the ball will cause enough upheaval to ruin the Cardinals' chances of returning to the Super Bowl?
Surprisingly, Arizona isn't the only Super Bowl team in recent memory to lose both coordinators. The 2004 Patriots beat the Eagles in Super Bowl XXXIX, then saw defensive coordinator Romeo Crennel leave to run the Cleveland Browns, while offensive coordinator Charlie Weis departed to pick up the reins at Notre Dame. In Pro Football Prospectus 2005, Ned Macey looked at what happened to teams when their coordinators left to take head coaching jobs. From 1989 to 2003, 35 coordinators were hired as head coaches in the NFL. The average unit (offense or defense) dropped 2.9 places in the NFL rankings for points and 3.3 places in the average rankings for yards.
One reason that units decline when they lose coordinators is regression to the mean; good teams are much more likely than bad teams to play worse the following year. Teams that placed in the top five in points scored or allowed and then lost their coordinator were worse by about 45 points the following season. Teams that ranked in the top five but retained their coordinators declined by 55 points. We don't suggest that all top-ranking teams fire their coordinators in an attempt to mitigate an inevitable decline, but this does show that retaining coordinators does not guarantee outstanding play.
Since that initial study, 15 more coordinators have taken NFL head coaching jobs (including Norv Turner twice). Six units improved in the rankings, most notably the Philadelphia offense, which jumped from 18th to sixth in points scored after Brad Childress left in 2005. Nine units declined, none moreso than the Patriots defense, which fell from second to 17th after Crennel left. Of these 15 units, the average team declined 3.8 spots in the rankings.
Some of you have no doubt spotted the MacGuffin in all this. This study only looks at coordinators who left to take head coaching jobs. It is not looking at coordinators who are fired, like Pendergast was. The Cardinals were 28th in points allowed last season; they were much better in DVOA, but still ranked 21st. This was, at best, a below-average unit last season, and improvement on this side of the ball is somewhat likely no matter who is in charge in 2009.
Everybody! The Cards could lose key players at quarterback (Kurt Warner), running back (J.J. Arrington), tight end (Leonard Pope), defensive end (Bertrand Berry), and linebacker (Karlos Dansby) to free agency (or, in Warner's case, retirement). Leading rusher Edgerrin James is also a candidate to retire or be released. And then there's Anquan Boldin, the No. 1 receiver disguised as a No. 2 who has been publicly angry with Cardinals management for nearly a year now and wants to be traded. All told, the Cardinals have only 41 players under contract for 2009.
Almost anybody! The Cardinals have a league-high $41 million in cap space. Most of that should go to Warner, who looked like an MVP in the first half of the season and again in the playoffs, and Dansby, the team's leading tackler. (Dansby was the only outside linebacker in the league involved in 20 percent of his team's run tackles.) Also, if Boldin can be convinced to return he won't come cheap. If he does leave, the Cards would not be desperate to replace him. Steve Breaston played well as a third receiver last season, and Arizona won the two games Boldin missed after shattering his face against the Jets, scoring 41 points against the Bills and 30 against the Cowboys. The Cardinals ranked 30th in run blocking last year and their tackles were abused in the Super Bowl; Jordan Gross or Jon Runyan could make immediate impacts. On the defensive side of the ball, they'd love to sign Albert Haynesworth, but so would 31 other teams. The Cards switched between a 4-3 and 3-4 last season; if they do that again, they'd be the perfect spot for athletic pass rushers like Julius Peppers or Terrell Suggs.
Forty tackles, four sacks. Not terrible numbers for a defensive end, but not what the Rams were hoping for when they took Chris Long with the second overall pick out of Virginia last April. Were St. Louis fans right to expect big things from Long right away? And can we expect a big jump from Long in his sophomore season?
Between 1996 and 2007, 16 defensive ends were selected in the top 10 picks of the draft:
|Defensive Ends Drafted in Top 10, 1995-2007|
|Name||Team||Rookie Year||Sacks||Second Year||Sacks||Difference|
In hindsight, Long's four sacks look like what we should have expected from a rookie pass rusher; nine of these ends had more sacks as rookies, seven had fewer. They averaged 5.1 sacks in their rookie seasons. The next season, nine of them saw some improvement, two matched their sack totals, and five saw their numbers drop. The average sacks in Year Two climbed modestly to 6.1, but only two players -- Mario Williams and Greg Ellis -- have made the kind of improvement it would take to get Long into double-digit sack numbers.
If we look at sacks per game instead of total sacks, things don't change much. These players averaged 0.35 sacks per game as rookies and 0.46 sacks per game the next year, an addition of 1.7 sacks over a 16-game season. It was not Williams or Ellis who made the biggest jump on a per-game basis however; that would be Courtney Brown. The Browns defensive end posted 4.5 sacks in each of his first two seasons, but he played 16 games in 2000 and only five in 2001. The good news is that his production (at least in terms of sacks per game) more than doubled; the bad news is that he missed 11 games.
Expect Long to make modest improvements this year and collect five or six sacks, but do not look for him to be a Pro Bowler anytime soon.
Incidentally, two other ends were picked in the top ten last season: Derrick Harvey (who had 3.5 sacks for the Jaguars) and Vernon Gholston (who had no sacks and just 13 tackles as an outside linebacker for the Jets). Again, only modest improvement should be expected from these two youngsters.
The Rams' first priorities are to re-sign free safety Oshiomogho Atogwe and cornerback Ron Bartell. Dante Hall set a career low in kickoffs last year (20.6 yards per return) but was still effective as a punt returner (10.3-yard average, right below his career rate of 10.5). The Rams could also lose veterans La'Roi Glover and Jason Craft, whose best days are long behind them. Beyond that, we're talking the Brock Berlins and Dane Lookers of the world, players who probably should be replaced anyway. Though not free agents, two of the greatest players in Rams history -- Torry Holt and Orlando Pace -- are likely to be cut for salary cap reasons. The Rams are only $14 million under the cap; releasing those two veterans would nearly double that room. (The Sporting News reports that the Rams turned down a first-round draft choice from the Titans in exchange for Holt last season. That's a trade both teams likely wish they pulled off now.)
The Rams are not overflowing with talent, but they at least have potential at defensive line and in the secondary. For what seems like the 27th season in a row, the Rams need to rebuild their offensive line. An interior lineman like Pete Kendall of the Redskins or Mike Goff of the Chargers, or a tackle like Jon Runyan or Tra Thomas of the Eagles, would help this team take a small step back to respectability.
Three years after being selected sixth overall in the 2006 draft, Davis is most famous for being thrown off the field by coach Mike Singletary after committing a personal foul against Seattle. He has sandwiched two horrible seasons (-30.4% DVOA in 2006, -24.6% in 2008) around one mediocre campaign (-8.6% in 2007, still below replacement level).
We can try to predict Davis' future success (or failure) by looking at players who have put up similar statistics early in their careers. Here are players most similar to Davis, limited to those between 23 and 25 years old and coming out of their third seasons:
|Tight Ends Most Similar To Vernon Davis|
As FO honcho Aaron Schatz noted upon compiling this list, "yeah, that's a pile of suck." The ten players averaged just 29 catches for 311 yards and 1.3 touchdowns the following season. If we loosen our age restrictions somewhat, we get Daniel Graham as a good comparison for Davis, except that Graham has much more value as a blocker. The best comp for Davis may be Irv Smith. Like Davis, Smith was drafted in the first round (20th overall) from a big school (Notre Dame), and the two were very similar in size (6-foot-3, 249 pounds for Smith; 6-3, 250 for Davis). Smith's best year, by the way, was his third season. It was all downhill from there.
Almost nobody of any value. J.T. O'Sullivan, our lowest-rated quarterback of 2008, is a free agent and, um, doubtful to return. The 49ers were strong on both kick and punt returns last year, but Allen Rossum is a free agent. He ranked seventh in average kickoff return in 2008. Roderick Green is an interesting pass rusher. In nine games last season, he had 3.5 sacks -- but only seven tackles. But the stars of the team -- guys like Patrick Willis, Michael Lewis, Frank Gore -- aren't going anywhere.
As General Manager Scot McCloughan put it on the team's own Web site, "When it’s all said and done, we’ll be in the top 10 [in cap room] in the NFL for sure. We’ll have room and we’ll be one of the teams with the most room. We have enough room to possibly be aggressive and there’s enough room to take care of our own guys. Whatever we want to do, we’re going to do it. We have some cap room where we can improve our roster because of the room we have. We’re in very good shape."
With the quarterback position likely to be determined by a Shaun Hill-Alex Smith training camp duel, it's time to find someone on the receiving end of those throws. T.J. Houshmandzadeh is far and away the best available at the position, at least in free agency. A dream scenario would be to somehow pull Anquan Boldin away from the Cardinals, which would be an immediate upgrade for the 49ers and also strike a blow to their top division rival. If they decide to bring in competition for Davis, there are some good tight ends available, most notably Owen Daniels and Bo Scaife.
For the first time since 1999, the Seahawks head coach is not Mike Holmgren. Instead it is Jim Mora, former coach of the Atlanta Falcons. In the past decade, Tom Coughlin, Tony Dungy, Bill Belichick, and Jon Gruden have all won Super Bowls in their second jobs as head coaches, but in general, is it a good idea to hire a retread, or should teams go with a fresh face instead? The following table shows the records for all 57 coaches with more than one coaching job since the 1970 merger (including a handful of guys like Art Shell, who had multiple jobs with the same team):
|Coaches with Multiple Jobs, 1970-present|
|First Job||Second Job||Third Job||Fourth Job|
|Avg. Years||Win %||Avg. Years||Win %||Avg. Years||Win %||Avg. Years||Win %|
To make this clear, those 57 coaches averaged 5.1 years at their first job, and had a winning percentage of .545. In the next go-round, the average tenure lasted 4.1 years, and the aggregate winning percentage dropped to .479. Eleven coaches managed to find a third job, and three men somehow pulled off a fourth stint in the NFL. Can you name them? The answer's at the end of this column.
So in general coaches see a slight dip in their second tenure, but that's skewed by guys like Mike Ditka, who got a second job because he had a long, successful tenure in his first gig. Mora's success in Atlanta was moderate -- a 27-21 record, but just one playoff berth in three years. Let's limit our sample size to the 26 coaches whose first tenure lasted four or fewer seasons:
|Coaches with Multiple Jobs, First Job Less Than 5 Years|
|First Job||Second Job||Third Job||Fourth Job|
|Avg. Years||Win %||Avg. Years||Win %||Avg. Years||Win %||Avg. Years||Win %|
These coaches usually have a little longer to prove themselves, but they don't necessarily fare any better. Looks like there's no real advantage to hiring a coach with prior experience.
Bobby Engram, one of the most dependable Seahawks of the Mike Holmgren era, will likely try to end his career on a winning team -- i.e., not in Seattle. Leroy Hill is coming off a disappointing season and a marijuana arrest, lowering his demand both with the Seahawks, and with other teams. It would probably hurt Seattle more to lose him than it would help another team to sign him. The Seahawks have only $9 million in cap space; they could free extra room by cutting or re-negotiating with players like Matt Hasselbeck, Walter Jones, or Deion Branch.
This is a team in serious transition, and a lot of what they do in free agency depends on what they do with players like Hasselbeck and Jones. They have clear and serious needs at wide receiver and along the offensive line, but the one position that would lend the most immediate help may be a cornerback to play opposite Marcus Trufant. Last season the combination of Josh Wilson and Kelly Jennings failed woefully at that task over and over again. Bryant McFadden, all six feet of him, would be a welcome sight in Seahawk blue.
By the way, the three coaches with four jobs were:
45 comments, Last at 24 Feb 2009, 3:17pm by phildo