Possibly the closest Super Bowl matchup in history also poses the question: how much does it mean when certain aspects of an NFL team improve dramatically in the second half of the season?
08 Mar 2005
by Aaron Schatz
The Buffalo Bills narrowly missed the playoffs in 2004 despite the league's best defense and special teams according to our Football Outsiders DVOA (Defense-adjusted Value Over Average) ratings (explained here). As long as those two units can stay close to their 2004 performance, all Buffalo needs to be playoff-bound in 2005 is a mild improvement from the offense. The Bills have a talented young receiver in Lee Evans and a talented running back in Willis McGahee who is a very good bet for improvement next season (most players need two seasons to fully recover from ACL surgery). At quarterback, the Bills recognized the limitations of Drew Bledsoe and turned over the starting quarterback position to last year's first round draft pick, J.P. Losman.
This all sounds promising except for one problem. After quarterback -- heck, possibly before quarterback -- the most important player on the offense is the left tackle. That's particularly true when you have a young quarterback who needs pass protection to keep him from getting his confidence shaken and a running game so the whole offense doesn't immediately fall on his shoulders. And the Bills have lost their left tackle, Jonas Jennings, signed by San Francisco very early in the free agency period.
How good was Jennings? He wasn't generally considered one of the best left tackles in the league, but at first glance the numbers indicate that he was by far the best member of the Buffalo offensive line.
Our adjusted line yards stats, which attempt to judge line performance by weeding out long runs from scrimmage, list Buffalo as a top five team rushing left in 2004, but one of the league's ten worst teams rushing to the right side or up the middle. There is some question here about why Buffalo was so good going left, however. In 2003, the Bills actually ranked worse going to the left compared to the other directions (though Jennings missed five games with injury). Now, look at the 2004 numbers split out between Travis Henry and Willis McGahee. The ranks given here represent where the Bills would rank if only this back's runs were considered. (The team's average going right is less than McGahee or Henry alone because third-stringer Joe Burns had a tendency to get stuffed on the right side when he managed to get the ball.)
|Buffalo Adjusted Line Yards by Direction and RB, 2004|
|Total ALY||Rank||Left ALY||Rank||Middle ALY||Rank||Right ALY||Rank|
|All Buffalo RB||3.50||22||4.32||4||3.31||25||3.17||26|
McGahee was far better running to the left than he was the other directions, but Henry really wasn't. Is this the result of Jonas Jennings blocking, or the specific tendencies of Willis McGahee? Probably a combination, but it is hard to know for sure.
There's no question, however, that the Bills will miss Jennings in pass protection. The conventional wisdom says Jennings has problems with speed-rushing defensive ends, but you would too if the quarterback you were protecting was Drew "The Monolith" Bledsoe. STATS, Inc. keeps numbers on sacks allowed (by "keeps", I mean "keeps fairly hidden from the public") and Jennings, protecting Bledsoe's blind side, is listed with two sacks allowed in 2003 and four in 2004. Over at right tackle, Mike Williams is listed with 9.5 sacks allowed in 2003 and 8.5 allowed in 2004. Yikes.
Who will replace Jennings? The Bills signed Mike Gandy from Chicago, but he's been more injury-prone than Jennings and is a more natural guard than he is a tackle. This is where the rumored trade of Henry to Arizona for tackle L.J. Shelton comes in, but that trade has gone from hot stove to gazpacho.
Since this is the last of the first round of Four Downs, many of the "upcoming free agents" are now "former free agents" and rumored cuts have become actual cuts.
Besides Bledsoe and Jennings, the Bills have lost defensive tackle Pat Williams, who played beside Sam Adams on the inside and has gone to Minnesota to help the Vikings corner the market on tackles named Williams. He has plenty of run-stopping talent, but he's getting old (32) and expensive, and only played on two-thirds of downs in Buffalo. The Bills drafted Tim Anderson in the third round of last year's draft, and he's the likely replacement, particularly since the Bills need to begin filtering younger players into their veteran defense.
The issue here isn't just age, but free agency, as the contracts on Buffalo's defense represent a ticking time bomb set to go off in 2006. After next season, more than half the starters on Buffalo's defense become unrestricted free agents, including Adams, cornerback Nate Clements, linebackers London Fletcher and Jeff Posey, safety Lawyer Milloy, and defensive end Chris Kelsay.
Buffalo's free agency needs, besides a replacement for Jennings, include a tight end and a backup running back if Henry is traded. But all those choices ahead in 2006 make it difficult for Buffalo to be active in this year's free agency market. The Bills have made one signing, but it is an odd one. Grabbing a veteran backup for Losman, they signed Kelly Holcomb to a four-year $6.6 million contract, with a $2 million signing bonus. It's not an odd move for the Bills, who needed a strong backup in case the rookie falters or gets injured while the defense is leading a playoff run. It's an odd move for Holcomb, who reportedly could have made more money to start in Cleveland.
The new Nick Saban administration had two possible courses of action going into this offseason. They either could decide that, since Miami's defense was still strong last season, the Dolphins just needed upgrades at running back or offensive line in order to become contenders again. Or, they could decide that since "playoff contenders" doesn't necessarily mean "Super Bowl contenders," a longer-term plan to transform the team in Saban's image was worth another year or two of poor performance.
The Dolphins have clearly decided to follow the latter path, revamping a defense that didn't need any revamping. Thanks to their free agent signings, the Dolphins now have an overabundance of talented defensive linemen. Unless a couple of these linemen can play on the other side of the ball, this is not going to make them a better team in 2005.
Miami has made three free agent signings so far this offseason. Two of those signings are defensive linemen (the other is safety Travares Tillman). Kevin Carter's comes over from Tennessee with a contract that will pay him at least $10 million during the first two years and could be worth $16 million during the first three. Vonnie Holliday arrives from Kansas City, signing a two-year deal with minimum salaries but a $1 million signing bonus and plenty of incentives. He'll try to regain the talented form he showed in Green Bay before disappearing into the defensive black hole that is Arrowhead Stadium.
The problem here is not that Carter is not plenty talented (he is) or that Holliday doesn't make a good risk for improvement outside Kansas City (he does). The problem is that Miami's defense was still very good last season. Some of those players have left in free agency (goodbye, linebacker Morlon Greenwood and safety Arturo Freeman) but all four of last year's starting defensive linemen are still here. Saban plans to play the defense in both 3-4 and 4-3 alignments, which means that the Dolphins either have two or three extra starting defensive linemen right now.
It is clear that Saban wants to build a defense that plays in his style with personnel that he has chosen. There's nothing wrong with that, and, if Saban follows in the footsteps of mentor Bill Belichick, these decisions will pay great dividends down the road. But Miami fans need to be patient and give Saban some time, because the first stop on that road is very likely another last place season filled with 20-10 losses.
The general belief among NFL observers is that Miami will fill its running back hole on draft day, where the Dolphins have the second overall selection and there are three running backs expected to go early in the first round: Ronnie Brown and Cadillac Williams of Auburn, and Cedric Benson of Texas.
That's bad news for Edgerrin James, the franchised Indianapolis tailback who still holds out hope of a return to the city where he played in college. Signing James would require relinquishing two first round draft picks or working out a trade with the Colts, and neither of those seems very likely.
Bruce Stram, in his series on Caponomics, pointed out that a veteran free agent is often a better value than a top draft pick, as top draft picks are far greater risks that still require paying superstar-level salaries. On the other hand, James will be 27 years old in the upcoming season, and running backs generally hit their peak at age 28. Paying James a high salary this season is a very good idea. Paying James a high salary four years from now may not be such a good idea.
James would disagree. He told USA Today, "The thing is I've got experience. I got the experience at an early age. I came in at an early age and I've only had one injury. Back then it was a major (knee) injury, but you see that you can come back. I really don't have much wear and tear I guess because of the way I play the game."
No disrespect meant to James, but this is why you don't put players in change of personnel decisions. James is subject to the same physical forces as all running backs. He ran 387 times in 2000 and like most backs who get that many carries in a season, he got injured the following year and it was not until 2004 that he finally returned to his pre-injury level. He may get less wear and tear playing in Indianapolis, where he has excellent blockers and faces defenses more concerned with the pass. If he came to Miami, "the way he plays the game" would significantly change. Just ask Ricky Williams.
One more note on the Miami offense: There have been reader comments on this site mentioning that the Dolphins might get a boost next year from the return of injured wide receiver David Boston. After failing his physical, however, Boston was cut yesterday.
Any franchise that seeks sustained dominance is going to eventually have to make the decision to let go some of the higher-priced veterans who led the way to the top. New England is no exception. Five different players who have been in New England for all three Super Bowl titles have left the team since they walked off the field in Jacksonville on February 6. Guard Joe Andruzzi and WR David Patten were free agents, and were offered larger contracts by Cleveland and Washington, respectively. LB Roman Phifer, CB Ty Law, and WR/CB Troy Brown were cut for salary cap reasons.
Brown's release was the most surprising and the most difficult for New England fans. A diminutive wide receiver taken in the no-longer existent eighth round of the 1993 draft, Brown has been a scrappy fan favorite since Bill Parcells was the head coach. But he never personified the franchise the way he did in 2004, as his willingness to switch from offense to defense in mid-season symbolized New England's "anything for the team" persona.
The Patriots had Brown practice with the secondary in pre-season as a precaution, in case they ever faced a scenario where injuries left the team without enough defensive backs to put on the field. After the Patriots lost their top three defensive backs at midseason, that unlikely scenario became reality. For the rest of the year and through a successful run to the Super Bowl championship, Brown served as the nickel back even though he had never played defense in his life. Though he sacrificed his role as New England's third receiver, he still played offense for a few plays each game and also returned punts.
Critics have attacked Brown's release as proof that New England's "team first" concept is really a mirage that masks a heartless front office. But Brown's release really proves that it doesn't matter how heartless or sentimental a front office may be, because you can't pay everyone. The salary cap is a zero sum game, and rewarding one player requires the merciless treatment of others.
Brown almost wasn't a member of the Patriots at all in 2004. He was set to become a free agent last year but renegotiated his contract to guarantee his 2004 salary and provide New England with cap relief. The new contract gave Brown a $2.5 million salary for 2005 with a $2.5 million roster bonus, basically made-up numbers that the Patriots never meant to pay Brown next season and Brown knew he would never get.
The question, "Should owner Robert Kraft express his gratitude to Brown for his selfless play by giving him a $5 million bonus" is a different question than "Should the Patriots front office express their gratitude to Brown by letting him take up $5 million of the team's salary cap." We don't know the answer to the first question, but we know the answer to the second is clearly "no." $5 million would make Brown one of the highest-paid receivers in the league, and he's far from that.
Critics of the Patriots have pointed out that Brown's willingness to go along with the defensive experiment cost him the chance to meet bonus incentives in his contract. But those bonus incentives would never have been reached even if Brown had spent the season as a full-time receiver. Brown had incentives in his contract for 50 and 60 receptions; on the season, the Patriots had only one receiver, David Patten, with at least 50 receptions, and no receiver with 60. Brown, after missing five of the first six games of the season, would not have matched him unless he had caught five passes per game for the final ten games. Not likely.
Many of the same people criticizing the Patriots for cutting Brown have also criticized the Patriots for not giving Tom Brady a new contract with a higher salary more befitting his status as the quarterback who won three Super Bowls by the age of 27. In reality, clearing Brown from the salary cap paves the way for Brady to sign a lucrative extension some time in the next few weeks. One fan favorite must suffer so that the other can profit.
Reports are mixed on whether Brown has taken this move personally, but it is very difficult to imagine him going to another team, because so much of his value is particular to his situation in New England. If Brown signs with another team, he will be learning a new offense; you can't expect him to learn a new offense and a new defense at the same time, which takes away his value as a two-way player. What you are left with is a declining 34-year-old slot receiver who missed the first half of last season with injury, asking for a salary more in line with a younger starting receiver.
As you might expect, rumors have Brown signing with his old coach Bill Parcells in Dallas or his old defensive coordinator Romeo Crennel in Cleveland. But signing Brown for starting receiver money makes sense for neither team. In Dallas, Brown would be fourth on the depth chart behind Keyshawn Johnson, Quincy Morgan, and Terry Glenn. In Cleveland, he would probably fall behind Andre' Davis, Antonio Bryant, and Dennis Northcutt.
If Brown does not return to New England, only four players will remain from the team that lost Super Bowl XXXI to Green Bay back in 1997: Kicker Adam Vinatieri and linebackers Ted Johnson, Willie McGinest, and Tedy Bruschi.
Bruschi, unfortunately, may not be coming back either. As readers know, Bruschi suffered a mild stroke roughly a week after the Super Bowl and the future of his career is uncertain. John Tomase of the Lawrence Eagle-Tribune wrote this article about Red Sox reliever Tom Burgmeier, who back in 1981 had a mild stroke similar to Bruschi's and came back to pitch the following season. But there is a great difference between a baseball pitcher and a football linebacker whose job is to smack people with his body on every play.
The Bruschi situation poses an interesting conundrum for those Patriots fans still obsessed with the idea that their team is underrated. We know that when some analysts (including me) pointed to the absence of Pro Bowl defensive tackle Richard Seymour as a reason why the Indianapolis Colts might beat the Patriots in this year's playoffs, many New England fans held this up as an example of "disrespect." When those NFL analysts make their preseason predictions for 2005, some who do not pick the Patriots to win a third straight title will mention Bruschi's absence as one of the reasons behind their choice. Will Patriots fans once again claim that this is a sign of "disrespect"? By doing so, they will basically be saying that Bruschi is an easily replaceable player. Those who believe that Bruschi was one of the most important players on the Patriots will have to acknowledge that picking one of the NFL's 31 other teams as the Super Bowl favorite is in part an acknowledgement of Bruschi's greatness, not a sign of disrespect for New England.
The main loss among the free agents is probably not former Pro Bowler Law, who the Patriots will try to replace with trade acquisition Duane Starks, but unheralded offensive lineman Andruzzi. The loss of Andruzzi leaves three guards on the roster: last year's starting right guard Stephen Neal, 2003 Super Bowl fill-in Russ Hochstein, and unknown Gene Mruczkowski. Will the Patriots go out and get an offensive lineman to replace Hochstein? Not necessarily. As Mike Reiss of the MetroWest Daily News points out, the Patriots tend to keep undrafted free agent offensive linemen on the practice squad for one or two years in order to teach these players the "Patriots way" of blocking. That leads to players like Neal and Hochstein being successful when finally inserted into the lineup. Mruczkowski may be the latest product of Patriots lineman school to move into the starting lineup.
The other major departures of the offseason, of course, are defensive coordinator Romeo Crennel (now Cleveland head coach) and offensive coordinator Charlie Weis (now Notre Dame head coach).
After a bit of a bidding war between Belichick and his disciples in Cleveland and Miami, the Patriots did manage to keep defensive backs coach Eric Mangini, believed by many to be the next coaching genius of the Belichick line. Mangini will now move up to defensive coordinator.
It appears that the Patriots will go without an offensive coordinator, splitting that role between numerous offensive assistants with Belichick likely calling offensive plays himself while one of those offensive assistants (possibly 29-year-old receivers coach Brian Daboll, according to Michael Felger of the Boston Herald) is trained to be the future coordinator.
What are the Patriots looking for in free agency? They've got their cornerback replacement in Starks, so the next defensive need is a linebacker in case Bruschi cannot return. Most people expect the Patriots to take one or two linebackers early in this year's draft as well. The Patriots are also in the hunt for a wide receiver to replace the departed Brown and Patten and had Derrick Mason in for a visit before he signed with Baltimore.
The Jets' attempt to deal for wide receiver Laveranues Coles has gone from on-again to off-again so many times that Jets fans may be wondering if Gang Green has confused Coles with Randy Johnson. After this weekend, though, it looks like the trade to bring the prodigal son back from Washington in exchange for receiver Santana Moss is back on, as long as both players can pass physical exams by tomorrow.
There is no question that in bringing back Coles, the Jets are adding a receiver whose speed and elusiveness contributed mightily to Chad Pennington's initial success as the starting quarterback in 2002. But there is significant evidence that the only differences between Coles and Moss are circumstances and personality. When it comes to performance over the past two seasons, Moss has actually been the better receiver.
Coles had 90 catches and 950 yards in 2004, while Moss had 45 catches and 838 yards. But Coles rang up those higher numbers because he was thrown more passes than Moss - he averaged only 10.6 yards per catch, compared to 18.6 yards per catch for Moss. Coles also caught fewer of the balls thrown to him (54%) than Moss (58%). And 35 of Moss's 45 catches were worth a first down or touchdown, compared to only 52 of Coles's 90 catches. According to our DVOA ratings, Moss ranked eighth among NFL receivers in value per play. Coles ranked no. 71.
Moss had a better year than Coles in 2003 as well, though the difference between the two was much smaller. Coles caught 82 passes for 1,204 yards, while Moss caught 74 passes for 1,105 yards. But Moss once again had a higher average of yards per catch (14.9 to 14.7), a higher percentage of passes caught (63% to 52%) and a higher rank in DVOA (16th compared to 36th for Coles).
True, Coles should not be blamed for the quarterback problems in Washington, where veteran Mark Brunell saw his abilities disintegrate before he was mercifully pulled from the lineup halfway through the year. Nor can he be blamed for the conservative nature of the Joe Gibbs offense. But if we give Coles the benefit of the doubt on these issues, we must do the same for Moss.
Moss's 2004 campaign clearly would have been even better if not for the rotator cuff injury that robbed Pennington of his ability to throw deep. Moss's yards per catch were cut nearly in half after Pennington's return from injury in Week 13, and he caught only two passes of 20 yards or more over the season's final five weeks, compared to 13 such catches in the first 11 games.
Additionally, Coles's Redskins threw 75 more passes than the Jets in 2004 and 31 more passes in 2003. And the Redskins play three or four wide receivers far less often than the Jets, so Coles is splitting responsibility in the passing game with fewer teammates.
|1st/TD|| 20+ Yd
|Moss, with Pennington pre-injury||31||18||381||58%||21.2||14||8||41.4%|
|Moss, with Carter/Bollinger||17||12||275||71%||22.9||11||3||82.3%|
|Moss, with Pennington post-injury||30||15||182||50%||12.1||10||2||-2.5%|
|Coles, with Brunell||78||35||417||45%||11.9||26||4||-23.6%|
|Coles, with Ramsey||87||54||518||62%||9.6||25||6||-10.6%|
|*Includes three option passes from Portis and Gardner|
Another overlooked facet of Moss's value is his role as a punt returner. Though he had an off-year in 2004, his punt returns over the past three seasons have been worth approximately 23 points of field position when compared to the NFL average; only Dante Hall has been more valuable in that role.
None of this is to say that Coles's numbers will not improve dramatically thanks to his reunion with Chad Pennington. Back in 2002, his last season with the Jets, Coles ranked fifth among NFL receivers in DVOA (Moss, as the third receiver on that team, ranked 22nd).
But why were the Jets in such a hurry to trade Moss in the first place? Moss is often referred to as "undersized," yet his listed height (5-foot-10) and weight (185 lbs.) are only one inch and eight pounds smaller than Coles. Critics allege that Moss is fragile, but the hamstring strain he suffered early in 2004 is not the kind of medical problem that constantly recurs, and his 2001 knee injury is about as meaningful to his 2005 performance as the case of chicken pox he had when he was 12.
I know that Al Bogdan, in Four Downs: NFC East, ran three-year similarity scores for Coles and Moss that showed a dramatically better set of comparable players for Coles. But the similarity scores are still somewhat a work in progress. Because they are based on standard statistics like yards and catches, they are subject to all the biases that DVOA attempts to filter out. That doesn't make them worthless, but they are just one piece of evidence.
You can also run the similarity scores for different career lengths: a single year, a two-year span, or a three-year span. Running Moss for three seasons, the list of comparables take into account that Moss was a part-time receiver in 2002. What if we run a list of comparables for Moss based on just two years? This would be a list of receivers who had a big star season early in their careers but then a down season afterwards with fewer yards and catches, though not fewer yards per catch. To make the list even closer to Moss, we'll limit it to players coming off their third, fourth, or fifth season (Moss is coming off his fourth season).
|YEAR ONE||YEAR TWO|
Fernandez, Connell, and Johnson disintegrated. Jackson is pretty good. Taylor had two more good seasons, 1991 and 1993. Hill, Hilliard, and Logan were mediocre. Owens you may have heard of. This list makes Moss look like a better player than the other list -- although not by as much as I expected. Is it a more accurate list? No, both are just pieces of evidence for the discussion.
I don't think that Moss is about to become Terrell Owens, but I do think that his numbers have been held down by circumstance. When Pennington has thrown him the ball, he has been one of the best receivers in football. On a per-play basis, he has been as good as Coles was in 2002. They just seem to call fewer plays for him.
Essentially, the Jets have traded one receiver who was upset with the conservative nature of his offense for a receiver of roughly equal talent who was upset with the conservative nature of his offense. The only advantage for the Jets is that Coles has a history of getting along better with the quarterback.
How much is that history worth in dollars? Moss is due to receive only $448,000 in salary for 2005, but the Redskins are expected to announce a contract extension sometime soon that will significantly raise his salary -- and the cap space he takes up. As for Coles, the Redskins will take the cap hit for a $5 million bonus he is due on April 1, but the Jets have agreed to guarantee $15 million in salary for Coles over the next three seasons.
If Coles costs the Jets only slightly more than Moss costs the Redskins, the Jets can safely argue that team chemistry and an attempt to recapture the magic of 2002 are worth a small premium. But if Moss's extension is notably smaller than the money the Jets have guaranteed to Coles, they will have exchanged two receivers of similar quality at a significant financial cost. Come December, Jets fans may find themselves wishing that the team had used that cap space on a cornerback or a tight end.
(Note: If you are wondering why this article did not appear until late Tuesday afternoon, I was waiting on announcement of Moss's new contract, which was originally expected today but now may take a few days or even weeks.)
The idea that the Jets could convince running back LaMont Jordan to accept a smaller contract to remain as Curtis Martin's backup was always a bit far-fetched. But the extent of the fantasy became pretty obvious once Oakland inked Jordan to a five-year deal worth $27.5 million, including a $7 million signing bonus.
The Jets moved quickly to replace Jordan with Kansas City backup running back Derrick Blaylock, who got a five-year deal worth $11.1 million, including a $3.2 million signing bonus. Blaylock rushed for 539 yards and eight touchdowns last year and on a per-play basis represented very little drop-off from starter Priest Holmes (all three Kansas City running backs had roughly similar DVOA ratings). He may be the perfect replacement for Jordan, used to playing behind a better-known back but just entering his prime after four years of experience. Any worries that his strong numbers might be a creation of the Kansas City offensive line are tempered by similar questions about whether Jordan's numbers might be a creation of the equally strong New York offensive line.
Replacing Jordan was easier than expected, but the same cannot be said for replacing defensive tackle Jason Ferguson. Ex-Giants tackle Lance Legree will be a step down. By franchising defensive end John Abraham and allowing Ferguson to leave, the Jets made a conscious choice of pass rush over run defense. While Abraham is talented, I don't believe this was the right choice. Run defense was the heart of the Jets defense this year, and the pass defense would be better served by putting some money into the secondary.
Ferguson and Jordan were just two of many free agents in New York this offseason. RT Kareem McKenzie traded places with Legree, moving to the crosstown Giants. His replacement is unknown. Backup quarterback Quincy Carter is probably gone as a free agent. Rumors have Vinny Testaverde returning to New York to once again back up Chad Pennington, although you wonder why the Jets are not going after ex-Miami quarterback Jay Fiedler, whose style tends towards higher accuracy passes and minimizing mistakes to let the running game and defense win for him ("Pennington/Brady Lite"). Well, minimizing mistakes when you have an actual offensive line blocking for you, which Fiedler did not have the last two seasons in Miami.
Starting TE Anthony Becht is an unrestricted free agent, his backup Chris Baker a restricted free agent. To replace Becht, the Jets signed Denver tight end and restricted free agent Jeb Putzier to an offer sheet. If Putzier comes to the Jets, it will mark one of the greatest days in the history of Yiddish-speaking Big Apple sports fandom.
A New York Jets offseason preview on the website scout.com wrote the following about new offensive coordinator Mike Heimerdinger:
"Under Heimerdinger, the running game is expected to remain the focus of the offense, but the passing game will undergo a significant transformation. Heimerdinger uses a vertical version of the West Coast offense, which should be a welcome addition in the New York locker room."
This is, I believe, makes it official: the phrase "West Coast offense" is now completely useless. Of course, the original, actual West Coast offense of Sid Gillman was a vertical passing game, but the phrase has now come to mean the opposite, an offense of precise short- and medium-range timing patterns. Which would make "vertical version of the West Coast offense" either an oxymoron or an incredible dose of nostalgia.
Next week: NFC West by Mike Tanier