Finally the Final 2008 Football Rankings

Posted on January 25, 2009 by

6


Just prior to the last game of the 2008 NFL regular season, Brian Burke – of Advanced NFL Stats (one of the best statistical sites on the web) – posted the following comment:

Dave Berri from Wages of Wins posted his weekly QB and RB rankings based on econometric models like my own. Note who the top QB is–Pennington. Dave may have an updated ranking by the time you read this. He also adds his thoughts on an ongoing discussion between amateur/internet sabermetricians and academic researchers.

I read this comment just as I was getting ready to put together my rankings for Week 16.  And then I had a thought.  Why not wait until the last game is played and then just post the Final Rankings for the 2008 season?

With that thought in mind I went and did something else.  And that was followed by a few more weeks of me doing “something else.”  Now the Super Bowl is just a week away and I think it is time I finally get around to posting those Final Rankings:

Table One: Final Quarterback Rankings for 2008

Table Two: Final Running Back Rankings for 2008

Brief Ranking Thoughts

Here are some thoughts on these rankings (and be warned, there are quite a few thoughts):

  • Pennington – the quarterback the Jets discarded in favor of Brett Favre – finished third in the rankings. Aaron Rodgers – who the Packers kept in favor of Favre – finished 11th (and was also above average). Favre, though, in what may be his last season, finished 27th. Only five quarterbacks who qualified for the rankings (minimum 224 pass attempts) offered less. Oddly enough – and I will touch on this again in a moment – Favre, not Pennington or Rodgers, was asked to go to the Pro Bowl.
  • One of the quarterbacks ranked below Favre was Ben Roethlisberger. Roethlisberger is also not going to Hawaii. Of course he will play in the Super Bowl next week. The opposing quarterback – Kurt Warner -finished in 9th place.
  • Speaking of Warner…Kerry Byrne at SI.com recently argued that Warner is a better quarterback than Peyton Manning (my co-author Stacey Brook – who know has an excellent blog of his own – sent me this link).  Byrne argues – via the NFL’s quarterback rating – that each player is essentially the same in the regular season. In the playoffs, though, Warner has done better. If we shift our focus from quarterback rating to Wins Produced (described in The Wages of Wins) we see that Manning’s career stats are worth 44.3 wins. Had Warner played as much as Manning – but maintained Warner’s per play production of wins – he would have produced 38.4 wins. Yes, this is close. But I think Manning has put up better numbers in the regular season. As for the playoffs… it is a much smaller sample. But I think you can say Warner has played better in the post-season.
  • Byrne is correct that both Warner and Manning’s career performances rank among the all-time greats. If we look just at the performances of 2008, we see that Manning did a bit more than Warner. In fact, Manning’s Net Points per Play of 0.262 ranks 38th among all quarterbacks (with 224 pass attempts in a season) since 1994. The following four performances from 2008 also rank in the top 40 since 1994.
    • Drew Brees: 0.303 Net Points per Play (ranks 10th)
    • Philip Rivers: 0.295 Net Points per Play (ranks 11th)
    • Chad Pennington: 0.284 Net Points per Play (ranks 20th)
    • Matt Ryan: 0.261 Net Points per Play (ranks 4oth)
  • Interestingly (perhaps only to me), three of these top 40 performances were not rewarded with an invite to Hawaii. In fact, of the seven quarterbacks invited, only four placed in the top 10 in 2008 in Net Points per Play. There three Pro Bowl participants who posted less than spectacular numbers includes Kerry Collins (ranked 12th), Eli Manning (ranked 17th), and Favre (see his low ranking above).
  • Table Two also ranks running backs in Net Points per Play. The top four backs listed – Derrick Ward, DeAngelo Williams, Maurice Jones-Drew, and Steve Slaton-were not originally invited to Hawaii. In fact, of the top eight running backs, only Chris Johnson is Pro Bowl bound.
  • When we look at the NFC running backs — Adrian Peterson, Michael Turner, Clinton Portis – we don’t see a single back in the top 10. Portis was actually below average on a per play basis.

Looking Back on the Lions

Okay, enough about the successful players in the league.  As a fan of the Lions I thought I would share some thoughts on the most miserable season I ever had a chance to witness (and as someone who got DirectTV just prior to the 2008 season, I got to witness almost every quarter of every game).

  • There is much talk that the Lions are going to select a quarterback with the first pick. Let me try and offer some perspective on such a choice. Last year the five quarterbacks with the Lions – Jon Kitna, Dan Orlovsky, Daunte Culpepper, Drew Stanton, and Drew Henson – participated in 589 plays and produced 1.5 wins. Peyton Manning also participated in 589 plays and produced 4.05 wins. This means that if Manning came to the Detroit – and posted exactly the same numbers he did with the Colts in 2008 (and that would be unlikely)- the Lions would only be 2.5 wins better. In other words, I am not convinced finding a “franchise” quarterback should be the focus of the Lions on draft day.
  • If quarterback is not the answer, where should the Lions look? Obviously the answer is Michael Crabtree. Currently, Scouts Inc. – at ESPN.com – ranks Crabtree as the number two prospect in the 2009 draft (with a grade that equals Andre Smith, the current top prospect). Crabtree is a 6-3 Wide Receiver, who caught 97 passes this past season for Texas Tech (including the pass that denied Texas the National Championship). The Lions already have Calvin Johnson, so Crabtree would give the Lions two major threats in the passing game. And since you can’t double team both players, one would always face single coverage and the Lions offense would be unstoppable.
  • Of course I am kidding. The Lions – under the direction of Matt Millen – already demonstrated the folly of trying to employ two number one receivers. And this year the Dallas Cowboys offered the same demonstration. There is only one football in the game. As Dallas discovered, if Terrell Owens gets all the passes thrown his way that he wants, Roy Williams will be unhappy. And if the Jason Witten gets what he wants, then Owens is unhappy. In sum, the idea that “if one great receiver is good then two must be better”, ignores the nature of wide receivers and football. On any given pass play, at most one receiver gets to be happy. The other receivers know that their pay is related to the number of passes they catch. And when the ball is not thrown their way, these other receivers are not happy to see their future earnings decline. And unhappy receivers tend not to help your team win games.
  • The failure of the “multiple number one receivers” strategy was but one failure in the regime of Matt Millen. Millen inherited a team in 2000 that was 9-7. In 2001 the Lions finished 2-14. Although the team did improve to 7-9 by 2007, the disaster that was 2008 put an exclamation point on the least impressive performance by a general manager in the modern history of the NFL. Let’s quickly review some of the major failures:
    • Millen hired three coaches in his tenure – Marty Mornhinweg, Steve Mariucci, and Rod Marinelli. All three had the same last initial as Millen (something I noted in 2007). By itself, that is odd (although, as James – a WoW Journal reader noted – there is evidence for “initialism”). All three, though, had another trait. Mornhinweg and Mariucci favored the West Coast offense. Marinelli favored the Tampa Two defense. All three made every effort to fit whatever talent they were given to their preferred scheme. In other words, rather than devise strategies that maximized the production of the talent on the roster, these three forced whatever talent they had into apparently the only strategy they knew.
    • Of course it might have helped if Millen provided these coaches with the talented players they needed. Millen’s record in the draft, though, was not exactly good. From 2001 to 2007, Millen drafted 53 players. The first two of these choices (from the 2001 draft) – Jeff Backus and Dominic Raiola – were still starting for the Lions at the conclusion of the 2008 season. Of the next 51 choices, only five were still starting when this last season ended. And none of these starters had ever gone to the Pro Bowl. It’s important to remember that Millen – because his team was so bad – often had very high draft choices. Despite this advantage, Millen often guessed wrong. And I think the word “guess” is appropriate. It did not appear Millen had a systematic approach to talent evaluation.
    • Millen supplemented his failures on draft day with an ability to sign unproductive free agents. This was highlighted by the disturbing pattern during Marinelli’s tenure of providing a home to people who used to play for the Tampa Bay Buccaneers. Perhaps the best example is Brian Kelly, who the Lions signed to be a starting cornerback in 2008. Kelly was so productive that he was cut from the roster before the season was over. Yes, Kelly was cut from a 0-16 team that finished last in defense.
  • In general, sport tends to reward success and punish failure. Certainly that is true in each individual game. Millen, though, had defied this tendency. Within days of Detroit finishing with the most regular season losses in NFL history, Millen signed with NBC. His new job is the same as his old job. Yes, Millen is going back to telling America all he knows about football. Although the last eight years would suggest this is not really Millen’s area of expertise, NBC has decided to ignore all the empirical data. What is amazing is that the other “experts” in the studio with Millen have not yet shown any skepticism with respect to Millen’s comments.

Looking Ahead with the Lions

There appear to be two approaches to the analysis of sports.  Millen appears to exemplify the approach that relies on personal experience and physical observation.  In other words, Millen did not strike me as a person who did any kind of serious data analysis.  And as we can see, Millen was not very successful.

The latest coach in Detroit – Jim Schwartz – appears to take a different approach to football.  Last November, Judy Battista of the New York Times wrote a profile — Titans’ Against-the-Grain Defense – of Schwartz.  Here is part of that story:

Schwartz, now the defensive coordinator for the Tennessee Titans, had an economics degree from Georgetown University, an abiding fascination with statistics and a preference for watching game film over television. That made him a kindred spirit with his first N.F.L. boss, Bill Belichick. But when Schwartz told Belichick his findings from an early N.F.L. research project almost 15 years ago, Belichick said he did not believe him.

“Fumbles are a random occurrence,” Schwartz said he told Belichick. “Being able to get interceptions or not throw interceptions has a high correlation with good teams. But over the course of a year, good teams don’t fumble any more or less than bad teams. Bill didn’t agree. He said, ‘No, good teams don’t fumble the ball.’ But actually, they fumble just as often as bad teams.”

With the Titans, Schwartz once encouraged the former offensive coordinator Norm Chow to run more on third-and-short because his research indicated that it was more effective than passing.

Unorthodox thinking like that has earned Schwartz, 42, a reputation as one of the N.F.L.’s leading practitioners of statistical analysis – “Moneyball” for the shoulder-pad set – using them in coaching the defense for the league’s only unbeaten team.

The Battista profile suggests Schwartz is cut from a different cloth from Millen (and Marinelli).  Millen often argued that winning is an attitude.  If your team simply believed it was going to win, it would be successful.  Schwartz seems to think that winning is not about your attitude, it’s about the choices you make during the game (and the talent you employ).  If you make better choices, you will be more successful.

That thinking is consistent with his choice of defensive and offensive coordinators.  With each choice Schwartz noted that he is not interested in a coordinator bringing in a specific scheme. 

“I don’t think you hire scheme,” Schwartz said. “I think you hire coaching ability. You mentioned some other coaches, and I think they came with a scheme. When you hired Coach X, you were getting that scheme. I’d like to steer away from that a little bit.

In addition – unlike the coaches hired by Millen – Schwartz will not be overly focused on one side of the ball.  Mornhinweg and Mariucci were offensive coaches and tended to ignore defense.   Marinelli was a defensive coach first and devoted most of his attention to that side of the ball.  Schwartz argues that it is up to the coordinators to manage the offense and defense.  The job of the head coach is to manage the game.  

All of this should give some hope to fans of the Lions.  Systematic analysis tends to trump the subjective approach.  And it looks like the Lions understand that point.

All that being said…. we must remember that systematic analysis in football is extremely difficult.  Unlike basketball and baseball – where player performance is largely independent of his teammates (at least for hitters in baseball) – the performance of football players depends on the performance of the player’s teammates.  These interaction effects severely hamper the objective analysis of the game. 

And that can be seen when we look at how much time I devote to discussing football.  Like I did with basketball, I also have a model to measure performance in football.  But the measurement of performance in football really only tells one story.  The interaction effects in football cause the performance statistics to be inconsistent.  So the players we see perform well today are not necessarily going to perform well tomorrow.  Although I like telling that story, it’s really about all I ever say about the NFL.

Consequently, this very long post (anyone still with me?) might be my last post on football.  At least, it will be my last post on football until the Lions win a game (or I change my mind). 

– DJ

The WoW Journal Comments Policy

For more on the Wages of Wins football metrics see

The New QB Score

Consistent Inconsistency in Football

Football Outsiders and QB Score

The Value of Player Statistics in the NFL