Week One QB Score and RB Score Rankings

The NFL season kicked-off this past week.  And with a new season we return to the weekly QB Score rankings. 

Two Changes

Two changes from last years rankings (which you can see on the page titled: QB Score Rankings for 2006).  The first change is the addition of running backs to the rankings.  Last January I introduced RB Score, and this metric will now be employed to evaluate my favorite position (growing up in the 1970s my favorite players were Tony Dorsett, Franco Harris, Earl Campbell, Walter Payton, and Billy Sims). 

The second change was noted in the following post a few days ago. 

A New QB Score

Yes, both the QB Score and RB Score metrics have been changed.  These changes were detailed in both the paperback edition of The Wages of Wins as well as in “Back to Back Evaluations on the Gridiron”, a paper published in Statistical Thinking in Sports (edited by Jim Albert and Ruud Koning).

Week One QB Rankings

Okay, with the changes noted, let’s get to the rankings.  Here are the QB Score rankings from Week One.

Table One: Week One QB Score Rankings

In looking at these numbers, we must first note the performance of an average QB.  Looking at all quarterbacks who attempted at least 224 passes in a season from 1994 to 2006, the average quarterback posts a

- 1.78 QB Score per play

- 0.153 Net Points per play

- 0.390 Wins Produced per 100 plays

- 80.0 QB Rating (NFL’s measure)

Looking over our list, we see that 18 of the 32 passers ranked (you must attempt 14 passes in a game to qualify for the NFL’s ranking), were above average in terms of The Wages of Wins measures (QB Score, Net Points, and Wins Produced). 

One of these, though, I question.  In the AP story describing the Lions crushing defeat of the Raiders (okay, it wasn’t crushing, but as a Lions fan you take whatever you can get), was the following sentence:

(Dewayne) White then stripped (Josh) McCown on the next play and recovered a fumble to set up a 14-yard score by Tatum Bell that iced the game.

I was listening to the game via NFL.com, but the Internet broadcast went down in the fourth quarter.  So I did not hear this play.  Still, it looks to me like Josh McCown, the Raiders quarterback, lost a fumble towards the end of the game.  But in the box score statistics, McCown is not credited with a lost fumble.  So his turnover total stands at two.  If we increase this total to three, then McCown – despite a QB Rating of 93.0 – was below average according to the Wages of Wins metrics (QB Score per play falls to 1.26).

McCown is not the only player where the QB Rating and QB Score disagree.  The NFL’s measure tells us that Chad Pennington was the second most effective quarterback in week one.  This is the same Pennington who lost 30 yards from sacks (only Brett Favre lost more) and only gained 5.4 yards per play.  Of the 32 quarterbacks ranked, sixteen gained more yards per play than Pennington.  Yes, he avoided turnovers, but he was not the second best quarterback in Week One.

One should note that while McCown and Pennington were overrated, Jay Cutler and Jason Campbell were underrated.  And I could talk about both players (or the amazing John Kitna), but I need to say something about the running backs. 

Week One RB Rankings

Here are the running back rankings from Week One.

Table Two: Week One RB Score Rankings

The quarterbacks were evaluated on a per play basis.  For running backs, though, the ability to keep carrying the ball is important.  So running backs are evaluated both in terms of rushing yards (which I think is the standard measure) and RB Score (as opposed to RB Score per play).

In looking over these numbers we need to – like we did with quarterbacks – start with the performance of an average player.  An average running back from 1995 to 2005 – who carried the ball at least 160 times in a season – averaged  

- 26.5 RB Score per game

- 1.30 RB Score per play

- 72.4 rushing yards per game

- 4.1 rushing yards per rushing attempt

When we look over the numbers we can see that fumbles lost are clearly a problem.  The 16 above average backs – in terms of RB Score per game – did not lose a fumble.  Conversely, no back with a lost fumble was above average in terms of RB Score per game or RB Score per play. 

This is especially clear when you look at Willie Parker, who gained 107 yards on 27 rushing attempts (and one reception).  When you factor in his lost fumble, though, his contribution to the Steelers efforts don’t look so good.  Of course, when you are playing the Browns, it doesn’t matter much if your running back has a bad game (and yes, it’s a bad sign when you trade your starting quarterback after week one). 

The Usual Disclaimer

The following observation concluded the post detailing the new QB Score and RB Score measures.  And it certainly is worth repeating…

The page detailing the final QB Score rankings for 2006 begins with the following disclaimer: “As we note in the book, statistics in the NFL do not necessarily represent a quarterback’s ability. So these evaluations should not been seen as conclusive evidence that one quarterback is “better” than another.”

Quarterbacks (and running backs) are quite inconsistent across time (unlike basketball players who are quite consistent across time).  This point about quarterbacks was made in the following posts:

Consistent Inconsistency in Football

Football Outsiders and QB Score

The Value of Player Statistics in the NFL

In The Wages of Wins we argue that the inconsistency on the gridiron is a reflection of the impact teammates and coaches have on player performance in the NFL.  A quarterback might have “good” statistics because he is “good”, his teammates are “good”, and/or he plays in a “good” system. The statistics cannot tell us who is ultimately responsible for the outcome we observe.  Hence the primary purpose of tracking statistics – it allows us to assign responsibility for outcomes – is not achieved in football.  This observation should always be kept in mind when we look at statistical evaluations of football players.

- DJ