PDA

View Full Version : The Fifth Down: Building a Better Passer Rating


El Minion
10-21-2009, 08:08 PM
October 21, 2009, 8:30 AM
Building a Better Passer Rating (http://fifthdown.blogs.nytimes.com/2009/10/21/building-a-better-passer-rating/)
By TONI MONKOVIC
Introducing a new feature on the Fifth Down: a weekly ranking of quarterbacks based on a more sensible passer rating.

Let’s get the caveats out of the way. No passer rating is perfect. Statistics in a moving-parts game like football can be overrated. This system won’t measure whether a quarterback is a good rusher. And lastly, feel free to chime in about what you think would make the ultimate passer rating.

Luis DeLoureiro, a pharmaceutical consultant who lives in the Boston area, is a self-described ‘’schlep who is using a geeky Web site as a way to work through an early midlife crisis.” He wrote on his site, NFLStatAnalysis.net (http://www.nflstatanalysis.net/): ”I’ll never have any impact on the N.F.L. changing their QB rating calculation. But I have taken a stab at it, anyway.”

As DeLoureiro sees it, the passer rating is arcane, without the benefit of being accurate. Using a regression model, he has found that passing yards per attempt is the biggest driver of points, which is the ultimate aim of any quarterback.

In a telephone interview last week, he said the biggest problem with passer rating was its reliance on completion percentage. ”It favors dinkers and dunkers, someone who completes five 3-yard passes instead of one 15-yard pass,” he said.

He has written up his methodology and sent it to us. Here is an excerpt:

At the end of the day, a quarterback’s job is to score points. The model also suggested that interception percentage plays a role, but not nearly as great as Y/A.

TD % also correlates, but, in my mind, TD % is what Brian Burke at AdvancedNFLStats.com (http://www.advancednflstats.com/) would call an “intermediate outcome.” This means that a player who’s good at other things (e.g. moving the ball down the field) is likely to get more TD opportunities and, therefore, more TD passes.

Using the aforementioned regression model as a base, I have created a new rating system that I believe is a better indicator of a quarterback’s success.


The rating takes into account four factors, with different weights applied to each:

1. The quarterback’s yards per attempt (adjusted for sack yardage)
2. The opponent’s average yards per attempt in other games played in the current season (adjusted for sack yardage)
3. Interception percentage
4. Touchdown percentage

The Y/A statistics were given higher weights than Int %. The TD % was given only a nominal weight to reward exceptional games.

The biggest differences between this rating and the “traditional” one are:

1. Completion % has been removed as a factor
2. Significantly different weights assigned to the factors that were included
3. Quality of opponent was factored into the calculation
4. Sacks are measured.
(editor’s note: The ProFootballReference.com blog has a post on the degree to which quarterbacks are responsible for their sacks — can’t blame it all on the offensive line)

The Formula:

* Y/A Differential - Y/A minus the opponent’s average Y/A yielded (times a factor of 2.5 - as determined by the regression model) MINUS
* Int % TIMES 70 PLUS
* TD % TIMES 20

It can also be stated as:
((Y/A - OPP Avg Y/A) * 2.5) - (Int % * 70) + (TD % * 20)

Notes: The opponent’s Y/A is bounded at a maximum of 7.2 and a minimum of 5.2. This is done because a team’s average Y/A yielded could be greatly affected by a single game (I call this the “Peyton Manning” effect) — especially early in the season.

Note: You can see more of his method here (http://www.nflstatanalysis.net/2009/10/new-qoa-b-rating-explained.html).
Here is the first set of Luis’s weekly rankings:

1. Drew Brees 8.28
2. Peyton Manning 6.84
3. Eli Manning 5.47
4. Tony Romo 4.03
5. Ben Roethlisberger 3.79
6. Matt Ryan 3.74
7. Matt Schaub 3.47
7. Philip Rivers 3.47
9. Kevin Kolb 3.21
10. Tom Brady 2.81
11. Kyle Orton 2.61
12. Brett Favre 1.08
13. Joe Flacco 1.01
14. Aaron Rodgers 0.95
15. Kurt Warner 0.85
16. Chad Henne .038
17. Jay Cutler 0.35
18. David Garrard -1.22
19. Byron Leftwich -1.43
20. Carson Palmer -1.48
21. Seneca Wallace -1.52
22. Jason Campbell -1.92
23. Donovan McNabb -2.15
24. Matthew Stafford -2.28
25. Marc Bulger -2.74
26. Matt Hasselbeck -3.02
27. Chad Pennington -3.68
28. Shaun Hill -3.83
29. Kerry Collins -3.98
30. Daunte Culpepper -4.41
31. Mark Sanchez -4.45
32. Matt Cassel -4.62
33. Josh Johnson -4.85
34. Trent Edwards -5.04
35. Jake Delhomme -5.06
36. Kyle Boller -5.56
37. JaMarcus Russell -6.30
38. Brady Quinn -6.96
39. Derek Anderson -8.39

BroncoBuff
10-21-2009, 08:12 PM
37. JaMarcus Russell -6.30
38. Brady Quinn -6.96
39. Derek Anderson -8.39

Figures.

DenverBrit
10-21-2009, 08:38 PM
10. Tom Brady 2.81
11. Kyle Orton 2.61
12. Brett Favre 1.08

Nice sandwich! ;D