This article is the summarizing of a blog I read this morning. You can read that blog in its entirety at http://www.pro-football-reference.com/blog/?p=1808
The article begins with a simple premise; if you have two field goal kickers who make 80% of their kicks in a season are they effectively equal. The author proposes that they are not equal based on percentages alone.
He argues that a field goal made from 40-50 yards is more valuable than a field goal made from 10-20 yards. Based on that premise he broke down the career field goals of kickers from various distances and compared what they made/missed versus what the average kicker would have made/missed.
The argument seemed logical enough and generated positive responses, so this year he decided to apply the same logic to the quarterback position.
The author basically tries to remove the defensive performance of the team from the win-loss equation, breaking it down in much the same way he did for place kickers.
The argument given is that a quarterback that wins a game when the defense gives up three points shouldn't be weighted the same as a quarterback that wins a game when the defense gives up 26 points.
For the purpose of this breakdown, he divided the groups based on points given up by the defense. 0 - 10 points, 11-20 points 21-30 points etc....
He then took the win/loss record of the QB in those situations and compared it with the average win/loss record.
He uses Joe Namath as his example and states that Namath played in games where the defense allowed 0-10 points 17 times.
Based on the average a QB would be expected to win 15.9 games. Joe Namath won all 17 so for that category his is +1.1
This analysis goes to show that a team's win/loss record can be a very misleading statistic.
The best example given by the author is that of Trent Dilfer (63-46) and Daunte Cullpepper.(43-56) 0-10 11-20 21-30 31-40 41-50
Daunte Culpepper | 5-0 1.000 | 7-4 0.636 | 17-6 0.739 | 6-7 0.462 | 8-39 0.170 |
Trent Dilfer | 30-5 0.857 | 9-7 0.563 | 13-8 0.619 | 9-12 0.429 | 2-24 0.077
In this example, Dilfer has played in 35 games where the defense gave up less than 10 points and Culpepper has played in 47 games where the defense has given up more than 41 points.
Cullpepper has actually won a higher percentage of games in every category but has a worse win/loss record.
It becomes fair to suggest that Trent Dilfer has an inflated W/L record courtesy of of the fact that nearly half of his wins came when his defense kept the other team to 10 points or less.
As for the player at the top of this ranking. It's probably no surprise that Peyton Manning tops the list.
It is almost a given that throughout his career, Manning has lifted his team to victory when they otherwise probably had no business winning. In fact he won an MVP last year for doing just that.
Just as the author points out, you have to acknowledge that the Colts offense as a whole deserves some of the credit for that number. Despite that, you simply can't dispute the fact that if you put any other quarterback on the Colts from 1998 to today, the numbers would suggest the Colts would have won less games. As many as 31.7 games if an average QB was in his place.
Clearly there is a component to any NFL game that reaches beyond the numbers. A human element that you can't quantify or measure. That unbelievable play that turns a game and the momentum.
The statistical analysis offered by the author needs to be taken with a grain of salt. That being said, I think it does an excellent job of putting the win/loss record into perspective.
He correctly concludes that winning in the NFL is a team effort. We place way too much emphasis on the quarterback when a team wins or loses.
The author believes that if people insist on using that statistic to measure a QB, then you need to look further into the numbers to see the quarterback's actual impact on that number.
This article did an excellent job of that for me, I enjoyed reading it, and sharing my thoughts on it. I am interested to hear the thoughts of others who have read it, given that the topic discussed tends to be a hot button issue here at B/R.