We compared the QBs (stats prior to this season) with more than 2000 career attempts. Since sacks have only been an offical metric in the NFL since 1982 (http://www.nfl.com/history/randf/records/indiv/sacks), we decided to abandon our "improved" passing yards per attempt algorithm so we could compare every QB on the same basis. We also decided to utilize utility weights for each metric (all metrics were scale normalized: metric(i)/ max(metric)). [Note, because of how we merged the database, we actually made Bob and Brian Griese one person, sorry Bob.]
Our metrics are:
TD/ INT (weight .35) - given the qb throws a TD or an INT, the likelihood that he throws a TD.
TD/ ATT (weight .25) - the probability that a QB throws a TD given a pass play.
1 - (INT/ ATT) (weight .1) - the probability that a QB does not throw a INT given a pass play.
Comp/ ATT (weight .15) - the probability that a pass is completed given a pass play.
Yards/ ATT (weight .15) - the number of yards thrown given a pass play.
The utility weights are subjective (namely my opinion) and a sensitivity analysis is an important future step, but as the results show, either we've have more good QBs in the last few years or these statistics are skewed towards modern QBs. Which makes Steve Young and Joe Montana that much more impressive. Interestingly, Archie Manning (who gets a lot of undeserved credit since siring two NFL QBs) is ranked below Kordell Stewart. Additionally, players who are over rated because they had the best offensive line, best running back, and best WR in the game some how found their way down our list. Aikman is ranked 73rd. I am not sure why this metric is so skewed toward modern QBs. Any thoughts?
| Name | Norm Score |
| S. Young | 0.943 |
| P. Manning | 0.907 |
| J. Montana | 0.880 |
| D. McNabb | 0.865 |
| T. Brady | 0.846 |
| J. Garcia | 0.829 |
| K. Warner | 0.820 |
| D. Marino | 0.818 |
| F. Ryan | 0.818 |
| D. Culpepper | 0.813 |
| M. Bulger | 0.809 |
| L. Dawson | 0.807 |
| D. Brees | 0.800 |
| S. Jurgensen | 0.798 |
| R. Gannon | 0.795 |
| B. Favre | 0.791 |
| T. Green | 0.788 |
| M. Brunell | 0.782 |
| M. Hasselbeck | 0.781 |
| R. Staubach | 0.781 |
| R. Cunningham | 0.778 |
| N. O'Donnell | 0.773 |
| J. Kelly | 0.768 |
| D. Lamonica | 0.762 |
| N. Lomax | 0.761 |
| F. Tarkenton | 0.758 |
| D. Meredith | 0.755 |
| D. Krieg | 0.751 |
| S. McNair | 0.749 |
| J. Unitas | 0.749 |
| B. Esiason | 0.749 |
| D. White | 0.748 |
| E. Morrall | 0.745 |
| N. VanBrocklin | 0.738 |
| B. Griese | 0.737 |
| S. Beuerlein | 0.731 |
| S. Baugh | 0.730 |
| B. Jones | 0.729 |
| B. Johnson | 0.729 |
| M. Rypien | 0.725 |
| K. Anderson | 0.724 |
| B. Kosar | 0.723 |
| B. Starr | 0.723 |
| C. Conerly | 0.722 |
| J. Elway | 0.721 |
| W. Moon | 0.719 |
| J. Hostetler | 0.718 |
| J. George | 0.718 |
| A. Brooks | 0.717 |
| J. Blake | 0.717 |
| R. Gabriel | 0.716 |
| P. Simms | 0.715 |
| Y. Tittle | 0.712 |
| B. Kenney | 0.705 |
| D. Fouts | 0.704 |
| T. Bradshaw | 0.704 |
| J. Theismann | 0.702 |
| E. Grbac | 0.702 |
| C. Chandler | 0.701 |
| K. O'Brien | 0.699 |
| D. Flutie | 0.696 |
| J. Everett | 0.695 |
| B. Kilmer | 0.694 |
| C. Miller | 0.692 |
| K. Stabler | 0.692 |
| B. Hebert | 0.690 |
| S. Bartkowski | 0.690 |
| M. Plum | 0.688 |
| C. Morton | 0.687 |
| E. Kramer | 0.683 |
| S. Mitchell | 0.682 |
| D. Bledsoe | 0.681 |
| T. Aikman | 0.681 |
| B. Wade | 0.680 |
| B. Sipe | 0.680 |
| J. McMahon | 0.679 |
| R. Jaworski | 0.676 |
| C. Johnson | 0.676 |
| J. Brodie | 0.676 |
| J. Hadl | 0.675 |
| G. Blanda | 0.675 |
| S. Grogan | 0.675 |
| G. Frerotte | 0.671 |
| V. Testaverde | 0.669 |
| T. Kramer | 0.666 |
| W. Wilson | 0.662 |
| J. Schroeder | 0.661 |
| B. Layne | 0.660 |
| G. Landry | 0.658 |
| J. Kitna | 0.654 |
| J. Harbaugh | 0.652 |
| D. Williams | 0.652 |
| S. Humphries | 0.651 |
| L. Dickey | 0.649 |
| B. Parilli | 0.648 |
| S. Deberg | 0.648 |
| J. Plummer | 0.646 |
| J. Ferguson | 0.644 |
| B. Brister | 0.641 |
| J. Plunkett | 0.637 |
| J. Namath | 0.637 |
| K. Collins | 0.630 |
| N. Snead | 0.629 |
| T. Banks | 0.629 |
| M. Wilson | 0.627 |
| T. Rote | 0.626 |
| J. Hart | 0.623 |
| R. Todd | 0.622 |
| T. Dilfer | 0.620 |
| M. Tomczak | 0.614 |
| R. Peete | 0.606 |
| K. Stewart | 0.605 |
| D. Carr | 0.603 |
| R. Johnson | 0.599 |
| J. Harrington | 0.598 |
| J. Zorn | 0.595 |
| A. Manning | 0.583 |
| J. Kemp | 0.580 |
| D. Pastorini | 0.548 |
| R. Mirer | 0.521 |
1 comment:
You have some interesting values that I did not take into account in my analysis of QB data, but you may be interested in my results.
I took the 4 elements that make up the NFL passer rating (Completion %, Yards Per Attempt, Touchdown %, and Int %) and compared them on a year-by-year base to what the leagues numbers were.
So as the eras may change, the QB is not punished for it. Rule changes involving everything from defensive players being able to bump or where stick 'em or what have you now does not work against a player
So I came up with a percentage of efficiency as compared to the league, that a passer performs. 100% is average.
Joe Montana performed at a 120.04% for instance.
So then, Archie Manning performed at 96.80% of what the league performed during his tenure.
Kordell Stewart performed at 90.15 % of what the league did.
The highest on my list is Sid Luckman who performed at 128%. The lowest superbowl winning quarterback is Trent Dilfer who was at 89%
I created a chart using this data as my Y axis.
For my X axis I used their attempts so that we could see how long they performed at that level.
I adjusted this number to accommodate for differing eras as well.
What I did was take the average number of throws that a team executed for a given year and gave that quarterback a years worth of attempts if he matched that.
So if the average team had 200 attempts in a given year, and in that same year a player passed for 150 attempts, he would get credit for .75 of a year.
The third element that seems to have impact in comparing Quarterbacks (wrongly or rightly) is Championship wins.
The chart that I used was a bubble chart so that the X axis is Adjusted Attempts, the Y axis is Comparative Efficiency and the size of the bubble corresponds to championships.
The last thing I did was color code the bubbles. The Yellow ones are hall of famers. The red ones are eligible, but not in. And the white ones are not yet eligible. I did, however, outline the white ones with either red or yellow based on my best guess as to if they are hall of famers.
If you would like me to send you the Excel file just let me know.
It is not an end all be all, but it shows some definite patterns. Based solely on the chart and admittedly excluding important data as quality of teammates, leadership, etc there are five names that appear to have an argument on some level to be included in the hall.
Conerly, Stabler, Simms, Anderson, and Gabriel
Anyway, let me know if you want the file.
brandonparis2004@yahoo.com
Post a Comment