Adjusting quarterback win-loss records, part I
Posted by Doug on March 25, 2009
Last summer I wrote a post that rated field goal kickers according to their accuracy compared to league average. And I also adjusted each kicker's field goal percentage to account for the distance of his attempts. If one kicker attempted a bunch of chippies while another was kicking a lot more long ones, then we should take that into account when comparing their percentages.
This is highly non-controversial. Obvious even.
So we're going to do it again. But instead of kickers, we'll look at quarterbacks. Instead of field goal percentage, we'll look at winning percentage. And as our measure of difficulty, we won't use distance but points allowed by the quarterback's team. Just as kicking a 45-yarder is more difficult than kicking a 23-yarder, it's harder for a quarterback to win a game if his defense gives up 30 than if they give up 10.
Highly non-controversial, right?
So here's the plan, which mirrors the kicker plan exactly:
STEP 1: compute each quarterback's winning percentage in each of six categories: (1) defense allows 0--10 points, (2) defense allows 11--15 points, (3) defense allows 16--20 points, (4) defense allows 21--25 points, (5) defense allows 26--33 points, and (6) defense allows 34+ points.
STEP 2: in each category, compute how many games that quarterback won compared to how many an average QB would have been expected to win.
STEP 3: sum up the QB's wins above or below average, across all six categories.
Let's run through Joe Namath as an example:
defense allows 0--10 points: Joe was the beneficiary of this kind of defensive performance 17 times. An average QB would be expected to win 15.9 of 17 games. Joe went 17-0. So he's +1.1 wins.
defense allows 11--15 points: 18 games. Joe is 16-2. An average QB would win 13.7. So Joe is +2.3 here.
defense allows 16--20 points: 23 games. Joe is 13-10. An average QB would win 12.1. So Joe is +0.9.
defense allows 21--25 points: 25 games. Joe is 10-15. An average QB would win 9.7. So Joe is +0.3.
defense allows 26--33 points: 24 games. Joe is 6-18. An average QB would win 4.1. So Joe is +1.9.
defense allows 34+ points: 25 games. Joe is 2-23. An average QB would win 0.9. So Joe is +1.1.
Add it all up (and ignore the rounding error) and Namath comes out at +7.6. Do that for every quarterback who has started 50 or more games since 1950 and you get the following list. Lots of commentary and fine print below:
G W ExpW Diff
=========================================
Peyton Manning 191 124 92.8 +31.2
John Elway 252 162 134.5 +27.5
Tom Brady 128 101 73.9 +27.1
Brett Favre 291 181 154.7 +26.3
Dan Marino 258 155 129.8 +25.2
Joe Montana 187 133 109.1 +23.9
Ken Stabler 158 103 80.7 +22.3
Johnny Unitas 194 124 106.7 +17.3
Daryle Lamonica 97 70 53.1 +16.9
Jim Kelly 177 110 93.2 +16.8
Steve Young 157 102 85.8 +16.2
Roger Staubach 131 96 79.9 +16.1
Norm Van Brocklin 105 63 47.7 +15.3
Terry Bradshaw 177 121 106.3 +14.7
Dan Fouts 178 89 76.4 +12.6
Randall Cunningham 144 85 72.8 +12.2
Danny White 102 67 55.5 +11.5
Bobby Layne 139 83 71.6 +11.4
Kurt Warner 112 65 53.6 +11.4
Y.A. Tittle 139 78 66.8 +11.2
Frank Ryan 90 58 47.8 +10.2
Bill Nelsen 79 42 32.1 +9.9
Fran Tarkenton 250 130 120.1 +9.9
Otto Graham 78 61 51.3 +9.7
Stan Humphries 87 53 43.7 +9.3
Joe Theismann 132 83 74.3 +8.7
Jeff Hostetler 88 55 46.6 +8.4
Steve McNair 163 96 87.7 +8.3
Ben Roethlisberger 81 59 51.0 +8.0
Rich Gannon 139 80 72.3 +7.7
Steve Grogan 138 75 67.4 +7.6
Joe Namath 132 64 56.4 +7.6
Dave Krieg 184 101 93.6 +7.4
Drew Brees 109 56 48.7 +7.3
Matt Hasselbeck 112 62 54.8 +7.2
Bert Jones 99 47 40.1 +6.9
Jim Hart 182 87 80.1 +6.9
Trent Green 115 56 49.1 +6.9
Philip Rivers 54 36 29.2 +6.8
Jake Delhomme 89 55 48.2 +6.8
Marc Bulger 90 41 34.2 +6.8
Jay Schroeder 104 64 57.6 +6.4
Mark Rypien 85 52 45.6 +6.4
Eli Manning 78 46 39.8 +6.2
Ed Brown 98 55 49.1 +5.9
Charley Johnson 124 59 53.2 +5.8
Billy Kilmer 121 63 57.3 +5.7
John Brodie 164 76 70.6 +5.4
Don Meredith 89 49 43.7 +5.3
Warren Moon 213 105 100.2 +4.8
Brian Sipe 113 57 52.2 +4.8
Jack Kemp 111 67 62.4 +4.6
Jim Plunkett 154 80 75.4 +4.6
George Blanda 108 55 50.5 +4.5
Dan Pastorini 122 59 54.6 +4.4
Tony Eason 56 31 26.9 +4.1
Daunte Culpepper 99 43 38.9 +4.1
Bob Griese 162 98 94.1 +3.9
Sonny Jurgensen 149 69 65.1 +3.9
Troy Aikman 180 105 101.3 +3.7
Phil Simms 169 101 97.3 +3.7
Len Dawson 167 99 95.6 +3.4
Earl Morrall 108 67 63.6 +3.4
Neil Lomax 102 47 43.7 +3.3
Bart Starr 167 103 99.7 +3.3
David Woodley 58 37 33.8 +3.2
Boomer Esiason 178 83 79.8 +3.2
Jim Zorn 106 44 40.9 +3.1
Jim McMahon 103 70 66.9 +3.1
Michael Vick 71 40 36.9 +3.1
Jake Plummer 142 71 68.1 +2.9
Donovan McNabb 143 91 88.5 +2.5
Jeff Garcia 122 60 57.6 +2.4
Wade Wilson 74 38 35.6 +2.4
Charlie Conerly 92 58 55.7 +2.3
Marc Wilson 61 32 29.7 +2.3
Neil O'Donnell 107 58 55.7 +2.3
Brad Johnson 132 76 74.1 +1.9
Bobby Hebert 103 56 54.1 +1.9
Rodney Peete 89 46 44.2 +1.8
Brian Griese 83 45 43.3 +1.7
Mike Phipps 73 38 36.3 +1.7
Pat Haden 60 37 35.5 +1.5
Mike Tomczak 78 45 43.5 +1.5
Roman Gabriel 159 86 84.8 +1.2
Vince Ferragamo 59 30 28.8 +1.2
Babe Parilli 104 50 48.9 +1.1
Tom Flores 67 31 30.1 +0.9
Kordell Stewart 86 50 49.1 +0.9
Gus Frerotte 95 45 44.2 +0.8
Tobin Rote 119 51 50.2 +0.8
Elvis Grbac 73 41 40.2 +0.8
Carson Palmer 66 32 31.4 +0.6
Aaron Brooks 92 39 38.5 +0.5
Billy Wade 86 41 40.7 +0.3
Ken Anderson 178 93 92.7 +0.3
Craig Morton 154 86 85.8 +0.2
Doug Flutie 68 38 37.9 +0.1
John Hadl 169 82 82.0 +0.0
Jeff Blake 100 39 39.0 -0.0
Tommy Kramer 114 56 56.0 -0.0
Mark Brunell 160 83 83.1 -0.1
Erik Kramer 70 32 32.4 -0.4
Jay Fiedler 63 38 38.4 -0.4
Scott Mitchell 73 32 32.5 -0.5
Bernie Kosar 115 56 56.6 -0.6
Chad Pennington 83 45 45.8 -0.8
Don Majkowski 57 26 26.8 -0.8
Steve Bartkowski 131 60 60.9 -0.9
Bubby Brister 77 38 39.4 -1.4
Jon Kitna 116 46 47.5 -1.5
Richard Todd 112 50 51.5 -1.5
Milt Plum 103 56 57.9 -1.9
Chris Chandler 155 69 71.0 -2.0
Bob Avellini 51 23 25.8 -2.8
Gary Danielson 61 28 31.0 -3.0
Eric Hipple 58 28 31.1 -3.1
Ken O'Brien 112 50 53.2 -3.2
Jim Everett 158 66 69.3 -3.3
Bill Kenney 77 34 37.7 -3.7
Joe Kapp 52 26 29.8 -3.8
Jeff George 127 47 51.0 -4.0
Mark Malone 55 24 28.0 -4.0
Joe Ferguson 175 80 84.3 -4.3
Dave M. Brown 60 26 30.8 -4.8
Chris Miller 94 35 39.8 -4.8
Kerry Collins 171 82 86.9 -4.9
Mike Livingston 75 31 36.0 -5.0
Drew Bledsoe 199 101 106.1 -5.1
Greg Landry 99 44 49.1 -5.1
Eddie LeBaron 81 26 31.3 -5.3
Tim Couch 59 22 27.5 -5.5
Steve Beuerlein 104 48 53.6 -5.6
Cotton Davidson 54 20 25.7 -5.7
Lynn Dickey 113 46 51.7 -5.7
Rick Mirer 68 24 29.8 -5.8
Bob Berry 52 20 25.8 -5.8
Doug Williams 88 42 47.8 -5.8
Bill Munson 66 27 33.2 -6.2
Jim Harbaugh 145 68 74.7 -6.7
Bobby Douglass 53 16 22.8 -6.8
Trent Dilfer 119 63 69.8 -6.8
Mike Pagel 54 17 23.8 -6.8
Tony Banks 78 35 41.9 -6.9
Lamar McHan 73 24 31.0 -7.0
Steve DeBerg 144 54 61.2 -7.2
Ron Jaworski 151 77 86.4 -9.4
Norm Snead 158 52 61.6 -9.6
Joey Harrington 76 26 36.2 -10.2
David Carr 79 23 34.1 -11.1
Vinny Testaverde 219 92 103.4 -11.4
Archie Manning 139 35 52.4 -17.4
Fine print:
1. This includes all regular- and post-season games.
2. You might be wondering about era/league effects. It's easier to win if your defense allows 20 points in the 1961 AFL than it is to win if you defense allows 20 points in the 1974 NFL. Though I didn't mention it above, I actually did attempt to account for this. What I did was to compute the "average quarterback's expected wins" for a given category by looking at all games in that category in the same league within two years. For example, if I'm examining Joe Namath's 1966 season, the expected winning percentage for category (1) games, e.g., is computed by looking at all category (1) AFL games from 1964 to 1968.
Thoughts:
1. Don't forget that the line labeled "Peyton Manning" is really Peyton Manning, Marvin Harrison, Edgerrin James, Jeff Saturday, Tony Dungy/Jim Mora/Tom Moore, Mike Vanderjagt, and a cast of thousands. What it isn't, though, is the Colts' defense. Or at least not as much as Manning's raw record is. More on this later.
I've said before that quarterbacks don't win games (teams do), and I still believe that. But if people are going to talk about QB wins --- and it looks like it's unfortunately too late to put the lid back on that can of snakes --- they may as well try to put them into context. This post is is an effort to do that. To oversimplify things a little (or maybe more), Peyton Manning's record is the product of the efforts of 22 guys. This exercise attempts to narrow that down to 11.
2. I think this exercise has provided me with a new all-time favorite example of Simpson's Paradox. Check out Daunte Culpepper and Trent Dilfer:
+-------------------------------------------------------------------+
| Points allowed |
+------------------+-------------+-----------+------------+------------+------------+
| QB | 10 or under | 11--15 | 16--20 | 21--25 | 26 or more |
+------------------+-------------+-----------+------------+------------+------------+
| 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 |
+------------------+-------------+-----------+------------+------------+------------+
Culpepper has him in every category, but Dilfer (63-46) has a better overall record (Culpepper's is 43-56) because their distribution of opportunities has been so different. Matt Hasselbeck and Bob Griese are another example:
+-------------------------------------------------------------------+
| Points allowed |
+-----------------+-------------+------------+-------------+------------+------------+
| QB | 10 or under | 11--15 | 16--20 | 21--25 | 26 or more |
+-----------------+-------------+------------+-------------+------------+------------+
| Bob Griese | 53-1 0.981 | 20-7 0.741 | 14-11 0.560 | 7-11 0.389 | 4-31 0.114 |
| Matt Hasselbeck | 25-0 1.000 | 6-2 0.750 | 12-5 0.706 | 11-8 0.579 | 8-35 0.186 |
+-----------------+-------------+------------+-------------+------------+------------+
And you can pair Marc Bulger with just about anyone. Bart Starr, for instance:
+-------------------------------------------------------------------+
| Points allowed |
+-------------+-------------+------------+-------------+-------------+------------+
| QB | 10 or under | 11--15 | 16--20 | 21--25 | 26 or more |
+-------------+-------------+------------+-------------+-------------+------------+
| Marc Bulger | 8-0 1.000 | 8-1 0.889 | 8-4 0.667 | 9-6 0.600 | 8-38 0.174 |
| Bart Starr | 51-3 0.944 | 17-5 0.773 | 20-10 0.667 | 10-15 0.400 | 5-25 0.167 |
+-------------+-------------+------------+-------------+-------------+------------+
3. Starr is barely above average in this analysis. In Part II, where we make a few more adjustments, he'll actually come out as below average. What are we to make of this? Those hoping for a Starr-is-a-fraud rant at this point will be disappointed. Maybe it's just because of the soft spot I have for Troy Aikman, who finds himself ranked similarly. Or maybe it's because I know that points scored and points allowed are correlated. Bart Starr wasn't playing defense, but he and the rest of his offensive teammates could have, and probably did have, an indirect impact on the number of points the defense allowed. How much of an impact? That's tough to say, but that's the case you have to make if you think Starr is an all-time great, or even, frankly, an all-time good. What this does, in my mind, is eliminate the argument that Starr was good at doing just enough for his team to win. He wasn't any better at that than Marc Bulger or Tony Eason or Randall Cunningham.
Lest you think this is a canned stat-head rant, let me also call attention to Ken Anderson. If there's one thing that all historical football stat nerds seem to agree on, it's that Ken Anderson is underrated. But this analysis says otherwise, ranking him dead average.
Stay tuned for part II, where I'll use a completely different method to derive a similar list.
March 25th, 2009 at 5:01 am
Interesting. But this is problematic. I modern NFL the salary-cap ensures that teams that go all in and tries to build an offensive powerhouse, as to pay for it on defense by playing league-minimum-guys. This, to an extend, makes the defense/offense symbiotic - at leats in terms of talent. It certainly doesn't render this post useless, but it takes away some of the instant meaning. Right?
March 25th, 2009 at 5:52 am
Nice post - the top 6 passes the sniff test then you see some of the names at the bottom like Ron Jaworski who I wouldn't have rated so low by "feel".
I presume Fouts rates so high because the San Iego (intentional - there was no 'D' in Coryell's San Diego) "defense" gave him so many opportunities to win despite conceding 30+ points. Mind you it was great to watch...
Slightly OT - Coryell should be in the HOF. Seriously. No Coryell, no "Greatest Show On Turf" no Dallas Cowboys '90s dynasty. His offensive insight changed the game as profoundly as Bill Walsh (a former SD OC under Coryell). Get the poor guy in before he dies.
March 25th, 2009 at 8:48 am
My only reservation would be that a lot of times the score would move into higher brackets after the game is decided, such as when a team is trailing 27-10 and gives up a junk TD late in the game, thus moving from the "16 to 20 point bracket" into the "21 to 25 point bracket", thus skewing the results a bit. In this case, the QB wouldn't get credit for the loss he realistically would have earned in the "16-20 bracket". Therefore, one could only put so much stock into this theory. (Please note my change of email address since last post.)
March 25th, 2009 at 10:06 am
Fantastic stuff, I love this!
I'd be interested in seeing some current QBs, particularly Jay Cutler. His 17-20 career record is thrown out as a stat often, but he's had horrendous defenses.
March 25th, 2009 at 10:54 am
Great piece, Doug (and the Anderson finding is fascinating). One thing that struck me in the Dilfer/Culpepper comparison was the distributions of their defense. In the same number of NFL starts, Dilfer's teams held their opponent to 10 or less 35 times, Culpepper's team's did it just five times. I'm wondering if you think this raw data, even though it measures the QB's team and not the QB himself, is a useful metric?
March 25th, 2009 at 12:01 pm
Very interesting stuff, but I don't think we can overemphasize that "the line labeled 'Peyton Manning' is really Peyton Manning, Marvin Harrison, Edgerrin James, Jeff Saturday, Tony Dungy/Jim Mora/Tom Moore, Mike Vanderjagt, and a cast of thousands."
Good QBs on good teams mostly rank very well. Good QBs on bad teams mostly don't. Mediocre QBs on good teams do fine. And everyone without a lot of games played is pretty close to the middle.
I think it would be fascinating to do something like this with offensive/defensive coordinators, even over a relatively short period of time (since going back to 1950, or even the merger, probably isn't possible).
March 25th, 2009 at 12:49 pm
JT,
Unofficially (meaning this is me, and not Doug, and I've excluded post-season data), I've got:
Cutler being +0.1, +1.0, +2.5, -1.1, +0.5, +0.5 in the six categories, for a score of +3.3. He's 1/5 in the 21-25 range, when he probably should have won 2.1 games. Other than that, he's been above average.
Rivers is +6.9, Ben is +5.0, Eli is +6.0 and Romo is +6.5. Vince Young is +2.1, Grossman is +2.0. Losman, Harrington and Carr are -5.9 or worse.
Roethlisberger, for his career, is 31 games over .500 in 71 games (51-20). Cutler is 3 games under .500 in 37 games (17-20). And FWIW, over the last 3 seasons, Cutler is at +3.3 (duh) and Roethlisberger is at 0.0. So if you want to make the case that Ben is better at "winning games" than Cutler, it's got to be based on either his '04/'05 numbers or his post-season numbers. Because Cutler's got him beat in '06/'07/'08.
Also worth noting that Plummer was 17 games over .500 since '04 but just +3.3 here; McNabb was 16 games over .500 since '04 but just +1.0.
On the flip side, Bulger was 20 games under .500 but just -0.3; Kitna was 14 games under .500 but only -1.7.
March 25th, 2009 at 1:34 pm
I feel kinda bad for Culpepper. His teams allowed 26+ points in 47 of his 99 career starts?! Wait, what am I talking about--Bulger's teams have given up 26+ in 46 of his 90 career starts! I mean, I never thought either guy would be a legend, but that's just ridiculous.
And Lamonica cracking the top 10! I'm a Jets fan, but LOVE those AFL guys and I always thought Lamonica especially was very underappreciated.
March 25th, 2009 at 3:35 pm
Enjoyed the analysis, but I have to disagree with a few of the point ranges. I think you have to go with 16-19, because I've found that good QB's win almost every game when the defense gives up less than 20. Take your Namath example. He was 13-10 in 16-20 pt games, but 5 of those losses came in games where the opponent scored exactly 20. If you stop it at 19, that means the team can win with 20 pts and don't have to get 3 TDs that day (2 TDs and 2 FG's would work). When you make that jump to 20, that means a minimum of 21 to win, and usually it's going to be more than that. Using the boxscore search here, I came up with 62 games that ended 21-20 (for reference, 20-17 had 220 results and 23-20 had 122 results).
I would have also shortened the range 26-33 into 26-30 to allow for those 65 31-28 victories. I don't think 31 is a very uncommon point total for a winner.
It was mentioned, but it would be nice to see the numbers with adjustments made for DEF/ST TDs, but that is a lot of work. One day I'd love to know which QB has thrown the most pick 6's.
March 25th, 2009 at 4:57 pm
Thanks Chase, those numbers are interesting. I realize that it's harder to compare some of the QBs with shorter tenures, but it helps put some things in perspective.
March 25th, 2009 at 5:45 pm
You just turned Bart Starr into the perfect NFL QB, IMO. Most of us ask our starting QB to be good enough to get our Team INTO the Playoffs (it no longer matters what seed). But, once in, we want him to be great enough to lead our Team to a Championship. In other words, he should be better than ever in the Playoffs--he should "step it up" a notch now that it's win-or-go-home. And nobody did that better than Starr. His Passer Rating went from a mediocre 80.5 in the Reg. season to a Record setting 104.8 in the Post-season. That's the mark of a GREAT QB. Oh by the way, his Playoff Record is 9 wins and only 1 loss, which includes 5 Championships.
March 25th, 2009 at 8:36 pm
Here is a list generated by using the point ranges suggested by Scott (#9):
March 25th, 2009 at 9:12 pm
Doug, three thoughts:
1) Could you pull the season-by-season numbers for Anderson?
2) Could you post the single season leaders (and losers)? That always is interesting. I've got no guesses on who would be up there, except Brady '07 is at least going to be positive
.
3) Would it be hard to generate a career list using 100% of best season score, 95% of second best season score, 90% of third best season score, etc., using only positive score seasons? That might be the best way to present the data.
March 25th, 2009 at 9:20 pm
Interesting to note that Manning's bookend this study. What most surprises me is Dad being the worst, but by a lot--not just a few decimal points.
March 26th, 2009 at 12:30 am
My first thought was of touchdowns scored on an interception return. I guess if a QB can overcome that to win then he should get some credit, but should he get potentially more value for giving up these points? Were the points allowed used strictly points given up by the defense or was the final score used including points given up on special teams and on offense as well?
March 26th, 2009 at 8:07 am
#15: Beyond INT returns, turnovers in general have an impact on points allowed by giving the opposing offense better field position than they'd get from a punt. So the team's points allowed aren't entirely independent of the QB's performance.
With that said, this is fantastic.
March 26th, 2009 at 9:42 am
Great stuff as always guys. I've got 1 question and 1 request in regards to this post.
1. Do you think the list discriminates against bad-weather QBs? If your playing indoors and your defense gives up 14 it is "easy" to win the game with a good offensive performance. However, if your playing in a blizzard and your defense gives up 14, I would say it is much harder to win the game. Perhaps giving up 14 in the snow is the same as giving up 28 in a dome? This probably isn't a big deal for most players but I bet guys like Favre and Brady get hurt a little bit by playing so many games outdoors in bad weather, while Peyton plays almost all his games inside.
2. I would love to see this list compiled on a per-start basis. Maybe like wins per 16 starts to see what each QB is adding to his teams win total per season compared with the average quarterback. Just looking at the chart quickly, it seems a guy like Drew Bledsoe, who has lost 5 games over 199 starts should rank higher than a Mark Malone who lost 4 games in just 55 starts. Currently all of the really good QBs who have a lot of starts are at the very top of the list.
March 26th, 2009 at 3:55 pm
Does anyone had the details of Manning's numbers according to these standards?
+31 is ridiculous and I'd love to see his winning record under the various circumstances.
I want to also make sure I'm reading the article right. These numbers represent the averages of the QB's career? Meaning, we don't divide the plus number by the amount of games played? Because someone tried to tell me that Brady has a higher average than Manning if you do that, wasn't sure if he read the article wrong.
March 26th, 2009 at 6:01 pm
It looks like Brady does come out ahead by rate... You could convert this to a sort of winning percentage as follows:
Wins above average / Games + .500
Brady's "winning percentage" is .706, Manning's is .663. Both are excellent, obviously; if you extend those over a full season, Brady gets an extra .7 wins. Of course, all of that has to be interpreted within the context of their respective offenses.
March 26th, 2009 at 7:29 pm
In addition to turnovers, bad weather, and the quality of the rest of the offense, these ratings will also be influenced by pace. QBs on teams that have few possessions per game will be rated too low, since the low scoring will be mistaken for good defense, when in reality it's because neither team had many chances to score. That makes Peyton's #1 rank even more impressive, since the Colts have consistently been near the bottom of the league in possessions per game.
March 27th, 2009 at 12:04 am
During the 2003 through 2006 seasons, the Colts went 8-5 in games they gave up 30+ pts. That's pretty incredible. I'd also point out one of the losses was a week 17 game against Denver that Manning only played the first series in. So he actually was 8-4.
I can't imagine many people doing better over a 12 game stretch like that. In fact I know plenty of QB's that never even got 8 career wins in games the defense gave up 30+.
Here are some numbers (some may not be updated for 2008) on a few QB's I have in 30+ pt games
Drew Bledsoe: 4-28
Tom Brady: 5-8
Drew Brees: 2-20
Aaron Brooks: 1-20
Marc Bulger: 2-24
David Carr: 0-23
Kerry Collins: 1-40
Daunte Culpepper: 3-21
Jake Delhomme: 0-8
John Elway: 7-39
Brett Favre: 9-41
Dan Fouts: 10-42
Jeff Garcia: 2-30
Trent Green: 7-25
Joey Harrington: 0-15
Matt Hasselbeck: 3-18
Eli Manning: 0-14
Peyton Manning: 9-25
Dan Marino: 7-49
Donovan McNabb: 3-9
Steve McNair: 5-26
Joe Montana: 6-13
Carson Palmer: 3-14
Chad Pennington: 0-8
Philip Rivers: 1-5
Ben Roethlisberger: 2-9
Alex Smith: 0-11
Michael Vick: 4-12-1
Kurt Warner: 5-20
Steve Young: 4-20
March 27th, 2009 at 3:16 am
It would be nice to add a column to the list showing each QB's percentage of wins over or under the expected projection. For example, Van Brocklin is about 30% over, while Favre is 15% over, but Favre comes out higher on the cumulative list because he played games. Both sets of stats are interesting in their own respect.
March 27th, 2009 at 5:09 am
Here's one to consider--only 1 QB WON a Super Bowl game in spite of his Defense giving up OVER 30 points. The answer: Terry Bradshaw.
March 27th, 2009 at 7:37 am
ammek,
OTOH, Favre played at ages 23, 35, 36, 37, 38 and 39. NVB did not. Guys who played mostly during their prime years (Lamonica, at this point Brady, Danny White, NVB) will have an advantage in any rate statistic. This is not a cumulative stat like passing yards that you can pad even when you're not very good.
March 27th, 2009 at 2:03 pm
Van Brocklin actually retired directly after his best season in 1960 (league MVP, NFL champions, 10-2 record, 1st team all pro, career-high 24 TD passes and 86.5 rating).
March 29th, 2009 at 10:31 pm
Dear Doug,
Nice list which reflects most of the QBs who we feel are the best of all-time. Two other stats catagories I feel are important when rating QBs are comeback wins and home & away records. These two catagories tell us a lot about how a QB reacts to a pressure situation---either trailing in the fourth quarter or playing before a hostile crowd upon foreign soil. For example, Joe Montana had 31 comeback wins during his career---22 of which came on the road. In comparison, Brett Favre had 29 comeback wins, but only six came in road games. As another example, Montana had the second highest road winning percentage in NFL history---second only to Otto Graham. John Elway, on the other hand, had a dismal losing road record. Of course, these stats, like all others, don't always tell the whole story. For example, Vinny Testaverde, the losingest QB in NFL history, ranks fifth on the all-time comeback wins list with 30---18 of which were scored on the road---tying Peyton Manning for second place only behind Montana. Anyway, food for thought.
March 29th, 2009 at 10:49 pm
To remove the effect of garbage points, it'd be interesting to see how it changes (if at all) if one uses halftime score or score at the end of the 3rd quarter. That'd remove some of the "42-7" that ends up "42-21" situations
May 16th, 2009 at 12:46 pm
Mattie: That would also remove some of the best 4th quarter comebacks in history, a sign of a great QB for sure.
Also, TB played with some of the best defenses ever, which certainly makes me question this formula.
May 16th, 2009 at 2:39 pm
The only thing I would add about Peyton Manning is there are 3 or 4 games where he played like a series or a quarter and then didn't play the rest of the game because it was week 16 or 17 and the Colts were resting starters. By my eyeball calculations, it looks like those were included for him. Then you get into the question of how long a QB has to be in a game for it to be considered a "win" for him as opposed to the back up. IMO that 4 games, 3 of which the Colts lost, probably should be thrown out for him. However, then you have to consider those sorts of things for every other QB and have a cut off as far as how long a QB has to be in the get a "win"
Also, this confirms what I always believed about Troy Aikman : He is an OK QB, but very much a product of the system. Only plus 3 or so for a guy that won 3 SBs. Danny White, who I always liked when I was younger, came out better than him.
June 3rd, 2009 at 1:54 pm
This was pretty interesting, and I looked at some of the top ranking QB's profiles, and was wondering -- why did they not match up with the stats (wins & games played) listed there? (P. Manning & Elway being the 2 I noticed it on).
Manning on the list has 124 wins, but his profile (updated to 08) says 117 (among other differences)
October 20th, 2009 at 7:02 pm
I like this a lot, but it hasn't been normalized yet for total games played. Right now, QB's who played more games will have higher numbers than QB's who played fewer, but who were just as proficient at winning. Do this with your original series and what happens? Brady rises to 1st, at +37%. Peyton Manning is 2nd, at +34%. But Elway drops to 14th, at +20%, and Favre is 23rd, at +17%. Daryle Lamonica climbs all the way to 5th, at +32%. And so on. The list changes a lot when one normalizes the calculation.
January 30th, 2010 at 6:40 pm
Where do Marino and Montana fall, personal curiosity, in your normalized list Steve#31?