≡ Menu
One of the two greatest quarterbacks of the first half of the 20th century

One of the two greatest quarterbacks of the first half of the 20th century.

The comments to Parts I and II of this series have been great, so let me start with a thank you. One of the more difficult parts of this process is comparing players across eras not just for efficiency, but for gross volume. In 2013, teams averaged 38.0 pass attempts (including sacks) per game, compared to just 24.5 in 1956. A great quarterback will be above average in either era, but it’s easier for great quarterbacks to accumulate above-average value when they play in a high-dropback era.

So what’s the solution? Simply pro-rating the numbers feels a bit too dramatic; we got into a similar issue with True Receiving Yards, and our solution there was to take a (literal) middle ground approach. I thought it would be fun to apply the same philosophy here. Over the course of the 96 league seasons in this study, the average number of league-wide dropbacks per game was 26.1. If we were going to do a 1:1 adjustment, we would then multiply each quarterback’s value in 2013 by 0.687, since that’s the result of 26.1 divided by 38. Instead, I decided to split the baby, and take the average of 0.687 and 1.000, which means modifying the VALUE metric for each quarterback in 2013 by 84.4%. On the other hand, a quarterback in 1956 now gets his VALUE multiplied by 103%, and a passer in 1937 sees his score multiplied by 129.0%.

The table below shows the revised single-season leader list. Here’s how to read it, which will explain why Dan Marino climbs back ahead of Tom Brady into the top spot on the list.  Under the old system, Marino had a value of 2,267 yards above average, but with the modifier, he gets downgraded to an adjusted value of 1981; of course, Brady’s modifier is more severe, which is why Marino vaults him.  Meanwhile, thanks to a 110.3% modifier, Sid Luckman’s 1943 season1 jumps ahead of Peyton Manning’s 2004 season, which has a modifier of 88.1%.  The table below shows the top 200 single seasons using this formula.

RkQuarterbackYearLgValueModifierAdj Val
1Dan Marino1984NFL226787.4%1981
2Tom Brady2007NFL227086.8%1971
3Sid Luckman1943NFL1739110.3%1918
4Peyton Manning2004NFL213288.1%1877
5Dan Fouts1982NFL194688.2%1716
6Otto Graham1953NFL187990.7%1704
7Cecil Isbell1942NFL1450113.8%1651
8Aaron Rodgers2011NFL187985.9%1615
9Steve Young1992NFL177990.2%1605
10Peyton Manning2006NFL173588.1%1528
11Peyton Manning2013NFL178084.4%1502
12Bert Jones1976NFL156495.3%1491
13Tom Brady2011NFL169685.9%1457
14John Brodie1970NFL151794.5%1434
15Ken Anderson1975NFL151693.7%1419
16Sonny Jurgensen1961NFL150493.9%1413
17Tom Brady2010NFL163286.3%1409
18Drew Brees2011NFL161585.9%1388
19Johnny Unitas1964NFL145292.4%1341
20Peyton Manning2003NFL152288%1339
21Tom Brady2009NFL153886.8%1335
22Steve Young1994NFL152286.6%1318
23Tom Brady2012NFL153285.2%1306
24John Brodie1965NFL138293.1%1287
25Ken Stabler1976NFL134595.3%1282
26Sammy Baugh1947NFL1241102.4%1271
27Daunte Culpepper2004NFL143888.1%1267
28Milt Plum1960NFL132995.2%1265
29Norm Van Brocklin1953NFL139590.7%1265
30Randall Cunningham1998NFL143087.5%1251
31Johnny Unitas1959NFL128696.7%1244
32Daunte Culpepper2000NFL143286.9%1244
33Johnny Unitas1967NFL135191.8%1240
34John Hadl1967AFL138489.2%1234
35Dan Fouts1981NFL139388.5%1232
36Roger Staubach1979NFL133791.6%1225
37Mark Rypien1991NFL137389.1%1224
38Johnny Lujack1949NFL129094.2%1216
39Ken Anderson1974NFL127295.5%1215
40Fran Tarkenton1976NFL127295.3%1213
41Ken Anderson1981NFL136988.5%1211
42Drew Brees2009NFL138386.8%1201
43Jim Hart1976NFL124895.3%1189
44Roger Staubach1978NFL124395.5%1187
45Jeff Garcia2000NFL136486.9%1185
46Philip Rivers2009NFL135586.8%1177
47Steve Young1998NFL134487.5%1176
48Peyton Manning2000NFL135086.9%1173
49Drew Brees2013NFL138484.4%1168
50Joe Montana1984NFL131887.4%1152
51Sammy Baugh1945NFL1026111.9%1149
52Roman Gabriel1973NFL115798.9%1145
53Dan Fouts1980NFL127289.7%1140
54Joe Montana1989NFL129787.9%1139
55Sid Luckman1941NFL990114.9%1138
56Cecil Isbell1941NFL988114.9%1136
57Joe Namath1967AFL127389.2%1135
58Matt Schaub2009NFL129986.8%1128
59Roger Staubach1977NFL115597.6%1128
60Donovan McNabb2004NFL128088.1%1127
61Tony Romo2009NFL129386.8%1123
62Sonny Jurgensen1967NFL122491.8%1123
63Carson Palmer2005NFL127487.9%1119
64Boomer Esiason1988NFL126488.5%1119
65Neil Lomax1984NFL127787.4%1116
66Tommy Thompson1948NFL1113100.3%1116
67Y.A. Tittle1963NFL119293.2%1111
68Ken Stabler1974NFL115895.5%1107
69Greg Landry1971NFL114196.7%1103
70Kurt Warner1999NFL128085.9%1100
71Drew Brees2008NFL124988%1100
72Steve Young1993NFL125087.8%1097
73Joe Theismann1983NFL124088.3%1095
74Drew Brees2006NFL122788.1%1081
75Joe Namath1968AFL117691.8%1079
76Fran Tarkenton1974NFL112895.5%1077
77Ed Danowski1935NFL798134.9%1076
78John Brodie1961NFL114693.9%1076
79Don Meredith1966NFL118390.9%1076
80Peyton Manning2007NFL123186.8%1069
81Peyton Manning1999NFL124485.9%1069
82Dan Marino1986NFL122487.3%1068
83Peyton Manning2009NFL122786.8%1066
84Brett Favre2009NFL122486.8%1063
85Bart Starr1966NFL116590.9%1059
86Kurt Warner2001NFL121387.2%1058
87Vinny Testaverde1998NFL120287.5%1052
88Daryle Lamonica1968AFL114691.8%1052
89Joe Montana1983NFL118888.3%1049
90Roger Staubach1971NFL107796.7%1041
91Len Dawson1966AFL116788.9%1037
92Norm Van Brocklin1950NFL111293.2%1037
93Frank Filchock1939NFL903114.1%1031
94Ron Jaworski1980NFL114789.7%1029
95Norm Van Brocklin1959NFL106196.7%1026
96Bernie Kosar1987NFL117187.5%1025
97Tom Brady2005NFL116587.9%1023
98Johnny Unitas1963NFL109593.2%1020
99Aaron Rodgers2010NFL117286.3%1012
100Johnny Unitas1958NFL107194.3%1010
101Sammy Baugh1942NFL886113.8%1009
102Peyton Manning2005NFL114787.9%1007
103Jim Everett1989NFL114187.9%1002
104Lynn Dickey1983NFL113488.3%1001
105Charlie Conerly1959NFL103496.7%1000
106Philip Rivers2008NFL113588%1000
107Tommy Kramer1986NFL114387.3%998
108Joe Montana1985NFL114287.2%996
109George Blanda1961AFL113887.4%995
110Aaron Rodgers2012NFL116385.2%992
111Steve Young1991NFL111189.1%990
112Peyton Manning2012NFL115985.2%988
113Charlie Conerly1948NFL985100.3%987
114Brian Sipe1980NFL110089.7%987
115Archie Manning1978NFL103195.5%985
116Richard Todd1982NFL111088.2%979
117Dan Marino1987NFL111887.5%979
118Sammy Baugh1940NFL849113.7%965
119Arnie Herber1936NFL713135.1%963
120Y.A. Tittle1953NFL106290.7%963
121Roman Gabriel1967NFL104991.8%962
122Len Dawson1968AFL104691.8%960
123Erik Kramer1995NFL112585.2%959
124Dan Marino1988NFL108288.5%958
125John Elway1993NFL108987.8%956
126Roman Gabriel1969NFL103292.3%953
127Fran Tarkenton1967NFL103791.8%951
128Steve Young1997NFL109586.9%951
129Trent Green2002NFL110486.2%951
130Trent Green2003NFL107888%949
131Sid Luckman1946NFL852111.3%948
132Boomer Esiason1986NFL108587.3%947
133Brett Favre1996NFL107986.7%936
134John Hadl1973NFL94198.9%931
135Dan Marino1992NFL103090.2%929
136Bert Jones1977NFL95197.6%928
137Philip Rivers2013NFL109884.4%927
138Sonny Jurgensen1970NFL97994.5%926
139Aaron Rodgers2009NFL106686.8%925
140Otto Graham1949AAFC870105.8%921
141Earl Morrall1968NFL96294.7%911
142Tom Flores1966AFL102188.9%907
143Joe Montana1987NFL103587.5%906
144Bobby Thomason1953NFL99890.7%905
145Bob Waterfield1951NFL95894.1%901
146Steve DeBerg1990NFL99990.1%900
147Tony Romo2011NFL104685.9%899
148Sid Luckman1945NFL799111.9%895
149Daryle Lamonica1969AFL97291.8%892
150Frank Ryan1966NFL97990.9%890
151Brett Favre1995NFL104385.2%889
152Ken Anderson1973NFL89898.9%888
153Warren Moon1990NFL98390.1%886
154Steve McNair2001NFL101387.2%884
155Billy Wade1961NFL94193.9%884
156Rich Gannon2002NFL102086.2%879
157Nick Foles2013NFL104184.4%878
158Fran Tarkenton1973NFL88398.9%874
159Eli Manning2011NFL101685.9%873
160Steve McNair2003NFL99188%872
161Otto Graham1947AAFC768113.5%872
162Daryle Lamonica1967AFL97289.2%866
163Otto Graham1952NFL94391.8%866
164Norm Van Brocklin1960NFL90595.2%862
165Brett Favre2007NFL99286.8%861
166Philip Rivers2010NFL99786.3%860
167Norm Van Brocklin1951NFL91494.1%860
168Craig Morton1970NFL90694.5%857
169Tommy Thompson1949NFL90794.2%854
170Sonny Jurgensen1966NFL93190.9%846
171John Elway1987NFL96187.5%841
172Daryle Lamonica1970NFL88994.5%841
173Fran Tarkenton1965NFL89993.1%837
174Scott Mitchell1995NFL98285.2%837
175Johnny Unitas1957NFL826101.2%836
176Boomer Esiason1985NFL95987.2%836
177Randall Cunningham1990NFL92890.1%836
178Dan Fouts1985NFL95487.2%832
179Otto Graham1955NFL86795.8%830
180Chad Pennington2002NFL96086.2%827
181Dan Fouts1983NFL92888.3%819
182Fran Tarkenton1970NFL86794.5%819
183Johnny Unitas1965NFL87893.1%817
184Glenn Presnell1933NFL578141.2%816
185Troy Aikman1995NFL95785.2%816
186Bobby Layne1958NFL86294.3%813
187Tony Romo2007NFL93486.8%811
188Fran Tarkenton1972NFL82298.6%810
189Billy Kilmer1975NFL85993.7%805
190Tom Brady2004NFL91388.1%804
191Y.A. Tittle1962NFL85893.6%803
192Johnny Unitas1960NFL84295.2%802
193Dan Fouts1978NFL83595.5%798
194Milt Plum1961NFL84993.9%797
195Joe Namath1972NFL80898.6%797
196Norm Van Brocklin1954NFL86991%791
197Joe Ferguson1975NFL84393.7%790
198David Garrard2007NFL90786.8%788
199Tommy Kramer1982NFL89388.2%787
200Brett Favre1997NFL90586.9%786

But let’s move on to the fun stuff: the career list, using the same 100/95/90 formula.  The table below shows where each quarterback ranked in Part II (the Old Value and Old Rk columns), along with where they rank using this adjusted dropbacks modifier (the New Val and New Rk columns).  I also included two other columns: using this new formula, I eliminated all seasons where a quarterback had below-average value.  So the New Val+ and Rk+ show how each quarterback fares when you only look at their positive (i.e., above-average) seasons.

QuarterbackFirst YrLast YrOld ValueOld RkNew ValNew RkNew Val+Rk+
Peyton Manning19982013127691111011111011
Tom Brady200020131006328710287422
Dan Marino19831999985038644386623
Fran Tarkenton19611978828957852478634
Steve Young19851999848747465576906
Johnny Unitas19561973788387417676925
Joe Montana19791994794866990769909
Dan Fouts19731987776096940869958
Drew Brees20012013794776877970947
Ken Anderson19711986691810643910657010
Otto Graham19461955650013637811637811
Norm Van Brocklin19491960662011618712620113
Sid Luckman19391950543019608113619314
Sammy Baugh19371952563218607114622812
Roger Staubach19691979599414571615605615
Brett Favre19912010650312564216581916
Sonny Jurgensen19571974589115548417563417
John Brodie19571973568317534918560918
Aaron Rodgers20052013578416497219509619
Len Dawson19571975486520438320451623
Y.A. Tittle19481964465525438121474621
John Hadl19621977472522432422483520
Philip Rivers20042013484321419923454222
Tony Romo20052013471423408524408530
John Elway19831998468724408525433926
Joe Namath19651977435426399226447224
Jim Hart19661984412030396527417928
Roman Gabriel19621977417928396028433127
Daryle Lamonica19631974419727388329414029
Cecil Isbell19381942334940384330384332
Bart Starr19561971404431375331396631
Boomer Esiason19841997417329365732441725
Arnie Herber19321945269054361233361238
Trent Green19972008394433344334364737
Ben Roethlisberger20042013396032343735352642
Warren Moon19842000384534341136370934
Donovan McNabb19992011382735336037365336
Kurt Warner19982009380936330738365335
Jeff Garcia19992009369537321539327944
Troy Aikman19892000355838310440378933
Steve McNair19952007355339309641342843
Bert Jones19731982325841308442354541
Charlie Conerly19481961310044305743356739
Bobby Layne19481962319642300744356340
Rich Gannon19872004317243274245303449
Bob Griese19671980283549273346290451
Tommy Thompson19401950265355265147280856
Daunte Culpepper19992009300045262748317546
Matt Schaub20042013299746260249297350
Jim Everett19861997290347255550308148
Greg Landry19681984257759250251289752
Vinny Testaverde19872007286848249552287153
Earl Morrall19561976265156249553269458
Carson Palmer20042013281250246254253762
Joe Theismann19741985275751245355282555
Jim Kelly19861996274552244956266659
Randall Cunningham19852001273553242557274457
Terry Bradshaw19701983257758237958285354
Mark Brunell19942011263557228759245565
Johnny Lujack19481951237562225660225669
Mark Rypien19882001252060224561262561
Ed Danowski19341941168481222362241866
Milt Plum19571969231364221563310947
Ken Stabler19701984222066218864323545
Bob Waterfield19451952217367218665263860
Matt Ryan20082013247961214166214174
Craig Morton19651982224365211967252664
George Blanda19491975235063208368198087
Billy Kilmer19611978215669207569253063
Bernie Masterson19341940160887202670202685
Chad Pennington20002010215868188471214175
Brad Johnson19942008214370185272227767
Steve Grogan19751990203573185073220572
Dave Krieg19801998209872184574212776
Michael Vick20012013213271183475207781
Frank Ryan19581970195976180276196088
Doug Williams19781989202074178877192691
Charley Johnson19611975179778174378212177
Bernie Kosar19851996197875173279227068
Danny White19761988184577163180185796
Bob Monnett193419381202981565811565110
Tommy Kramer19771990179079156482202984
Billy Wade19541966155689148383178998
Brian Sipe19741983161385148384223271
Steve DeBerg19781998166884148285207880
Phil Simms19791993168182148086191394
Doug Flutie198620051691801471871673106
Matt Hasselbeck19992013167683146988223770
Keith Molesworth1932193610061081444891444114
Jim McMahon198219961610861411901474113
Neil Lomax19811988160888140891216473
Bill Munson196419791425931368921709105
Ron Jaworski197419891502911348931645107
Wade Wilson198119981473921314941717104
Russell Wilson201220131514901284951284127
Joe Ferguson19731990134096125096208279
Jeff Hostetler198619971381941220971386116
Ace Parker193719469991091178981351121
Cam Newton201120131361951162991162133
Jim Zorn1976198711331021077100183097

Some thoughts:

  • Obviously the quarterbacks most harmed by this adjustment will be modern quarterbacks, but there’s not that much movement at the top.  Manning, Brady, Drew Brees, and Aaron Rodgers all see their values drop by at least 250, but that doesn’t mean too much.  Perhaps a more significant modifier is necessary?  Then again, are Brees and Rodgers already too low at 9 and 19, respectively?  One might quibble with Brees’ stats because of dome effects, but that should be handled outside of era adjustments, in my opinion.
  • What if we eliminate all negative seasons?  Ken Stabler moves up significantly.  You probably didn’t even notice it (and if you did, you didn’t comment about it), but the Snake ranked just 66th in Part II.  After adjusting for dropbacks, he only moved up to 64th, but by removing negative years, he jumps up to #45.  I’m sure fans of a certain age (or certain part of the Bay area) would be incredulous at such a low ranking for Stabler.  But Stabler was about as bad as you can be from 1980 to 1984: he ranked 41st out of 43 quarterbacks in ANY/A over that period, and had significantly more attempts than the two passers below him. In the ’70s, Stabler had a career ANY/A average of 5.23, which was pretty good; his work in the ’80s dropped his career ANY/A to 4.74, despite rising league averages during that time. A 45th-place ranking is much more representative of Stabler’s career; combined with his great postseason (see below), and there’s enough there for Raiders fans to be annoyed that the Snake isn’t in the Hall.  Of course, Lamonica gets my vote as best Raiders quarterback of all time.
  • Troy Aikman falls from 40th overall to 33rd once you remove negative seasons. In his middle nine seasons, Aikman averaged 6.19 ANY/A.  But in his first, second, and final seasons, he produced an ANY/A between 3.0 and 4.0 each year, and had between 250 and 400 attempts. As a result, his career ANY/A is “just” 5.66.  When we think of Aikman, we tend to think of Troy Aikman, not the guy who was miserable in his first, second, or final years, so a 33rd place ranking (as may be boosted by his playoff production) is probably more appropriate.
  • Of course, for some, putting Aikman 33rd is still too low.  We’ll get to the playoffs in a moment, but let’s look at Aikman’s regular season numbers.  We all know his gross numbers are not impressive, but a run-heavy offense shouldn’t have harmed Aikman’s efficiency numbers. And for the eight-year period from 1991 to 1998, Aikman posted an excellent ANY/A of 6.25, which gave him an ANY/A+ index of 114.  In ANY/A+, a quarterback gets credit for his standard deviations above average, with one standard deviation being equal to 15 points, and average being equal to 100.   So Aikman was essentially one full standard deviation above average during that period. Which is really, really good. He may not have posted outstanding gross numbers, but he was very efficient. Of course, over the last eight years, Tony Romo has had an ANY/A of 6.96 and an ANY/A+ of 115.

Of course, the real “issue” with Aikman isn’t his efficiency numbers, but that his playoff performances were outstanding, which are ignored by the above analysis. So let’s take a look at the postseason.


I’ve done a couple of quantitative studies about quarterback play in the postseason. Joe Montana has been, quite clearly, the most productive postseason passer in NFL history.

After him, Terry Bradshaw, Kurt Warner, Peyton Manning, and Aikman probably round out the top five of best postseason quarterbacks. The next tier would be a big one, and would include Brett Favre, Steve Young, Stabler, Drew Brees, Bart Starr, Jim Plunkett, John Elway, and Tom Brady. Quarterbacks like Daryle Lamonica, Aaron Rodgers, Mark Rypien, Roger Staubach, Dan Marino, Ken Anderson, Len Dawson, and Phil Simms aren’t too far behind, either.

Bradshaw, Warner, Aikman, Stabler, and Plunkett are probably the ones most investigating.  Bradshaw ranks just 54th in my formula of the best regular season quarterbacks, but the 54th best regular season quarterback + the 2nd best playoff quarterback is enough to get you to the Hall of Fame.  And while Bradshaw’s numbers look bad now, they aren’t nearly as bad as they seem.

Warner’s numbers look great, but he doesn’t fare all that well in this formula.  There are a couple of reasons for that.  He faced one of the easiest schedules of any quarterback (average of -.21), so his raw numbers are overstated (and that’s before you get to the dome/supporting cast argument). He also rarely played full seasons, which limited his ability to produce significant value. Only two of his seasons cracked the top 150, and Warner also didn’t have many “good” seasons, as he tended to either be very good or average (or worse). His incredible playoff numbers means he’ll make it to Canton, and that’s probably appropriate, but I think he tends to be overrated by many analysts.

Aikman is like a softcore version of Bradshaw.  He’s not as extreme as Bradshaw, but he fits the “outstanding playoff numbers, just okay regular season numbers” paradigm. He’s a worthy Hall of Famer — at least, statistically — because he’s got good enough regular season stats and then superlative playoff numbers.  But figuring out how to compare him to Romo to Staubach is too tall a task for a summer Friday.

Finally, we get a pair of Raiders quarterbacks.  Both were great in the playoffs, and combined to win three titles.  Plunkett is mostly a non-starter when we take about all-time great quarterbacks: Plunkett has just three seasons (’74, ’86, ’77) where he produced more than 100 yards of value above average, and zero where he had more than 350 yards. Jeff Hostetler was a better quarterback if you want to really focus on mediocre quarterbacks (who played for the Raiders!) who won a Super Bowl. As for Stabler, he lags behind Aikman in the regular season and Aikman and Bradshaw in the playoffs. Not every quarterback with great playoff numbers gets to make the Hall of Fame, but it just so happens that three of the best (including Lamonica) were Raiders. Stabler was better than someone like Mark Rypien, but there’s a pretty big gulf between players like Rypien and the Hall of Fame. Perhaps Raiders fans will feel less bothered when Eli Manning doesn’t make the Hall of Fame (right now, Manning’s argument is terrible; ignoring the “will he” question, he’s a “not even close” on the “should he” question).

As for the top of the list? It’s pretty clear to me that Peyton Manning‘s the greatest quarterback of all time. If you don’t agree with that, well, there’s probably nothing I can say to convince you otherwise.

  1. Note that there is already a 25% deflation rate built into all seasons during World War II. Luckman’s numbers that year were insane. The Bears averaged 9.2 ANY/A, while the rest of the seven teams averaged just over two ANY/A. And even that understates things, as Luckman’s backup significantly deflated Chicago’s average. []
  • Red

    I like the new method better, although you’re right that it doesn’t change the rankings very much. I’m not in favor of eliminating negative seasons, because avoiding horrible years is a skill, and they’re already being downplayed by the 100/95/90 adjustment. When Eli Manning throws 20+ INT’s every other year, should we just pretend it never happened? I don’t think so.

    However, I have one minor concern – the baseline dropback rate of 26.1 seemed awfully low at first glance, and that bears out when you look at the chart. Of the 200 QB’s listed, fully 181 of them have their value adjusted down, indicating they played in a pass-heavy era. And it’s not just today’s players, but almost everyone from the 50’s, 60,’s, and 70’s is adjusted downward as well. Mathematically, that doesn’t even seem possible, especially for the dead-ball era of the mid-70’s. Please explain…

    • Chase Stuart


      I include the AFL and the AAFC as separate seasons from the NFL. So you’ve got 82 NFL seasons, 4 AAFC seasons, and 10 AFL seasons in the study. It felt low to me, too, but the math checks out. Since we’re using the 50% modifier, I’m happier with the low baseline.

      Re: Eli Manning, that’s true. But then again, he only moves up to 89th if we eliminate negative seasons.

      • bill

        I to go back to the importance of big games. Eli Manning has probably the 2 greatest post season runs ever. The runs contain 5 road wins. Two wins in Green Bay over Favre and Rodgers. Two Super Bowl wins over Brady. He won over a 15-1 Packer team and an 18-1 Patriot team. He beat a great and #1 seed in Dallas in 2007 and a great 49er team in SF in 2011. I don’t care if he has had ten 50 interception seasons. He has done twice what few have ever done once.

        Let me make this comparison for you. In Favre’s one Super Bowl season he went up against Elvis Grbac (Young was injured), Kerry Collins and Drew Bledsoe. No road games. Favre was 2-8 versus “great” quarterbacks in his post season career. Had he not had this run against mediocre opponents, he would never have won one Super Bowl. He could never have won the big games Eli won.

        I’m not even a fan of Eli. I’m a Packer fan. Eli Manning is an all-time top 20 quarterback. Marino, Tarkenton, Fouts, Anderson, Jurgensen, Brodie, Tittle, Hadl, Rivers and Romo are not. I think Eli has more great wins than those 10 guys combined. Yes, I’m saying Eli is better than Marino and Favre and Tarkenton and Fouts. Statistics must be put in context. Some games just don’t matter. Some games count as 100 games. That is pressure. That is what determines greatness.

        Doesn’t the fact that you have John Hadl in your all-time top 25 show you that something is very wrong?

    • James

      Eliminating negative seasons really depends on your goal. If you want to simply rate how well quaterbacks did during their career, then you should include negative seasons. If, however, you want to look at greatness or HOF type questions, then eliminating negative seasons is fine.

      Warner is probably a good example. When people remember Warner and debate about putting him in the Hall it’s going to be centered around the GSOT and Cardinals SB run, which it should be. Warner was the talk of the league and won two MVPs, as far as fame goes that matters a hell of a lot more than sucking it up with the Giants.

      Favre is another good example. When people think of his time in Minnesota, what will they think of – the excellent season and playoff run, or his poor, injury shortened final season?

      • Chase Stuart

        Good comment, James. I agree with you. I’m not quite sure what the question we’re trying to answer is, so it is difficult to come up with the right formula. My inclination is to eliminate negative seasons, just because of where guys like Aikman and Bradshaw rank.

        • Red

          Have you ever considered using an age curve to weigh a QB’s seasons? With different baselines based on age, a bad season at age 22 or 39 wouldn’t hurt nearly as much as a bad season during the prime years of 27-30. That would pretty much fix the Aikman/Bradshaw issue and many other cases similar to them.

  • Arif

    Someone’s going to come barreling in asking about Peyton Manning’s positive playoff ranking, so you might as well get ahead of the game…

    • Chase Stuart

      Yeah, I’m not too worried.

  • james

    Why isn’t there a weather and home/road adjustment? In your last series of GOAT QB, you adjusted for it. You have to adjust for Warner’s weather advantage. Will your next article also have a combined playoff/regular season list?

    • ubrab

      Is there any existing study that attempted to quantify the weather impact over QB performance? I can’t remember seeing one.

    • Chase Stuart

      I’d like to include a weather adjustment, but I need to fine-tune the process before I’m comfortable running those numbers. That’s probably a next year project.

  • Im not crazy about the removing the negative seasons deal, but its also something that probably should be considered. I guess if its one of those deals where the negative years are coming as a rookie or say the last season or two of a career I think its fair to remove because I would not want to penalize Peyton Manning if he got injured again and tried to hang on another two years and it pulled him down. Hes probably a bad example since he’ll still rank number one, but the first active guy that jumped into my head.

    The one thing I do like about pulling them out is it helps point out how good players were at one point in time. I was always a Boomer Esiason fan regardless of how bad his teams were on the Jets. For a period of time in the 80s he was up there with some terrific QBs and he just gets completely lost in the shuffle because the peak was short and he never won the Super Bowl. I kind of feel for a player like this the non-negative season does more justice.

    On that topic would you ever consider doing the chart in a form with average value per season with standard deviations to see how much variability there was in the players performance?

    • Chase Stuart

      Agreed on Boomer. In one of these iterations, he was 2nd to Ken Anderson on the list of best QBs not in the HOF.

      I’m not sure we would get much out of standard deviations — I think Peyton Manning might wind up #1 because of his high peaks. But it might be interesting to examine.

  • James

    Chase, isn’t the point of the 10/95/90 system to eliminate compliers and weight the best seasons more?

    But if you’re removing all seasons where a player is below average, you’ve already set a replacement level at league average, so I don’t think there’s any need to use the 100/95/90 system at that point.

    Of course, I’m not even sure what kind of difference that would make. Is it possible you could show the results without the 100/95/90 modifiers, or at least run it and look at them to see if there’s any differences in the ordering or magnitude?

  • Ty

    Would the personnel used by the defense factor into the league average ANY/A? Since more teams before the 2000s used more traditional (run-oriented) defenses, whereas teams today are built more to stop the pass, using nickel or dime defenses. I say this, because Dan Marino’s 1984 season seems to be a big outlier compared to his whole career (his next best season is almost half of his best season, according to your list). I’m not trying to discredit Marino, I think he is one of the best QBs ever, I am just curious. I’ve read debates about how the 1984 Dolphins using 3+ wide receivers while defenses were still using traditional looks, but I can’t prove that, myself.

    • Chase Stuart

      Sure, personnel on both sides of the ball factors into ANY/A. Marino’s ’84 season was a big outlier, but then again, a lot of record-setting seasons are big outliers. For example, none of Graham’s NFL seasons really compares to ’53, either.

  • Red

    Chase, I’m curious as to where the most egregious compilers rank on this list; specifically Bledsoe, Collins, and Testaverde.

    Are you going to publish the playoff rankings soon? I would love to see the SOS adjustments for playoff QB’s. And weather adjustments, too. Yes I’m greedy, but judging by the number of comments, GQBOAT is by far your most popular series, so you might as well milk it for as many articles as you can…

    • Chase Stuart

      Bledsoe is 141 in the old study, 147 in the new, and 126 in + only.

      Testaverde is 48/52/53 (he’s on the above table)

      Collins is 154/159/122.

      For now, I’m still working on the best formulas to handle those issues, so probably a 2015 issue.

  • Chris

    Great work Chase, I’m a bit late here but as some mentioned earlier, why get rid of the negative seasons, isn’t that what the 100/95/90 formula is for?

    Also, along with SOS adjustments/weather adjustments, is there any possible way of doing a supporting cast adjustment? Because obviously guys like say, Kurt Warner benefitted from playing with Bruce/Holt/Faulk, on the other end you have Elway who was stuck throwing to the 3 Amigos. I know that disentangling that would be very difficult and is basically the Holy Grail of football analysis, but is there any way to do it?

  • Dr Death

    Just now saw this and I guess if nobody else will, then I will speak up for Raider fans, in particular your vote that Lamonica is the Raiders greatest QB of all time. Sure, Lamonica had a great W-L record… before defenses started playing zones and changing things up. A guy who completes less than 50.0% and has a play-off record of 4-5 is hardly the greatest, no matter what era.

    Stabler was a very clutch QB, with three (3) 4th quarter comeback wins in the play-offs (’74 Miami – ’76 New England – ’77 @ Baltimore) and if not for the Immaculate Reception, would have had 4. Snake was also much more accurate and whilst this has nothing to do with your formula(s), Stabler was much more respected by his teammates and more feared by opponents. And while you can say that that is my opinion, it’s also the opinion of many former Raider players as well as former Steelers, Chiefs, Dolphins and Broncos.

    Anyway, this is an interesting thing that you’re doing, I look forward to seeing where it goes from here. Just wanted to stand up for Raider fans and also give some input as to the differences between Lamonica and Stabler.

  • Don

    I doubt anyone is going to see this, but I’ll try anyway.

    1. Am I missing a way the formula deals with seasons when the QB doesn’t play at all? I’m not necessarily talking about injury seasons – I can see an argument that avoiding injury is a skill, even if I don’t really agree – but, rather, ones like Brady’s first season or Rodgers’ first three seasons where they just didn’t play because of an entrenched starter. I know you’ve weighted seasons and/or removed negative seasons altogether, but without a minimum attempt limit it seems to me that those types of years could still impact the result. Rodgers’ 2007 year, for instance, would be included because of his high ANY/A, but he still only threw 28 passes while sitting behind Favre. Unless I’m misunderstanding the method (not even close to out of the question), that season still gets included as 60% of 40 or so yards above average (Rodgers went for 6.9 ANYA in his 28 attempts, compared to 5.5 league average).

    2. Assuming those limited action seasons are included, would it be fair to remove those seasons and figure an average value per season for each player? I ask because I was looking at pro-football-reference after Manning’s record breaking TD pass, and saw that Rodgers is ahead of him – by a good margin in some cases – in every non-counting stat except sack rate. Then I came to this list and saw that Rodgers is also way ahead of Manning in “average value per season” if you only include the 6 seasons he actually played.

    3. Basically, I’m asking if we’re someday going to be arguing counter-factuals about Rodgers’ first three seasons in some type of reverse Jordan/LeBron situation. If Rodgers plays to age 36 with three more years at his current level, and then three good but not great years after that, he’s going to be at least 5th. If he’s better than that or plays longer, he could get to second or third depending on when Brady retires. He’ll have to really fall apart to not finish first in QB rating, top 5 in TDs, and top 10 in passing yards. But he’ll always have big zeros (or close to it) for three full seasons at the start of his career. I know it’s not that rare (Romo, Rivers, Brees, and Brady all come to mind as guys who sat at least their first year), but I think it’s relevant in some of those cases.

    • Chase Stuart

      Good comments. Your view on this is correct.

      The thing to remember is that a 32-year-old QB is almost always going to be the career leader in rate stats. That’s because they get the double benefit of playing all their time in the most pass-happy era (always the trailing 10 years, it seems) and they haven’t yet entered the decline phase of their careers. Rivers is 3rd in career passer rating; Romo is 5th.

      As for Rodgers, he can certainly wind up in the GOAT discussion, although you’re right that it seems unlikely he’ll ever be able to match Manning’s sustained level of dominance. Of course, nobody can: Manning has 7 seasons where he ranked 1st or 2nd in ANY/A: http://www.footballperspective.com/eli-manning-and-the-hall-of-fame/

  • Albert Heisenberg

    I love your work Chase, but to assert Manning as the greatest is nonsensical. I’m hoping 3 years of critical observation has changed your tune. Tom Brady is the greatest QB of all-time, period. So much of your analysis cherry picks weighted values and ignores the context surrounding how we define the qualitative ‘value-added’ variable of superior QB play. In other words, how does one weigh passing averages when the pass catchers are Deion Branch and David Patten as opposed to Jerry Rice and John Taylor? It’s silly. You don’t even have an internally consistent criteria. His ANY/A in the playoffs is as equally weighted as his ANY/A in the regular season, when competition is considerably weaker? This is laughable.

    Brady > Manning…..If Brady had Manning’s weapons his entire career, this wouldn’t even be a statistical argument. In 2007 he got a 30 year old, past his prime, Moss and an unproven slot special-teamer in Welker and broke the TD record…and he isn’t done yet.

    Does 2017 Chase Stuart STILL believe Manning is better than Brady after the season he just had? 5 Super Bowls HAS to mean something; or are we just incentivizing stat-padding so a few statisticians will move a player up a notch on lists like these? After all, it’s been shown that QBs who have to play with poor defenses – and are behind – throw more. Volume stats will go up, even though one would have to correct for efficiency. It’s insane that you’re putting Peyton Manning – and the incredible talent he’s been surrounded by his entire career (more Pro-Bowlers, All-Pros, Offensive/Defensive Player of the Year teammates, AND future HOFers than any other QB of his generation).