(Dec 2022 revisit, update Brady, Rodgers)

this is just my way to highlight a few QBs with unusual stats, some in the Hall of Fame

It is targeted to produce a small rating with worse performance.

I just picked workable weights to give a rating approx 0-1

rating = [ (PC-50)/70 + (WP-40)/70 + (TD-INT)/(2*INT)+ RYD/PYD + SBW/(5*SBA) ] /2.0 +0.1

where SBW=superbowl wins   SBA=superbowl appearances

          PC=percent completion   WP=win percentage  TD=passing TD  INT=interceptions

          RYD =rushing yards   PYD=passing yards

A high performing QB might achieve PC<70, WP=70,   (TD-INT)/INT=2  RYD/PYD=1/40

I will list Tom Brady, along with some other notables:

Rodgers* 1.41 = ( (0.653-.50)/.7 + (0.664-.40)/.7 + ( 473-104)/(2*104) + 3454/58691 + 1/(5*1) )/2.0 + 0.10

Brady* 1.05= ( (0.643-.50)/.7  + (.754-.40)/0.7  + (649-212)/(2*212)  +  1123/89214  +  7/(5*10)  )/2.0   +0.1 (typ.)

McNabb 0.613 =  ( (0.590-.50)/.7  + (.613-.40)/0.7  + (234-117)/(2*117)  +  3459/37276  +  0/(5*1)  )/2.0 +0.1

Favre  0.537 = ( (0.620-.50)/.7  + (.624-.40)/0.7  + (508-336)/(2*336)  +  1844/71838 +  1/(5*2)  )/2.0 +0.1  (HOF)

Starr  0.464 =  ( (0.574-.50)/.7  + (.623-.40)/0.7  + (152-138)/(2*138)  +  1308/24718  +  2/(5*2)  )/2.0 +0.1 (HOF)

Bradshaw 0.454=  ( (0.519-.50)/.7  + (.677-.40)/0.7  + (212-210)/(2*210)  +  2257/27989  +  4/(5*4)  )/2.0 +0.1  (HOF)

Stabler  0.427 = ( (0.598-.50)/.7  + (.662-.40)/0.7  + (194-222)/(2*222)  +  93/27938  +  1/(5*1)  )/2.0 +0.1

Len Dawson  0.409 = ( (0.571-.50)/.7  + (.623-.40)/0.7  + (239-183)/(2*183)  +  1293/28711  +  0  )/2.0 +0.1  (HOF)

Anderson 0.350= ( (0.593-.50)/.7  + (.529-.40)/0.7  + (197-160)/(2*160)  +  2220/32838  +  0/(5*1)  )/2.0 +0.1

Morrall 0.321 = ( (0.513-.50)/.7  + (.636-.40)/0.7  + (161-148)/(2*148)  +  878/20809  +  0/(5*3)  )/2.0 +0.1

Sonny Jurgensen  0.312 = ( (0.571-.50)/.7  + (.493-.40)/0.7  + (255-189)/(2*189)  +  493/32224  +  0  )/2.0 +0.1  (HOF)

Plummer   0.259  = ( (0.571-.50)/.7  + (.507-.40)/0.7  + (161-161)/(2*161)  +  1853/29253  +  0/(5*3)  )/2.0 +0.1

Sipe  0.249 = ( (0.565-.50)/.7  + (.509-.40)/0.7  + (154-149)/(2*149)  +  762/23713  +  0/(5*1)  )/2.0 +0.1

Namath   0.218=   ( (0.501-.50)/.7  + (.496-.40)/0.7  + (173-220)/(2*220)  +  140/27663  +  1/(5*1)  )/2.0 +0.1 --Hall of Fame !!???

Kemp  0.179=  ( (0.467-.50)/.7  + (.637-.40)/0.7  + (114-183)/(2*183)  +  1150/21218  +  0  )/2.0 +0.1

Parilli  0.144=   ( (0.466-.50)/.7  + (.516-.40)/0.7  + (178-220)/(2*220)  +  1522/22681  +  0  )/2.0 +0.1 (his stats improve once he plays for Boston)

Blanda  0.135= ( (0.477-.50)/.7  + (.515-.40)/0.7  + (236-277)/(2*277)  +  344/26920  +  0  )/2.0 +0.1

AManning  0.012 = ( (0.552-.50)/.7  + (.257-.40)/0.7  + (125-173)/(2*173)  +  2197/23911  +  0/(5*3)  )/2.0 +0.1

Snead  0.026 = ( (0.523-.50)/.7  + (.344-.40)/0.7  + (196-257)/(2*257)  +  521/30797  +  0  )/2.0 +0.1

Reverse Elite (min 100 games )

3 w/  Losing %>60%,    AManning, Snead, George, DeBerg  (0.74,0.66,0.63,0.62)

3 w/   INT/attempts >=6.0%    Blanda, Parilli, Kemp (6.9,6.6,6.0)

9 w/   INT/attempts >5.5%   Blanda, Parilli,Kemp,Snead,Stabler,Namath,Grogan,Dickey,Hadl  (6.9,6.6,6.0,5.9,5.9,5.8,5.8,5.7,5.7)

4 w/  PC<50%  Parilli,Kemp,Blanda, Layne (46.6,46.7,47.7,49.0   )   (Namath 50.1, Hadl 50.4)

5 w/  Reverse Rating (above)<0.20  AManning, Snead, Blanda, Parilli, Kemp   (0.012, 0.026, 0.135, 0.144, 0.179)     (Namath  0.218 )

9 w/  NFL rating<=67.5   Kemp,Pastorini,Namath,Snead,Hart,AManning,Zorn,Hadl,Plunkett (57.3,59.1,65.5,65.5,66.6,67.1,67.3,67.4,67.5)

It is not for me to draw conclusions, as a QB may have played on teams with less than outstanding players or whatever.  

Some would also argue a QB winning despite that (and poor stats) is indicative of special quality.

For example, YA Tittle and Namath have more career interceptions than touchdowns, and they both have lower QB ratings (ref: Google Bard 4/26/2023, could be wrong, numerics != PFR )