Content Post All-Time QB Rankings / QB Hall of Fame Monitor

[Tony2046]

I've been trying to put together a ranking using playoff point averages. In that one I assigned points 1-5 based on how far a QB went in the playoffs and used the average to come up with a ranking. The obvious problem there is that players with less trips to the playoffs are rewarded more than player with more trips.

This is the weird result of that...

Otto	Graham
Patrick	Mahomes
Tom	Brady
Roger	Staubach
Joe	Montana
Jack	Kemp
Daryle	Lamonica
Terry	Bradshaw
Jim	Kelly
Jay	Schroeder
Russell	Wilson
Bart	Starr
Troy	Aikman
Norm	Van Brocklin
Kurt	Warner
Danny	White
David	Woodley
Steve	Young
Peyton	Manning
John	Elway
Donovan	McNabb
Frank	Ryan
Phil	Simms
Mark	Rypien
Johnny	Unitas
Aaron	Rodgers
Earl	Morrall
Ben	Roethlisberger

How about adding a value for each additional trip to the playoffs?

 

[1960Pats]

How about adding a value for each additional trip to the playoffs?

That's possible. I'll have to check on some options.

 

[chris_in_sunnyvale]

I'm mulling ways to normalize winning it all across eras. For example:

If in Graham's day the top two seeds from each conference made the playoffs and had to win 2 playoff games to be champs, how many future QBs did the same? Brady's Pats were a top-2 seed and won 2 playoff games nine times. Peyton, like him or not, did it 3 times and that excludes his 1st title. Russell's Seahawks did it twice. Staubach...5 times? Elway 4 times, excluding his 1st title. Jim Kelly...3 times?

Anyway, this is one example that would fit into a larger methodology of looking at the "he's a winner" aspect.

Regards,
Chris

 

[venecol]

You could lodge a complaint with Timmy. He has also accused me of hating Brees. It's like you two are actually the same guy.

Cowherd just said Brees is a top 10 QBOAT. He made good points.

 

[Ice_Ice_Brady]

I'm mulling ways to normalize winning it all across eras. For example:

If in Graham's day the top two seeds from each conference made the playoffs and had to win 2 playoff games to be champs, how many future QBs did the same? Brady's Pats were a top-2 seed and won 2 playoff games nine times. Peyton, like him or not, did it 3 times and that excludes his 1st title. Russell's Seahawks did it twice. Staubach...5 times? Elway 4 times, excluding his 1st title. Jim Kelly...3 times?

Anyway, this is one example that would fit into a larger methodology of looking at the "he's a winner" aspect.

Regards,
Chris

I'm using a multiplier for playoff games, so that they add quite a bit of extra weight and change the overall winning pct, and the winning pct is then used to calculate a wins added type of metric.

In terms of the all-or-nothing Super Bowl era championship issue, what if we reduce the championship points and add in the conference title to make it whole for SB era QBs? This could solve the problem with double dipping for SB winners and also give a little advantage for the Kellys and Tarkentons. You can change the prorpotions below if you want...I think the trick is making it so that you're rewarding Super Bowl losers but not to the extent that someone like Montana is punished disproportionately either.

Championship - 2.5 points
Conference Championship (Hunt/Halas Trophy only) - 0.5 points

Does this look like a pretty fair point distribution? AAFC converted to conference championship, plus a playoff win. AFL does not give conference champion points.

@chris_in_sunnyvale
@Bleedthrough

 

[DarrylS]

I have a great deal of difficulty comparing Luckman ('39 to '50), Graham ('46 to '55) and Baugh ( '37 to '52) even Starr and Unitas to any of the QB's today as the game has completely changed, imo it is better to break it down by era...

Started following the NFL in the 50's and what happened during the game is nothing like what happens today... I remember the game where this Pix of YA Tittle in 1964 against the Steelers..

 

[Ice_Ice_Brady]

I'm taking it a step further and applying my adjusted winning pct formula; it uses actual winning pct., postseason weight (trying to get a clutch factor with that), and an accolade winning pct. It's an attempt to capture how responsible each QB is for those points. This is based on career average...it doesn't calculate each specific Super Bowl run. Thoughts?

 

[1960Pats]

I'm using a multiplier for playoff games, so that they add quite a bit of extra weight and change the overall winning pct, and the winning pct is then used to calculate a wins added type of metric.

In terms of the all-or-nothing Super Bowl era championship issue, what if we reduce the championship points and add in the conference title to make it whole for SB era QBs? This could solve the problem with double dipping for SB winners and also give a little advantage for the Kellys and Tarkentons. You can change the prorpotions below if you want...I think the trick is making it so that you're rewarding Super Bowl losers but not to the extent that someone like Montana is punished disproportionately either.

Championship - 2.5 points
Conference Championship (Hunt/Halas Trophy only) - 0.5 points

Does this look like a pretty fair point distribution? AAFC converted to conference championship, plus a playoff win. AFL does not give conference champion points.

@chris_in_sunnyvale
@Bleedthrough

View attachment 31144

That one looks pretty accurate.

 

[1960Pats]

I'm taking it a step further and applying my adjusted winning pct formula; it uses actual winning pct., postseason weight (trying to get a clutch factor with that), and an accolade winning pct. It's an attempt to capture how responsible each QB is for those points. This is based on career average...it doesn't calculate each specific Super Bowl run. Thoughts?

View attachment 31159

Where's Mahomes?

 

[Ice_Ice_Brady]

Where's Mahomes?

I had to take him out to make an adjustment. He is at 2.34 points, or 84% responsible the championship points. I apply a filter to players with only a few years when it comes to their efficiency scores...so he would still get a good amount (similar to Young or Manning), but if I don't adjust it, I'll get some absurd multiplier due to small sample size.

 

[Ice_Ice_Brady]

Cowherd just said Brees is a top 10 QBOAT. He made good points.

It’s easy to say anyone is a top 10 or top 5. It‘s a lot harder when you have an actual list of the top 10 or top 5 players and have to whittle it down.

I have Brees ranked top 12-14 on almost every list. If you’re not including the old timers, he’d be around 10-11. There’s no bias here...I have no issues with Brees in that range. I have issues with other players ranking too high, like Roethlisberger and Griese in the top 20. They shouldn’t be.

The only thing I’ve said negatively about Brees re: the rankings is that I prefer Rodgers because his peak play is higher. That’s hardly a scathing criticism. Generally I prefer to rank by greatness level rather than longevity...and longevity should just be a tiebreaker.

But if we’re really going with all-time and going all the way back, who are you removing from the top 10 to make room for Brees? Everyone in that top 10 was a dominant force who wrecked the league...there’s no era tier 2 guys there, which makes that top 10 consistent.

 

[venecol]

It’s easy to say anyone is a top 10 or top 5. It‘s a lot harder when you have an actual list of the top 10 or top 5 players and have to whittle it down.

I have Brees ranked top 12-14 on almost every list. If you’re not including the old timers, he’d be around 10-11. There’s no bias here...I have no issues with Brees in that range. I have issues with other players ranking to high, like Roethlisberger and Griese in the top 20. They shouldn’t be.

The only thing I’ve said negatively about Brees re: the rankings is that I prefer Rodgers because his peak play is higher. That’s hardly a scathing criticism. Generally I prefer to rank by greatness level rather than longevity...and longevity should just be a tiebreaker.

But if we’re really going with all-time and going all the way back, who are you removing from the top 10 to make room for Brees? Everyone in that top 10 was a dominant force who wrecked the league...there’s no era tier 2 guys there, which makes that top 10 consistent.

I think it's practically impossible to try to have a list of QBs across 100 years when the game, players, equipment, facilities, rules, competition, etc. has changed so dramatically. I think you're doing a great job trying it, but you're having to come up with a bunch of adjustments to bring some level of parity to it all, which ultimately leads to a lot of subjectivity.

I looked up that Duke guy at the bottom of your HOF QBs and only thing I found listed him as a TB, who played defense and called plays. Those guys did it all.

I would separate this into 3 or 4 eras. Whatever makes sense. Then pick the best guys of each one of those eras. Then all you can do at that point is compare the leaders of each era with each other on stuff like how dominant they were amongst their peers and stuff like that to maybe establish a top 5 to 7 QBs among those eras, based on quasi subjective criteria.

Idk, you're doing a great job though. How high is your grass bro?

 

[Ice_Ice_Brady]

I think it's practically impossible to try to have a list of QBs across 100 years when the game, players, equipment, facilities, rules, competition, etc. has changed so dramatically. I think you're doing a great job trying it, but you're having to come up with a bunch of adjustments to bring some level of parity to it all, which ultimately leads to a lot of subjectivity.

I looked up that Duke guy at the bottom of your HOF QBs and only thing I found listed him as a TB, who played defense and called plays. Those guys did it all.

I would separate this into 3 or 4 eras. Whatever makes sense. Then pick the best guys of each one of those eras. Then all you can do at that point is compare the leaders of each era with each other on stuff like how dominant they were amongst their peers and stuff like that to maybe establish a top 5 to 7 QBs among those eras, based on quasi subjective criteria.

Idk, you're doing a great job though. How high is your grass bro?

Right now I'm working an era/decade setting to make it really easy. You type in the years you want to look at. You might say 1960-1970 and then fade out at 2 years (progressively it fades out 1958-60 and 1960-62 to zero). That way there's no hard stops. What it will do is assign a percentage to each player based on the criteria...so Johnny Unitas, who played all those years, would get 100%; a guy who played from 1965-75 would get something like 60% of his score.

It won't count only championships/games won after a cutoff, so it's taking player rating and applying it to whatever percentage he played there. It will be interesting to see how it turns out; you could also just take someone's career and put those years in, and then maybe add a 3-4 year cusion to the beginning and end of his career, and therefoe get a good sense of how he ranked "for his era." I'll post some stuff as soon as I finish it.

My goal here is not just a master ranking system to rank all-timers against each other, though it's cool...but it's more about using that as a calibration tool; it has so many complexities that doing intra-era stuff should be a lot easier.

 

[Zarozzor]

It’s easy to say anyone is a top 10 or top 5. It‘s a lot harder when you have an actual list of the top 10 or top 5 players and have to whittle it down.

I have Brees ranked top 12-14 on almost every list. If you’re not including the old timers, he’d be around 10-11. There’s no bias here...I have no issues with Brees in that range. I have issues with other players ranking to high, like Roethlisberger and Griese in the top 20. They shouldn’t be.

The only thing I’ve said negatively about Brees re: the rankings is that I prefer Rodgers because his peak play is higher. That’s hardly a scathing criticism. Generally I prefer to rank by greatness level rather than longevity...and longevity should just be a tiebreaker.

But if we’re really going with all-time and going all the way back, who are you removing from the top 10 to make room for Brees? Everyone in that top 10 was a dominant force who wrecked the league...there’s no era tier 2 guys there, which makes that top 10 consistent.

This is also my opinion on Brees. I have no problem saying he's top 15. He played at a high level for a long time. My issue is there is only one season where you could say he was the best quarterback in football, and even then it's not like he was far away and the best that season because Favre and Manning were right there with him. I know some of his defenders like to say it's because he played in the best quarterback era but that didn't stop Brady, Manning, and Rodgers from having multiple seasons of separating themselves from the pack.

I also use longevity as more of a tiebreaker. But guys like Staubach, Young, and Rodgers were just so much better at their peak than Brees that his longevity doesn't even come into play for me against them.

 

[venecol]

Just watched the Manning vs Brees SB. Brees was so much better.

 

[Zarozzor]

Cowherd just said Brees is a top 10 QBOAT. He made good points.

He’s around top 10 if you’re doing quarterbacks who played a majority of their career post 1966.

 

[Zarozzor]

Are Brees and Elway the only two quarterbacks who are usually in the average fans top 10 where you could say they never had at least one season where they dominated compared to their peers? Elway has an MVP in 1987 but it’s suspicious as hell. Montana was better that year.

 

[Ice_Ice_Brady]

A much needed score snapshot.

 

[Zarozzor]

A much needed score snapshot.

View attachment 31209

So top 5 All Time

1. Tom Brady
2. Joe Montana
3. Bart Starr
4. Otto Graham
5. Roger Staubach

top 5 Super Bowl Era

1. Tom Brady
2. Joe Montana
3. Roger Staubach
4. Peyton Manning
5. Steve Young

I don’t see any issues with either so the formula is doing a good job. It’s also showing that the legend of Marino and Favre is greatly exaggerated.

 

[Ice_Ice_Brady]

So top 5 All Time

1. Tom Brady
2. Joe Montana
3. Bart Starr
4. Otto Graham
5. Roger Staubach

top 5 Super Bowl Era

1. Tom Brady
2. Joe Montana
3. Roger Staubach
4. Peyton Manning
5. Steve Young

I don’t see any issues with either so the formula is doing a good job. It’s also showing that the legend of Marino and Favre is greatly exaggerated.

Yep...I think it is.

The key is, on the left, you see the "Career" setting at 20%. At 0%, it's pure efficiency...so it breaks down everyone's average season and assigns that average equally to all players. At 100%, it gives full win shares per season played, so if two players have the same efficiency rating, but one played twice as long, that player will have twice as many points. This is for the Franchise and Era Rating scores (not championships or peak.) I was setting it at about 50/50 before, and that's why Manning was often #2, Favre #11, etc. It rewards longevity. I think that these guys are already rewarded by longevity by having more chances to win championships...I won't give players more points for winning more championships in less years, as I don't believe that's the right way to go. On the other hand, I've found that the heavy lean towards efficiency makes these rankings better, though there is some collateral damage with Kurt Warner at 16...but overall few inconsistencies.