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

[Zarozzor]

@Ice_Ice_Brady @Deus Irae

 

[Deus Irae]

@Ice_Ice_Brady @Deus Irae

View attachment 31246

I'd like to congratulate Ryan Clark on his successful demonstration that his opinion on quarterbacks is useless.

 

[sean10mm]

I'd like to congratulate Ryan Clark on his successful demonstration that his opinion on quarterbacks is useless.

Too many helmet to helmet hits lol

 

[Ice_Ice_Brady]

@Ice_Ice_Brady @Deus Irae

View attachment 31246

6. Drew Brees
7. Trevor Lawrence
8. Dan Marino
9. Brett Favre
10. Johnny Unitas (always thrown in to make it seem like a lot went into it)

 

[venecol]

6. Drew Brees
7. Trevor Lawrence
8. Dan Marino
9. Brett Favre
10. Johnny Unitas (always thrown in to make it seem like a lot went into it)

Surprised he left Big Ben out. Didn't they win a SB together?

 

[Zarozzor]

6. Drew Brees
7. Trevor Lawrence
8. Dan Marino
9. Brett Favre
10. Johnny Unitas (always thrown in to make it seem like a lot went into it)

Always cracks me up when someone throws Unitas in at the end of their list. Sometimes they’ll even show they’re a true connoisseur of quarterbacks when they put Unitas AND Graham at the end.

 

[Ice_Ice_Brady]

I’ll have a big update for you guys tomorrow.

Based on a ton of feedback, I added a simple way rank players from an era, or exclude years, etc., create an all-decade or or all-whatever period list or to rank any quarterback against his peers, just by typing in his name. I think it works well after some tinkering and figuring out how to know when to count certain players when they’re borderline on whether they fit into the timeframe.

Also continuing to work on the ranking display itself and how to display the data points that went into the rankings so it’s not overly complex but also breaks down the rating instead of just spitting out a number.

Hoping the sheet will be operational on Google within the next week or two. It’s frustrating because on Excel, it can literally re-rank the entire population and update in about one second, while it lags on Google sheets. Trying to reduce the file size, etc, and continuing to make the interface intuitive and not easy to destroy by accident.

 

[rogerthat]

I've only read half the thread, sorry if this has been covered. But what about adjusting for dome and/or cold weather? It hurts guys like Brees and Manning, helps the Favres and Brady's?

The other suggestion I would make is blowout %. This is a measure that I've found works great for quarterbacks and coaches. The great ones rarely get blown out- they are always in the game.

 

[venecol]

I've only read half the thread, sorry if this has been covered. But what about adjusting for dome and/or cold weather? It hurts guys like Brees and Manning, helps the Favres and Brady's?

The other suggestion I would make is blowout %. This is a measure that I've found works great for quarterbacks and coaches. The great ones rarely get blown out- they are always in the game.

With all respect, I disagree with this. Why should there be an adjustment over things the QB has no control over. Should there be an adjustment for cold weather games, or artificial turf vs natural grass? What about all the old players that played in mud. This is why I think it's impractical (and impossible) to rank players across different eras. Trying to "equalize" all these factors brings too much subjectivity into this, imo.

 

[rogerthat]

With all respect, I disagree with this. Why should there be an adjustment over things the QB has no control over. Should there be an adjustment for cold weather games, or artificial turf vs natural grass? What about all the old players that played in mud. This is why I think it's impractical (and impossible) to rank players across different eras. Trying to "equalize" all these factors brings too much subjectivity into this, imo.

The dome thing is huge. Coldhardfootballfacts (if its still around) made the argument years ago that Manning especially was overrated because of the dome.

 

[Zarozzor]

This thread has caused me to research a lot of Bart Starr stuff recently, and has me contemplating if I had him too low at #6. Now I'm thinking of putting him #3 or #4.

 

[Ice_Ice_Brady]

The dome thing is huge. Coldhardfootballfacts (if its still around) made the argument years ago that Manning especially was overrated because of the dome.

It might help passer rating/stats some, but winning percentage and a whole bunch of other factors should be able to wash out the advantage. That's why I think winning percentage is very importand and should balance out passer rating...because it's about how you did versus the guy on the other sideline, both of you having the same weather and conditions.

 

[Ice_Ice_Brady]

This thread has caused me to research a lot of Bart Starr stuff recently, and has me contemplating if I had him too low at #6. Now I'm thinking of putting him #3 or #4.

Highly underrated...people who rank him in the top 10 and think they're doing him a favor are misguided. One really interesting thing I came across in researching all of these QBs: there have been so, so many attempts to win 3 straight championships, teams that absolutely should have, but for whatever reason, they didn't. I think that reason alone puts Starr in the top 5...and the deeper dive statistics confirm it.

 

[Ice_Ice_Brady]

Okay guys...I had a lot of refining stuff to do yesterday so wasn't able to post...but I am going to start right now posting a bunch of lists.

I'll start by just running through some updated stuff with the chart itself.

• You can see that to the left of the big overall score, there's a percentage on some players. This designates that some performance stats are being reduced/normalized due to the controls on the left, "Pre-1960, Pre-Merger Confidence" and "Small Sample Outlier Stats."
• There's also a percentage way to the right in the Championship area; that designates the proportion a player gets for the championship points based on an index (awards, winning pct, clutch). It does not go through each individual championship but rather applies a career score. It's adjustable and doesn't have to be used, though.
• Please note that the Grading system (A/B/C/D) is a work in progress. It's difficult because we're comparing well-known, all-time great and franchise QBs mainly too each other, so I tried to account for what they are compared to the norm, but it isn't perfect and there are some gaps. But also, note that it is a percentile based on the entire population; so if you lookup 1981-1990 and see that Jim McMahon is a C- in a category like Era Rating, that's him compared to everyone all-time. The rank number (first column) is the rank compared to the selected players.
• On the Franchise category, Pure Win % is the major factor, and Postseason & Accolades are two ways of adjusting that number. So you might say "why is this guy 0-4 in the playoffs but has a B+ playoff score"? It's because that player probably has an A+ winning pct to begin with, so geting knocked down to a B+ is a really big deal.

I'm going to start posting some lists now, really more to show the functions of this. I'm not going to re-calibrate each list/era, unless it's because Excel crashed or there's a technical problem or glaring mistake. But for now, I'll leave Joe Theismann and Jim McMahon above Warren Moon and there will be some other issues like that. Obviously we will find some issues and calibrate from there. Some of the issues might be due to preferences and values that you select and others just glaringly bad.

Feel free to request a certain player or era....I'm trying to remember some of the specifics about eras, players, etc. and what has already been requested.

@Bleedthrough @venecol @BaconGrundleCandy @robertweathers @chris_in_sunnyvale @sean10mm @1960Pats @Deus Irae @betterthanthealternative @Tony2046 @DarrylS @stinkypete


(Post-Merger version)

 

[Ice_Ice_Brady]

Player Comparison (versus peers)

I found with this one, the only way to do it was with an "all-or-nothing" model when comparing peers. You simply can't compare Elway to 50% of Montana's score because that's their overlap. So there's a cutoff point where a guy is considered close enough to be a peer and likely matched up against the spotlighted player...and if the player meets that threshhold, he gets 100% of his score. The percentages you see (McMahon, Hostetler, etc.) are some chunks taken out because of their shorter careers, leading to less confidence that their efficiency ratings reflect their play.







The Brady one is the most difficult and will serve as "the acid test" here. Such a long career spanning two eras; I don't see Warner there...will have to look into how he fell through the cracks.

 

[Ice_Ice_Brady]

Eras

Eras is a lot more tricky...you're really not just comparing one player against peers; rather, you're tallying up which guys were the best during a certain timeframe. I'm using the "Phase" here to give a sliding amount of leeway. So, for example, if you the era is set from 1980-2000, it will default to phase out at 5 years with 1975-1980 and 2000-2005 giving some partial credit. You can set it longer or just cut it off.

This also uses a system whereby a player meets some qualifications to be included on the list; after that, though, it gets pretty complex as to how much credit should be assigned. Again, I'm not able to split championships and accomplishments by years so am applying the average back into the portions of the whole. It smooths it well but doesn't allow for nuances either. Instead, the player gets 5% of his total score for each year he contributed to the era...so playing the entire 1980s will help your ranking, but guys aren't being included unless they played a good chunk of their career during the designated period.

Pre-1950



1950s



1960s



1970s



1980s



1990s



2000s



2010s



Here's how it deals with a smaller timeframe. It isn't going to be ideal for these small periods of time, but it will at least tell you who the best players in the league were (just not how well those guys performed within the period.)

 

[IllegalContact]

f*ck that.........we all truly know in our hearts that Steve Grogan is number 1....the rest of them all suck.....except Cam Newton

 

[Ice_Ice_Brady]

Some of the grade rankings aren't synced with the place rankings...this doesn't affect the actual scores and is supposed to be a guide, but I'll fix it. Part of the issue there is that I've set the grade rankings (A+, B-, etc) to grade the player's pure efficiency and not the adjusted career-based ranking that results from a setting change. It doesn't change the list order, but I'll working on syncing those up. It's kind of a double edged sword; Steve Young has an A+ in era rating, but if you make this heavily career based on his total score (which makes the era rating a win shares type of value) then it might take him down to an A-...I think the A+ better reflects his actual ranking, and if you're looking to actually up-rank him, you'll want to know where to increase the value.

...it’s a work in progress...

 

[Ice_Ice_Brady]

@Bleedthrough

I'm thinking about this for other leagues...want to get your advice:

-1960s AFC championships count as championships but not as additional conference championship (so in my point system, it's 2.5 points but not 3 points.)
-AAFC count as conference championships and playoff wins...so quite a downgrade but can't give credit for that.

For all rival leagues with fairly close competition (AAFC, USFL, AFL pre-merger), an award is valid but one level down.
-An MVP award in the rival league is equivalent to an all-pro selection.
-An all-pro selection in the rival league is equivalent to a pro bowl selection.
-A pro bowl selection does not transfer.
-Players get credit for each season, but their winning pct in that league is irrelevant. They're given their NFL winning winning pct. So for example, if Steve Young averaged 10.7 wins per season in the NFL and played one year in the USFL, he would be given one additional season of 10.7 wins and 4.3 loses.

The one that's difficult is Warren Moon and the CFL. I don't know if that league is really anything comparable to the NFL. Ideas I've come up with are: prorating the winning pct and giving the season credit, like the example above - but it can actually hurt your score if you don't have the awards. Maybe go two levels down, so Moon's MVPs there are pro bowls in the NFL. The other thing is, I wonder if I should give him one playoff win for each CFL Championship.

 

[Zarozzor]

@Bleedthrough

I'm thinking about this for other leagues...want to get your advice:

-1960s AFC championships count as championships but not as additional conference championship (so in my point system, it's 2.5 points but not 3 points.)
-AAFC count as conference championships and playoff wins...so quite a downgrade but can't give credit for that.

For all rival leagues with fairly close competition (AAFC, USFL, AFL pre-merger), an award is valid but one level down.
-An MVP award in the rival league is equivalent to an all-pro selection.
-An all-pro selection in the rival league is equivalent to a pro bowl selection.
-A pro bowl selection does not transfer.
-Players get credit for each season, but their winning pct in that league is irrelevant. They're given their NFL winning winning pct. So for example, if Steve Young averaged 10.7 wins per season in the NFL and played one year in the USFL, he would be given one additional season of 10.7 wins and 4.3 loses.

The one that's difficult is Warren Moon and the CFL. I don't know if that league is really anything comparable to the NFL. Ideas I've come up with are: prorating the winning pct and giving the season credit, like the example above - but it can actually hurt your score if you don't have the awards. Maybe go two levels down, so Moon's MVPs there are pro bowls in the NFL. The other thing is, I wonder if I should give him one playoff win for each CFL Championship.

I like those. I was tinkering around with the idea of how many points to give for AAFC and AFL championships. I definitely think at times I made them either too low or too high, but your point values seem good. The MVP/All-Pro/Pro Bowl also seems like a good idea. I have been giving Graham credit for 5 MVP’s, even though a couple came in the AAFC. So with your system they would be counted as an all-pro selection.

I honestly have no idea about the CFL. That’s probably the toughest one. An MVP being worth a pro bowl, and a CFL championship being worth a playoff win sounds like a good start.