Jump to content

Total QBR?


JetNation

Recommended Posts

jn_logo_100.gif

33 QBs played in NFL football games this past Sunday. ESPN ranked them all using their new “Total QBR” system. The results are in, and they are very stupid.

Before I get started, let me stress that this will not simply be about our own Mark Sanchez being ranked 30th by this system, although that will obviously play a role.

According to ESPN’s system, Ryan Fitzpatrick of the Buffalo Bills was the best quarterback in the NFL in week one. Now everyone who knows me knows I am no fan of Tom Brady. While I begrudgingly admit he is a great QB, I do not put him in the same class as Peyton Manning and I think he is pretty freaking far from being the best QB of all time. That being said, any system that says that anyone other than Tom Brady was the best QB in week one is completely flawed.

The guy had a game for the ages. He threw for 517 yards and four TDs, including a 99-yard TD pass to scrappy little Wes Welker. It was the sixth best single game by yardage and only the third time ever that a QB threw for 500 yards and four TDs in a game. I hate the guy with every fiber of my being but he was undoubtedly the best QB in the NFL in week one. ESPN’s Total QBR ranks him third!  And I feel so completely dirty for defending this whiny pretty boy.

Now that I’ve deloused myself, let’s examine the rest of ESPN’s ratings shall we?

Aaron Rodgers was ranked 2nd by ESPN. Hey guys, they actually got one right!

Now, I’m not going to pretend to completely understand this new system yet but ESPN has stressed what they call “clutch plays” and that plays are weighted based on how much they contributed to a win. Please then explain to me how Cam Newton, Drew Brees and Chad Henne were all ranked in the top 10 when they lost. If you lost, technically zero of your plays contributed to a win.

Michael Vick, who completed only 44% of his, passes ranked 11th. Tony Romo, who had almost identical statistics to Mark Sanchez and personally lost the game with a late fumble and a back breaking INT that led to the Jets game-winning FG, ranked 13th while Sanchez ranked 30th.

30th? Out of 33 QBs? Behind Donovan McNabb who threw for less than 40 yards? Behind Tavaris Jackson and Colt McCoy who were both terrible? Behind Matt Ryan who had arguably the worst game of his career?

I think even the biggest Mark Sanchez hater alive can admit he was better on Sunday than all of those guys. If I can admit Brady was the best over the weekend, you can admit that Sanchez was better than a guy who threw for 36 yards.

Another big issue I have with these rankings (and Football Outsiders participation in developing them) is their refusal to allow for defensive adjustment. If you’re going to create a “total quarterback rating” you have to take everything into account if you want to accurately rate the player’s performance.

Football Outsiders, who helped ESPN develop the Total QBR, has their own QB ratings system that does take strength of opponent into account and passes the eye test much better than ESPN’s the second you employ your brain: http://footballoutsiders.com/stats/qb2011 .

Funnily, Ryan Fitzpatrick does rank first still in DVOA (which means a QB with more value per play) while Brady ranks first in DYAR (which is a QB with more overall value.) That seems about right to me. FO’s rankings also list guys who played like crap in losses, like Ryan, McNabb, McCoy and Jackson near the bottom where they belong.

So ESPN set out to redefine how we rate QBs. It took them all of one week to completely discredit themselves. That sounds about right to me too.

di</img>di</img>

V9FY4wYLy8c

View the full article

Link to comment
Share on other sites

http://www.jetnation.com/2011/09/14/total-qbr/

Total QBR?

by Joe Brod on September 14, 2011

in Featured Editorials

jn_logo_100.gif

33 QBs played in NFL football games this past Sunday. ESPN ranked them all using their new “Total QBR” system. The results are in, and they are very stupid.

Before I get started, let me stress that this will not simply be about our own Mark Sanchez being ranked 30th by this system, although that will obviously play a role.

According to ESPN’s system, Ryan Fitzpatrick of the Buffalo Bills was the best quarterback in the NFL in week one. Now everyone who knows me knows I am no fan of Tom Brady. While I begrudgingly admit he is a great QB, I do not put him in the same class as Peyton Manning and I think he is pretty freaking far from being the best QB of all time. That being said, any system that says that anyone other than Tom Brady was the best QB in week one is completely flawed.

The guy had a game for the ages. He threw for 517 yards and four TDs, including a 99-yard TD pass to scrappy little Wes Welker. It was the sixth best single game by yardage and only the third time ever that a QB threw for 500 yards and four TDs in a game. I hate the guy with every fiber of my being but he was undoubtedly the best QB in the NFL in week one. ESPN’s Total QBR ranks him third! And I feel so completely dirty for defending this whiny pretty boy.

Now that I’ve deloused myself, let’s examine the rest of ESPN’s ratings shall we?

Aaron Rodgers was ranked 2nd by ESPN. Hey guys, they actually got one right!

Now, I’m not going to pretend to completely understand this new system yet but ESPN has stressed what they call “clutch plays” and that plays are weighted based on how much they contributed to a win. Please then explain to me how Cam Newton, Drew Brees and Chad Henne were all ranked in the top 10 when they lost. If you lost, technically zero of your plays contributed to a win.

Michael Vick, who completed only 44% of his, passes ranked 11th. Tony Romo, who had almost identical statistics to Mark Sanchez and personally lost the game with a late fumble and a back breaking INT that led to the Jets game-winning FG, ranked 13th while Sanchez ranked 30th.

30th? Out of 33 QBs? Behind Donovan McNabb who threw for less than 40 yards? Behind Tavaris Jackson and Colt McCoy who were both terrible? Behind Matt Ryan who had arguably the worst game of his career?

I think even the biggest Mark Sanchez hater alive can admit he was better on Sunday than all of those guys. If I can admit Brady was the best over the weekend, you can admit that Sanchez was better than a guy who threw for 36 yards.

Another big issue I have with these rankings (and Football Outsiders participation in developing them) is their refusal to allow for defensive adjustment. If you’re going to create a “total quarterback rating” you have to take everything into account if you want to accurately rate the player’s performance.

Football Outsiders, who helped ESPN develop the Total QBR, has their own QB ratings system that does take strength of opponent into account and passes the eye test much better than ESPN’s the second you employ your brain: http://footballoutsiders.com/stats/qb2011 .

Funnily, Ryan Fitzpatrick does rank first still in DVOA (which means a QB with more value per play) while Brady ranks first in DYAR (which is a QB with more overall value.) That seems about right to me. FO’s rankings also list guys who played like crap in losses, like Ryan, McNabb, McCoy and Jackson near the bottom where they belong.

So ESPN set out to redefine how we rate QBs. It took them all of one week to completely discredit themselves. That sounds about right to me too.

Link to comment
Share on other sites

Did you actually read it or did you just assume that's what its about?

Yes. The reason your argument is poor is because you're attempting to evaluate a multivariate statistic through the lens of a single conventional variable. The entire point of a number like QBR is to provide you with exactly the opposite. That your argument is since Player A or Player B threw for totals of X or Y yards, and the inference I interpret from that does not match up with what this measure says does not make it a poor measurable.

Link to comment
Share on other sites

Yes. The reason your argument is poor is because you're attempting to evaluate a multivariate statistic through the lens of a single conventional variable. The entire point of a number like QBR is to provide you with exactly the opposite. That your argument is since Player A or Player B threw for totals of X or Y yards, and the inference I interpret from that does not match up with what this measure says does not make it a poor measurable.

No, the eye test makes it a poor measurable. If you actually believe Mark Sanchez was worse on Sunday than Donavan McNabb, you don't know anything about football and there is no mathematical formula that will make you right.

Link to comment
Share on other sites

No, the eye test makes it a poor measurable. If you actually believe Mark Sanchez was worse on Sunday than Donavan McNabb, you don't know anything about football and there is no mathematical formula that will make you right.

No, it really doesn't. In any study where factors are being quantified, there are always going deviations and results that stray from conventional norms. That these exist does not discredit the studies.

Link to comment
Share on other sites

Ah yes, the never-ending battle between efficiency measures and how they don't rank Mark Sanchez highly enough. Not since the Jedi fought the Sith has the universe seen such a struggle.

actually not since they ranked pennington too high :)

qbr is a step in the right direction.. Glad it's getting mainstream appeal .. Anything that values a 7 yard pass more on 4th and 6 then on 3rd and 9 is better the what the casual fans get fed

Link to comment
Share on other sites

are there similar type ratings for every position?

no, and if there were any they would have the same flaws and limitations. you simply can't allow for every variable, because there are literally thousands if not more. I'm old fashioned, the QB's job is to get the team in the endzone and win. based on that, sanchez had a middle of the road performance, but he certainly didn't finish 30th based on that, more like 14th- 18th, which also coincides with about where I would rank him

Link to comment
Share on other sites

actually not since they ranked pennington too high :)

qbr is a step in the right direction.. Glad it's getting mainstream appeal .. Anything that values a 7 yard pass more on 4th and 6 then on 3rd and 9 is better the what the casual fans get fed

I perused the methodology before the season. Seemed pretty solid. From what I remember it's more of a population statistic. The deviations should standardize over the long term.

Link to comment
Share on other sites

no, and if there were any they would have the same flaws and limitations. you simply can't allow for every variable, because there are literally thousands if not more. I'm old fashioned, the QB's job is to get the team in the endzone and win. based on that, sanchez had a middle of the road performance, but he certainly didn't finish 30th based on that, more like 14th- 18th, which also coincides with about where I

would rank him

sure.. All these stats attempt to do is recreate the "eye test" and grade everyone on the same scale while removing eye bias.. (well the eye test if a smart fan anyway) I didn't see the buff game so I have no basis to compare to Brady other then stats .., also bias is obv real, op really thinks all jets are hof material.. Of course his eye test isn't going to be realistic

Link to comment
Share on other sites

Ah yes, the never-ending battle between efficiency measures and how they don't rank Mark Sanchez highly enough. Not since the Jedi fought the Sith has the universe seen such a struggle.

Klecko has a point. It's just not a very good point. Basically what he's saying is that when you use a generalized metric to answer a specific question (which quarterback had the best game last weekend?), it usually isn't do as good a job as just going with the eyeball test. Which is true, but so what? If you wanted to design a stat to answer that question, you could, but nobody does because that's stupid.

Link to comment
Share on other sites

sure.. All these stats attempt to do is recreate the "eye test" and grade everyone on the same scale while removing eye bias.. (well the eye test if a smart fan anyway) I didn't see the buff game so I have no basis to compare to Brady other then stats .., also bias is obv real, op really thinks all jets are hof material.. Of course his eye test isn't going to be realistic

That's not true and you know it and quite frankly, I'm getting pretty ******* sick of you saying it.

Link to comment
Share on other sites

Klecko has a point. It's just not a very good point. Basically what he's saying is that when you use a generalized metric to answer a specific question (which quarterback had the best game last weekend?), it usually isn't do as good a job as just going with the eyeball test. Which is true, but so what? If you wanted to design a stat to answer that question, you could, but nobody does because that's stupid.

That too. That's a recurring problem for efficiency measures in any sport though. They take longer to compile, and people are impatient. Can't say the blowback doesn't suck. I'm hoping Brad Pitt and Jonah Hill clear some of this up in a few weeks.

Link to comment
Share on other sites

no, and if there were any they would have the same flaws and limitations. you simply can't allow for every variable, because there are literally thousands if not more. I'm old fashioned, the QB's job is to get the team in the endzone and win. based on that, sanchez had a middle of the road performance, but he certainly didn't finish 30th based on that, more like 14th- 18th, which also coincides with about where I would rank him

I feel the same way. 30th was ridiculous. How could a guy who threw for 49 yards and lost get a better rating?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...