Passer Rating vs QBR, Other Interesting Facts

[loofasisgeek]

I thought this was interesting:
Andrew Luck is like 20th in "Passer Rating" but #1 in "QBR". Which stat do you prefer?


A few other interesting facts:

4 penalties called in Bills game. 24 last week in Ravens game.

Ben Jarvis Green Ellis has 3 fumbles in Cincy this year. Didn't see that coming.

The Jets suck and Rex Ryan is a windbag. DID see that coming.

 

[Fixit]

I thought this was interesting:
Andrew Luck is like 20th in "Passer Rating" but #1 in "QBR". Which stat do you prefer?

I think passer rating is flawed, but QBR is absolutely useless.

 

[Sparetire]

I think passer rating is flawed, but QBR is absolutely useless.

care to explain about QBR?

 

[ivanvamp]

care to explain about QBR?

Read all about it here: NFL - Total Quarterback Rating shifts the way we see the position - ESPN

Tries to take into account running the ball as a QB. And that's not a silly thing to include, especially when taking into consideration guys who do a lot more than just throw the ball.

Like Tom Brady: New England Patriots quarterback Tom Brady TD run - NFL Videos

 

[jmt57]

It could be worse; for example, consider that Pro Football Focus gives quarterbacks a bonus for throwing into double coverage with their ratings.

 

[RodThePat]

Read all about it here: NFL - Total Quarterback Rating shifts the way we see the position - ESPN

Tries to take into account running the ball as a QB. And that's not a silly thing to include, especially when taking into consideration guys who do a lot more than just throw the ball.

Like Tom Brady: New England Patriots quarterback Tom Brady TD run - NFL Videos

It's not nearly as objective as that. TQBR includes a number of abstract concepts in it's formula. It tries to take into consideration passing in the clutch, and attributes different values to throws made in different situations. It's a nice try, but it leads to some bizarre anomalies, like Andrew Luck leading the league, when by all accounts he's been solid, but unimpressive.

I really see no issues with standard passer rating, aside from the overemphasis it puts on completion percentage. Passer rating measures a QBs efficiency, because it takes into account the five most important factors for a passer to succeed: completion percentage, TD%, INT%, yards per attempt and TD/INT ratio. Obviously, when considering a small sample size, passer rating can lead to some distortions as well, but in the long run (say, in the course of a season), it's much more likely to accurately represent the quality of a quarterback's play than TQBR. For example, according to TQBR Peyton Manning's 2010 season was better than Brady's 2007. Obviously, passer rating says otherwise, and correlates much better to what we see on the field.

 

[ivanvamp]

It's not nearly as objective as that. TQBR includes a number of abstract concepts in it's formula. It tries to take into consideration passing in the clutch, and attributes different values to throws made in different situations. It's a nice try, but it leads to some bizarre anomalies, like Andrew Luck leading the league, when by all accounts he's been solid, but unimpressive.

I really see no issues with standard passer rating, aside from the overemphasis it puts on completion percentage. Passer rating measures a QBs efficiency, because it takes into account the five most important factors for a passer to succeed: completion percentage, TD%, INT%, yards per attempt and TD/INT ratio. Obviously, when considering a small sample size, passer rating can lead to some distortions as well, but in the long run (say, in the course of a season), it's much more likely to accurately represent the quality of a quarterback's play than TQBR. For example, according to TQBR Peyton Manning's 2010 season was better than Brady's 2007. Obviously, passer rating says otherwise, and correlates much better to what we see on the field.

Right. But tell me which of these two QB had a better game (I'm making up the numbers to make a point):

Smith: 22-35 (62.9%), 330 yds, 3 td, 1 int, 139.4 passer rating
Jones: 18-30 (60.0%), 290 yds, 2 td, 1 int, 132.6 passer rating, but also 10 carries for 110 yds and 2 td

Smith accounts for 330 yards on 35 plays (9.43 per play) and 3 scores, versus one turnover.

Jones accounts for 400 yards on 40 plays (10.00 per play), and 4 scores, versus one turnover.

By passer rating, Smith was better, as he should be - his completion % was better, he threw for more yards, and also for more TD. But you factor in Jones' rushing, and it's clear who had a better overall performance.

EDIT: I'm not arguing for QBR...I'm just saying that QBs can bring more to the table than just passing, and a true rating of a QBs performance and value should take that into account.

 

[Palm Beach Pats Fan]

A lot of the parameters that go into QBR make good sense, taking into account "clutch" situations, for example, but others are bizarre (to me anyway).

For instance, you get more favorable scores for passes that travel further in the air, as opposed to a short pass with run after the catch by the receiver.

Makes no sense except that they are trying to credit YAC more to the receiver than to the QB. That's baloney. A QB who hits the WR right in stride with perfect timing on a 5 yards (in the air) pass to a WR who gains an extra 20 yards has done a lot better job for his team than the QB who lofts it 15 yards downfield for a diving catch and no YAC.

 

[RodThePat]

Right. But tell me which of these two QB had a better game (I'm making up the numbers to make a point):

Smith: 22-35 (62.9%), 330 yds, 3 td, 1 int, 139.4 passer rating
Jones: 18-30 (60.0%), 290 yds, 2 td, 1 int, 132.6 passer rating, but also 10 carries for 110 yds and 2 td

Smith accounts for 330 yards on 35 plays (9.43 per play) and 3 scores, versus one turnover.

Jones accounts for 400 yards on 40 plays (10.00 per play), and 4 scores, versus one turnover.

By passer rating, Smith was better, as he should be - his completion % was better, he threw for more yards, and also for more TD. But you factor in Jones' rushing, and it's clear who had a better overall performance.

EDIT: I'm not arguing for QBR...I'm just saying that QBs can bring more to the table than just passing, and a true rating of a QBs performance and value should take that into account.

As I said in my post, when considering a small sample size, of course you can have results that do not accurately reflect the play on the field. In your example, it's clear Jones had a better game, but over the course of a season, if the difference between his passer rating and Smith's is considerable, odds are the latter had a better year. Really, in today's NFL running ability is valuable for a QB because it helps him evade sacks and get key first downs. Pretty much nobody is running for considerable yardage, or a ton of TDs. Sure, when you look at a guy like RGIII or Cam Newton you have to say they are better players than comparable passers with no rushing skills, but those guys who are better at throwing the football than they are, are mostly better quarterbacks than they are, even if they can't run.

Seriously, use passer rating+ as a parameter for evaluating QBs. That is a formula that accounts for how much better a QB's passer rating was than the average NFL rating in a given year. If you use that, you'll find a remarkable correlation between what the stats show and what we perceive by watching football games. It's extremely rare that an average passer will have a fantastic career passer rating+.

 

[NSPF]

Right. But tell me which of these two QB had a better game (I'm making up the numbers to make a point):

Smith: 22-35 (62.9%), 330 yds, 3 td, 1 int, 139.4 passer rating
Jones: 18-30 (60.0%), 290 yds, 2 td, 1 int, 132.6 passer rating, but also 10 carries for 110 yds and 2 td

Smith accounts for 330 yards on 35 plays (9.43 per play) and 3 scores, versus one turnover.

Jones accounts for 400 yards on 40 plays (10.00 per play), and 4 scores, versus one turnover.

By passer rating, Smith was better, as he should be - his completion % was better, he threw for more yards, and also for more TD. But you factor in Jones' rushing, and it's clear who had a better overall performance.

EDIT: I'm not arguing for QBR...I'm just saying that QBs can bring more to the table than just passing, and a true rating of a QBs performance and value should take that into account.

I disagree, there is no reason to include rushing performance into a composite passing stat. Why do we need to reduce a QB's performance to one stat? Passer rating is useful because it's transparent and doesn't try to do too much.

 

[ivanvamp]

I disagree, there is no reason to include rushing performance into a composite passing stat. Why do we need to reduce a QB's performance to one stat? Passer rating is useful because it's transparent and doesn't try to do too much.

The point I was trying to make is that the traditional passer rating stat is just about passing, period. QBR is about more than that - taking everything into account.

It may be a flawed stat but at least it tries to evaluate the complete performance of a QB, not just the passing aspect of it.

 

[Deus Irae]

The point I was trying to make is that the traditional passer rating stat is just about passing, period. QBR is about more than that - taking everything into account.

It may be a flawed stat but at least it tries to evaluate the complete performance of a QB, not just the passing aspect of it.

Trying to take something into account, and failing miserably, isn't really something that lends itself to a strong recommendation. QBR is terrible, IMO, and I laugh anytime I see it being used. It's as bad as PFF stats. The difference is that ESPN is enough of a big boy outfit that it should have done better if it was going to go down this route.

QB rating is flawed, especially in a single game situation, but it tends to get it right over the course of a season.

 

[Sparetire]

Read all about it here: NFL - Total Quarterback Rating shifts the way we see the position - ESPN

Tries to take into account running the ball as a QB. And that's not a silly thing to include, especially when taking into consideration guys who do a lot more than just throw the ball.

Like Tom Brady: New England Patriots quarterback Tom Brady TD run - NFL Videos

I know what QBR is, i just dont understand why is it useless according to fixit and how come andrew luck is #1 in it and is 24# in the passer rating table, i had the same thought in other forum but i was given answers like "its only week 4" and that passer rating gives INT's much more wiegh than QPB and stuff like that.

long story short: i dont understand what the hell QBR is ment to give us and in what aspects it differs from the passer ratings.

 

[RodThePat]

The point I was trying to make is that the traditional passer rating stat is just about passing, period. QBR is about more than that - taking everything into account.

It may be a flawed stat but at least it tries to evaluate the complete performance of a QB, not just the passing aspect of it.

But the passing aspect of it is, by far and away, the most significant one. And TQBR doesn't merely include rushing stats in it's formula. Those numbers are also processed through the same arbitrary, subjective components that the passing numbers are. A good example of that is Cam Newton, who rushed for 700 yards and 14 TDs in 2011, and yet ended up lower on the TQBR list (17th) than he did on the passer rating list (15th). It's not merely a case of adding up rushing totals.

 

[ivanvamp]

Trying to take something into account, and failing miserably, isn't really something that lends itself to a strong recommendation. QBR is terrible, IMO, and I laugh anytime I see it being used. It's as bad as PFF stats. The difference is that ESPN is enough of a big boy outfit that it should have done better if it was going to go down this route.

QB rating is flawed, especially in a single game situation, but it tends to get it right over the course of a season.

Agreed. But it's only a passer rating. Leaves a big chunk of some QBs contribution out. Which is fine, because all it is supposed to do is measure passing effectiveness, not *total* quarterback play.

 

[ivanvamp]

But the passing aspect of it is, by far and away, the most significant one. And TQBR doesn't merely include rushing stats in it's formula. Those numbers are also processed through the same arbitrary, subjective components that the passing numbers are. A good example of that is Cam Newton, who rushed for 700 yards and 14 TDs in 2011, and yet ended up lower on the TQBR list (17th) than he did on the passer rating list (15th). It's not merely a case of adding up rushing totals.

I don't disagree.

 

[Deus Irae]

Agreed. But it's only a passer rating. Leaves a big chunk of some QBs contribution out. Which is fine, because all it is supposed to do is measure passing effectiveness, not *total* quarterback play.

True, but the problems with adding running into a QB rating go beyond finding which numbers to use. Blending in the running data is going to be almost impossible, because you can't really add it in with a set formula. Running the ball means one thing for Michael Vick and another for Aaron Rodgers, for example.

 

[ivanvamp]

True, but the problems with adding running into a QB rating go beyond finding which numbers to use. Blending in the running data is going to be almost impossible, because you can't really add it in with a set formula. Running the ball means one thing for Michael Vick and another for Aaron Rodgers, for example.

Yep, good point. There has to be a way of factoring it in, though, because Newton's rushing adds another incredible dynamic to his performance. But that's way above my pay grade.