Correlation of Passer Rating to Winning - game by game wk 6

[San-O]

They sure did. Just look at the sheer number of receptions Roger Craig had during those years. Hell, he led the league in catches one year with 92. And I'd wager the vast majority of those receptions were caught within 5 yards of the LOS. They just killed on passes to the flats, most of which were caught behind the LOS. That's why I think the "Air Yards Per Attempt" stat is one of the most useless stats out there. Doesn't tell us anything pertinent about a QB or the system he operates; completely disregards it, actually.

 

GO BILLS!!!

 

Disagree, it told us E.J. couldn't throw down field worth a damn.

[Alphadawg7]

Passer rating is often derided as a "misleading statistic". And this information is a very small one week sample size. But 13 of the 15 games were won by the team with the QB with the higher passer rating in week 6. Kyle Orton was pretty good, with a 94 rating, 14'th best out of 30 QBs this weekend. Problem was that Mr. Brady had a 139 rating, second best of the week.

 

At the other extreme Nick Foles was pretty weak with a 79 rating, but his counter part Eli Manning had a 76 rating.

 

The two outliers are wierd, some Ryan Fitzpatrick ended up ahead of Andrew Luck. 109, to 97. That was a weird game over all. And Charlie Whitehurst (87) finished ahead of Blake Bortles (88) m and that was a close rating in a 2 point game.

 

A one week sample is very small, I get that. But 87% is a pretty decent correlation (and I do know the difference between correlation and causation)

 

Guess what passer ratings don't count in the stat?...Fumbles and many other meaningful stats that determine the outcome of the game. ESPNs QBR is way better than the old passer rating and Ortons is not good. Orton did not play like a 94 rated QB, he lost a fumble that hurt this Bills a lot too.

 

Orton is the 31st ranked QB by using the QBR measure in the NFL at a pathetic 39.1 which is terrible. He should be 0-2 as a starter if it were not for no one telling the Detroit Kicker they won and all he had to do was make one kick.

 

His QB rating is a lie at 90, but his QBR shows his real effectiveness at a sad 39.1 that takes into consideration RELEVANT stats like first downs, all turnovers, etc etc. The traditional rating is a stat that not one GM, Scout, Analyst, etc ever used to measure a QB's performance. It was the least used stat in all of sports because it was meaningless. That is why the QBR was developed to have a better way to gauge how effective a QB is and its actually used quite a bit now as a barometer. No one stat is the be all, end all stat...but the old rating is as useless as they come and the modern day QBR is substantially more insightful. And according to that, he sucks...just like Manuel did who is only 2 spots behind Orton.

Edited October 17, 2014 by Alphadawg7

[Greg F]

But why use quarterback rating at all when YPA is better? (Especially when you mix three parts YPA with one part interception percentage.)

 

Copied what I posted before for convenience:

I compared the first 2 weeks of the season and found one game the first week (N. Foles - 87.45, C. Henne - 89.87). There were three games the second week:

 

B. Hoyer - 81.67 ~ D. Brees - 89.27

D. Stanton - 66.30 ~ E. Manning - 82.96

N. Foles - 84.4 ~ A. Luck - 89.3

 

For the first 32 games this year the team with the highest quarterback rating won 87.5% of the time.

 

Using just YPA for the same 32 games I found the loosing team had a higher YPA in 8 of the games. IOW, the winning team had a higher YPA in 75% of the games. OTOH, in those 8 games only one of them had no interceptions by either QB. In the other 7 games the loosing team had more interceptions than the winning team.

 

I am not quite sure how you would "mix three parts YPA with one part interception percentage". If you could give me some direction on how the interception percentage should combine with the YPA I would be happy to run the numbers.

 

As a further bonus I modified the quarterback rating formula by zeroing out the TD/att from the quarterback rating and got 3 games in the first week and none in the second where the team with the higher quarterback rating lost the game. The difference in quarterback rating for those three games was 2.60, 4.65, and 1.73 as opposed to 4.93, 16.6, 7.6, and 2.4 for the 4 games listed above.

[K-9]

Disagree, it told us E.J. couldn't throw down field worth a damn.

 

Tells me Wilson, Brees, Dalton, and Foles aren't very good at it either.

 

Ridiculous stat.

 

GO BILLS!!!

[Greg F]

Guess what passer ratings don't count in the stat?...Fumbles and many other meaningful stats that determine the outcome of the game.

 

Nobody is arguing anything different. It is a dubious stat put out by the NFL so idiots like us can argue about it.

 

ESPNs QBR is way better than the old passer rating and Ortons is not good.

 

ESPNs QBR is a black box that that they admit is composed of numerous subjective inputs. Subjective inputs make it prone to the biases of the person(s) who create the data. The subjective nature of the inputs also make it susceptible to adjustments when the results don't live up to expected results. Let me point out that the use of subjective data is a well known problem that often leads to studies that come to complete opposite conclusions. It should also be pointed out that this doesn't imply the bias is malicious, it is simply a fact of human nature. It is the old garbage in garbage out problem. To assume that it is "way better than the old passer rating" is assuming that the data is free of bias, which by its very nature cannot be. The proprietary nature of the rating makes 3rd party validation impossible and thus judging its reliability impossible. If ESPN was interested in the integrity of their rating they would release it for 3rd party analysis.

 

While the NFL passer rating is clearly too simplistic at least we know the input data is objective and easily verified. The ESPN rating, OTOH, is composed of inputs that are subjective in nature and could be complete garbage for all we know. In short ESPN's rating it is nothing but a sales gimmick.

[PlayoffsPlease]

Guess what passer ratings don't count in the stat?...Fumbles and many other meaningful stats that determine the outcome of the game. ESPNs QBR is way better than the old passer rating and Ortons is not good. Orton did not play like a 94 rated QB, he lost a fumble that hurt this Bills a lot too.

 

Orton is the 31st ranked QB by using the QBR measure in the NFL at a pathetic 39.1 which is terrible. He should be 0-2 as a starter if it were not for no one telling the Detroit Kicker they won and all he had to do was make one kick.

 

His QB rating is a lie at 90, but his QBR shows his real effectiveness at a sad 39.1 that takes into consideration RELEVANT stats like first downs, all turnovers, etc etc. The traditional rating is a stat that not one GM, Scout, Analyst, etc ever used to measure a QB's performance. It was the least used stat in all of sports because it was meaningless. That is why the QBR was developed to have a better way to gauge how effective a QB is and its actually used quite a bit now as a barometer. No one stat is the be all, end all stat...but the old rating is as useless as they come and the modern day QBR is substantially more insightful. And according to that, he sucks...just like Manuel did who is only 2 spots behind Orton.

 

This is an example of missing the point completely. 13/15 games last week were won by the team with QB with the higher passer rating. I have no idea if that holds up over the course of a season, but if its an indicator its hugely meaningful. Will it cover a 100% of cases, no. But 13/15 is pretty compelling.

[Orton's Arm]

Copied what I posted before for convenience:

 

 

For the first 32 games this year the team with the highest quarterback rating won 87.5% of the time.

 

Using just YPA for the same 32 games I found the loosing team had a higher YPA in 8 of the games. IOW, the winning team had a higher YPA in 75% of the games. OTOH, in those 8 games only one of them had no interceptions by either QB. In the other 7 games the loosing team had more interceptions than the winning team.

 

I am not quite sure how you would "mix three parts YPA with one part interception percentage". If you could give me some direction on how the interception percentage should combine with the YPA I would be happy to run the numbers.

 

As a further bonus I modified the quarterback rating formula by zeroing out the TD/att from the quarterback rating and got 3 games in the first week and none in the second where the team with the higher quarterback rating lost the game. The difference in quarterback rating for those three games was 2.60, 4.65, and 1.73 as opposed to 4.93, 16.6, 7.6, and 2.4 for the 4 games listed above.

 

> If you could give me some direction on how the interception percentage should combine with the YPA I would be happy to run the numbers.

 

Step 1: determine a reasonable range for each variable. Range for yards per attempt: 6.0 - 8.0. Range for INT percentage: 0% - 6%.

 

Step 2: determine a QB's actual performance as a percentage of possible performance. A yards per attempt of 7.0 represents 50% of what he could have achieved on that 6.0 - 8.0 scale. Turning that into a mathematical formula, you have the following:

 

Maximum performance (8.0) - actual performance (7.0)

_________________________________________________ =

Maximum performance (8.0) - minimum performance (6.0)

 

8.0 - 7.0

________ =

8.0 - 6.0

 

1/2 = 50%

 

You can use that same formula to calculate how well a QB is doing in terms of interception percentage. If his INT rate is 2%, then he's 66.67% of the way toward performing at his theoretical maximum. (Given that a 6% INT percentage is the theoretical worst-case scenario, and a 0% INT percentage is the best case.)

 

Suppose a QB's yards per attempt is 55% of the theoretical maximum, and his INT rate is 62% of the theoretical best-case scenario. To finish this process, do the following.

 

0.75 x (yards per attempt score) + 0.25 x (INT score) = total score.

(0.75 x 55%) + (0.25 x 62%) = 0.4125 + 0.155 = 0.5675.

Multiply by 100 to get 56.75. That's the QB's score on a scale of 0 - 100. It's possible for a QB's score to fall outside that 0 - 100 range, because some will have YPA or INT percentages outside the range I specified. But cases like that will be extremely rare.

Edited October 18, 2014 by Orton's Arm

[Orton's Arm]

Disagree, it told us E.J. couldn't throw down field worth a damn.

 

You are exactly right. Manuel was the beneficiary of excellent YAC. A two yard pass to Fred Jackson would go for a 40 yard gain. Or a 20 yard pass to an insanely open Mike Williams would go for an 80 yard gain and a TD. By comparing air yards per attempt with yards per attempt, it was possible to determine that 67% of Manuel's passing yardage had come via YAC. Given that most of his throws were to stationary targets, it's not like he was doing anything special to help generate all that YAC. IIRC, Manuel was the 31st best QB in terms of air yards per attempt.

 

As a consequence of all this, both his quarterback rating and yards per attempt stats greatly overstated the quality of his play. There is another way to distort both those stats: by taking a lot of sacks. Rob Johnson is a good example of that. In fact, he's the example of that, given that his sack percentage is twice as high as any other QB in NFL history. Both his yards per attempt and quarterback rating look very impressive. But, again, both those stats have been distorted due to all the sacks he took. If other quarterbacks took sacks every time a play was less than perfect, their YPA and quarterback rating stats would go up too.

 

The moral of this story is that we should avoid religious reliance on any given stat. All stats of which I'm aware are subject to distortion. This specifically includes my favorite stats (yards per attempt and air yards per attempt). Nor is the correct response to throw up one's hands and give up on stats completely. Instead of all that, we should use a number of different stats--plus the eyeball test--to keep digging, until the truth has been discovered.

[K-9]

You are exactly right. Manuel was the beneficiary of excellent YAC.

 

Please tell me what QB in the history of the game was NOT the beneficiary of excellent YAC. As mentioned up thread, Joe Montana benefited more than any other QB in the game from receivers, RBs mostly, that could turn a short pass or even a pass for negative "air yards" into nice gains. This was the RULE in their offense, not the exception.

 

Utterly superficial stat for anyone that relies on QB stats to evaluate one's play.

 

GO BILLS!!!

[Orton's Arm]

Please tell me what QB in the history of the game was NOT the beneficiary of excellent YAC. As mentioned up thread, Joe Montana benefited more than any other QB in the game from receivers, RBs mostly, that could turn a short pass or even a pass for negative "air yards" into nice gains. This was the RULE in their offense, not the exception.

 

Utterly superficial stat for anyone that relies on QB stats to evaluate one's play.

 

GO BILLS!!!

 

If you click here and sort the list by percentage of air yards, you'll see that E.J. Manuel is dead last (among the 32 QBs listed) in terms of percentage of air yards. 33% of his total passing yards came because of air yards. The other 67% was due to YAC. Compare that to the top QB on the list: Mike Glennon. Glennon gained 69.4% of his total yardage through the air; and only 30.6% due to YAC. While all QBs benefit from YAC, not all benefit equally.

 

If you then sort the list by air yards per attempt, you'll see Manuel is dead last, with an average of 2.13. The second-worst QB on the list--the 31st ranked guy--is Ryan Tannehill, with 2.56 air yards per attempt. From this link:

 

> 68.6 percent of [Manuel's] rookie season passes traveled fewer than 10 yards in the air.

 

That 68.8 percent doesn't represent Montana-like quick slant passes to targets moving horizontally. No. The vast, vast majority of those passes were to stationary targets. Targets less than 10 yards away. I could have thrown those passes! And thrown them at least as accurately as Manuel. It wouldn't even have been all that difficult. Those are routine, boring passes to make--at least for a random fan like me.

 

On two thirds of his throwing attempts, Manuel isn't doing anything more than a random fan could do. On the other hand, Kyle Orton routinely completes throws I'd never dream of attempting. Throws that are way, way beyond my pay grade or ability level. Throws that put Orton in a different league than me. (Both literally and figuratively.)

 

So what do we do about all those short throws to stationary targets Manuel makes? If I could make those throws--and I can--then I have to believe that almost any QB who's ever been on an NFL roster or practice squad could make those throws too. These are really elementary, basic throws! It's not like Manuel is doing anything special to help generate YAC that those other QBs wouldn't have done. The YAC is really, really high because of play design, and because guys like Fred Jackson and Sammy Watkins are exceptional at gaining yards after the catch. Giving Manuel credit for that YAC confuses the issue, and distorts what would otherwise be a painfully clear picture. Air yards per attempt tells the true story of Manuel's performance.

[K-9]

If you click here and sort the list by percentage of air yards, you'll see that E.J. Manuel is dead last (among the 32 QBs listed) in terms of percentage of air yards. 33% of his total passing yards came because of air yards. The other 67% was due to YAC. Compare that to the top QB on the list: Mike Glennon. Glennon gained 69.4% of his total yardage through the air; and only 30.6% due to YAC. While all QBs benefit from YAC, not all benefit equally.

 

If you then sort the list by air yards per attempt, you'll see Manuel is dead last, with an average of 2.13. The second-worst QB on the list--the 31st ranked guy--is Ryan Tannehill, with 2.56 air yards per attempt. From this link:

 

> 68.6 percent of [Manuel's] rookie season passes traveled fewer than 10 yards in the air.

 

That 68.8 percent doesn't represent Montana-like quick slant passes to targets moving horizontally. No. The vast, vast majority of those passes were to stationary targets. Targets less than 10 yards away. I could have thrown those passes! And thrown them at least as accurately as Manuel. It wouldn't even have been all that difficult. Those are routine, boring passes to make--at least for a random fan like me.

 

On two thirds of his throwing attempts, Manuel isn't doing anything more than a random fan could do. On the other hand, Kyle Orton routinely completes throws I'd never dream of attempting. Throws that are way, way beyond my pay grade or ability level. Throws that put Orton in a different league than me. (Both literally and figuratively.)

 

So what do we do about all those short throws to stationary targets Manuel makes? If I could make those throws--and I can--then I have to believe that almost any QB who's ever been on an NFL roster or practice squad could make those throws too. These are really elementary, basic throws! It's not like Manuel is doing anything special to help generate YAC that those other QBs wouldn't have done. The YAC is really, really high because of play design, and because guys like Fred Jackson and Sammy Watkins are exceptional at gaining yards after the catch. Giving Manuel credit for that YAC confuses the issue, and distorts what would otherwise be a painfully clear picture. Air yards per attempt tells the true story of Manuel's performance.

 

You keep insisting on making the Air Yard Per Attempt stat an indictment of EJ Manuel's ineptitude. Well, it's an indictment of several damn good QBs as well. That's the point and the reason the stat is useless to anyone that's honest about evaluating stats in general and QBs in particular. Mike Glennon is a good QB because of this stat? Mike Glennon? Really?

 

If you insist on using this stat to make some salient point about QB play, then Joe Montana is one of the worst QBs in history.

 

GO BILLS!!!

[Orton's Arm]

You keep insisting on making the Air Yard Per Attempt stat an indictment of EJ Manuel's ineptitude. Well, it's an indictment of several damn good QBs as well. That's the point and the reason the stat is useless to anyone that's honest about evaluating stats in general and QBs in particular. Mike Glennon is a good QB because of this stat? Mike Glennon? Really?

 

If you insist on using this stat to make some salient point about QB play, then Joe Montana is one of the worst QBs in history.

 

GO BILLS!!!

 

> If you insist on using this stat to make some salient point about QB play, then Joe Montana is one of the worst QBs in history.

 

Please provide a link to a site indicating Montana's career air yards per attempt.

 

> That's the point and the reason the stat is useless to anyone that's honest about evaluating stats in general and QBs in particular.

 

It's possible to poke holes in any statistical measure, including your favorite: yards per attempt. For example:

 

John Elway career YPA: 7.1

Jay Cutler career YPA: 7.2

Rob Johnson career YPA: 7.2

 

Any of the tools we're discussing are subject to distortion. Statistical distortion over the course of an entire career (as above) is harder to explain away than a few games' worth (the Mike Glennon example). The Rob Johnson/John Elway example isn't the only case in which YPA would have led to incorrect conclusions about the quality of QB play.

 

*************

One quarterback led his team to a 12-4 regular season record after starting slow at 3-3. He won the Associated Press's MVP award. Another quarterback was replaced due to ineffectiveness early in the season by an aging journeyman best known for headbutting a stadium wall. Which one would you want on your team? You wouldn't be able to tell from their official stats.

 

Although one didn't play the entire season, both passers had nearly identical "per attempt stats." The first QB threw for 7.2 yards per attempt and the second QB threw for 7.1. They both had a 95 NFL passer rating. Actually, Tavaris Jackson's was 95.4 and Peyton Manning's was 95. But then again, I might be able to approach a 95 rating if I were throwing dump-offs to Adrian Peterson.

************

 

From the same article:

 

****************

Here are the QBs from 2007 who led the league in percent of their passing yardage as YAC: Croyle, Testaverde, Greise, Harrington, Favre, McCown, Losman, and Lemon. . . . There's isn't a single guy on that list who we can call a legitimate starter. [This was written long, long after Favre had been in his prime.]

 

The 2008 season's list of leaders in %YAC include Cassel, O'Sullivan, Campbell, Favre (again), Losman, and Wallace. But Matt Cassel is good, right? Maybe not.

****************

 

The air yards per attempt/YAC yards proved to be a more accurate indication of Cassel's future career trajectory than did the high regard in which he was generally held.

Edited October 19, 2014 by Orton's Arm

[K-9]

Please provide a link to a site indicating Montana's career air yards per attempt.

 

No need. Everybody knows Joe Montana made a living off of short passes, especially swing passes to his RBs in the flats BEHIND the LOS. If you didn't have the pleasure of watching him, his games are readily available, so feel free. Sorry some stat-geek with too much time on his hands wasn't around to create the useless "Air Yards" stat back then, but it wasn't needed to qualify anything he did as a QB.

 

But hey, how about that Mike Glennon? Montana can't hold his jock. Or, if you need a contemporary instead, neither can Russell Wilson or Drew Brees for that matter. Mike Glennon is a GREAT QB according to his Air Yards, better than some of the best QBs in the game. That is ludicrous by ANY stretch.

 

I get it, you're a statistician. You live and love the world of stats. And that's great. But THE GAME should inform you about the stats, and NOT the other way around. I suggest you stop trying to shoe-horn stats to fit your understanding of the game and try learning a bit more about the game itself. Stop confusing "multiple reads" with a QB going through his progressions. Stop harping on "mental bandwidth" or the "ability to process lots of information quickly" when the game exists in a much simpler, less complex realm. Stop treating the high percentage play or the short pass that attempts to get the ball to your best playmakers as quickly as possible like some sort of defect. Maybe you could try appreciating YAC as an attribute of a good receiver rather than an indictment of the QB for a change.

 

GO BILLS!!!

[rafter]

Isn't passer rating differential often cited as one of the best predictors of super bowl teams?

 

Are there stats on this anywhere? This is an interesting subject. Also, id like to see the last 10 years passer rating winning percentage. As in how often did the team with the higher passer rating win that particular game.

Edited October 19, 2014 by rafter

[Orton's Arm]

No need. Everybody knows Joe Montana made a living off of short passes, especially swing passes to his RBs in the flats BEHIND the LOS. If you didn't have the pleasure of watching him, his games are readily available, so feel free. Sorry some stat-geek with too much time on his hands wasn't around to create the useless "Air Yards" stat back then, but it wasn't needed to qualify anything he did as a QB.

 

But hey, how about that Mike Glennon? Montana can't hold his jock. Or, if you need a contemporary instead, neither can Russell Wilson or Drew Brees for that matter. Mike Glennon is a GREAT QB according to his Air Yards, better than some of the best QBs in the game. That is ludicrous by ANY stretch.

 

I get it, you're a statistician. You live and love the world of stats. And that's great. But THE GAME should inform you about the stats, and NOT the other way around. I suggest you stop trying to shoe-horn stats to fit your understanding of the game and try learning a bit more about the game itself. Stop confusing "multiple reads" with a QB going through his progressions. Stop harping on "mental bandwidth" or the "ability to process lots of information quickly" when the game exists in a much simpler, less complex realm. Stop treating the high percentage play or the short pass that attempts to get the ball to your best playmakers as quickly as possible like some sort of defect. Maybe you could try appreciating YAC as an attribute of a good receiver rather than an indictment of the QB for a change.

 

GO BILLS!!!

 

> But hey, how about that Mike Glennon? Montana can't hold his jock.

 

Until you can quantify both Glennon's and Montana's air yards per attempt, there's no way for you (or any of the rest of us) to know if the above statement is true. Even the quarterbacks who have the highest percentage of air yards to total passing yards still got 33% of their passing yards from YAC. I'll grant it's possible that Montana got more than 33% of his passing yards from YAC. But for him to have been among the most YAC-intensive QBs of the game, he'd have to have gotten about 67% of his total passing yards from YAC. I don't see how it's possible for someone--without looking at any data--to be certain his YAC percentage was closer to 67% than to 33%.

 

> Stop harping on "mental bandwidth" or the "ability to process lots of information quickly"

 

The Bills front office didn't harp on mental bandwidth when they took Losman. Nor (as best I can tell) did they look for on-field evidence of the ability to process lots of information quickly when they drafted Manuel. How did that work out for them?

 

The prototypical example of a first round bust is a guy with great physical tools, with good passing statistics in college, but who's considered "raw" A "project." Whereas, most successful first round quarterbacks are considered "polished pocket passers" at the college level. "NFL-ready."

 

There are plenty of NFL front offices who think the way you do. (Or else those prototypical first round busts wouldn't have been drafted in the first round to begin with.) But those front offices are wrong. If I sense the Bills have a front office like that--which they do--my hope is for them to be replaced. The front office which chose Manuel should not be given the opportunity to choose his replacement. If, as paid professionals, they're not smart enough to see things that I as a casual fan can see, they do not constitute the long-term answer to the Bills' problems.

 

> Maybe you could try appreciating YAC as an attribute of a good receiver rather than an indictment of the QB for a change.

 

The main question about YAC is whether credit for it should be attributed to the quarterback or the WR. If the QB hits a moving target in perfect stride, thereby setting him up for really good YAC, then some or all the credit for that play's YAC belongs to the QB. If on the other hand the QB throws the ball to a stationary target, the QB doesn't deserve credit for the YAC.

 

Suppose you were to give all quarterbacks full credit for their air yards. But only credited them with YAC yards on plays when they hit a moving target in perfect stride. At the end of the day, a guy like Joe Montana would receive credit for a much greater proportion of his team's YAC yards than a typical QB would receive. This would accurately reflect the fact that Montana did much more to help create YAC yards than most QBs do.

 

Unfortunately, data like the above are not readily available. But if you see that a particular quarterback very rarely his receivers in perfect stride, there's no reason to worry about how that quarterback might have fared in the above-described analysis. Instead, you throw away the YAC yards--those are the receivers' doing, at least in this case--and look only at his air yards per attempt. By comparing this quarterback's air yards per attempt stat against that of other QBs around the league, you'll have a reasonably accurate gauge of his true contributions to the team. This is why Manuel's air yards per attempt is far more reflective of his actual performance than his yards per attempt.

Edited October 19, 2014 by Orton's Arm

[dave mcbride]

 

 

You are exactly right. Manuel was the beneficiary of excellent YAC. A two yard pass to Fred Jackson would go for a 40 yard gain. Or a 20 yard pass to an insanely open Mike Williams would go for an 80 yard gain and a TD. By comparing air yards per attempt with yards per attempt, it was possible to determine that 67% of Manuel's passing yardage had come via YAC. Given that most of his throws were to stationary targets, it's not like he was doing anything special to help generate all that YAC. IIRC, Manuel was the 31st best QB in terms of air yards per attempt.

 

As a consequence of all this, both his quarterback rating and yards per attempt stats greatly overstated the quality of his play. There is another way to distort both those stats: by taking a lot of sacks. Rob Johnson is a good example of that. In fact, he's the example of that, given that his sack percentage is twice as high as any other QB in NFL history. Both his yards per attempt and quarterback rating look very impressive. But, again, both those stats have been distorted due to all the sacks he took. If other quarterbacks took sacks every time a play was less than perfect, their YPA and quarterback rating stats would go up too.

 

The moral of this story is that we should avoid religious reliance on any given stat. All stats of which I'm aware are subject to distortion. This specifically includes my favorite stats (yards per attempt and air yards per attempt). Nor is the correct response to throw up one's hands and give up on stats completely. Instead of all that, we should use a number of different stats--plus the eyeball test--to keep digging, until the truth has been discovered.

It's important to note that TEAM passer rating factors in sacks. It's the basic rating with sacks added in as incompleted attempts and sack yardage subtracted from the total passing yardage. That's partly why it's such a good predictor. If what Alphadog says is true -- that teams don't pay attention to collective passer rating stats -- then there a lot of dumb teams out there.

Edited October 19, 2014 by dave mcbride

[Orton's Arm]

It's important to note that TEAM passer rating factors in sacks. It's the basic rating with sacks added in as incompleted attempts and sack yardage subtracted from the total passing yardage. That's partly why it's such a good predictor. If what Alphadog says is true -- that teams don't pay attention to collective passer rating stats -- then there a lot of dumb teams out there.

 

> If what Alphadog says is true -- that teams don't pay attention to collective passer rating stats -- then there a lot of dumb teams out there.

 

Prediction and measurement are two different things. Meaning, that the stat you described may be relatively good at predicting a team's success. But that does not necessarily mean it measures individual players' contribution to that success. For example, I agree that sacks are bad, and that any conglomerate measure which takes them into account will yield a higher r^2 (if winning is the dependent variable) than would have been the case had sacks not been included as one of the independent variables. But while all of that is true, it doesn't tell you whether those sacks should be blamed on the QB, the offensive line, or (in some cases) are coverage sacks blameable on the WRs.

 

If the stat you've described is being used as a tool to measure quarterbacks' performance, the underlying assumption is that quarterbacks are completely to blame for the sacks they take. An assumption like that isn't even accurate of Rob Johnson. (Mostly because Johnson played behind a terrible OL, and many of the sacks he took were the line's fault. But given that Rob Johnson's sack percentage was twice as high as any other QB in NFL history, there was plenty of blame to go around.)

 

My preference is typically for "one ingredient" stats over conglomerate stats. Don't tell me what a QB's collective passer rating is. Tell me his air yards per attempt, his YAC yards per attempt, his INT percentage, his sack percentage, etc. Give me the individual ingredients one by one, and let me put together my own analysis of what he is or isn't accomplishing. The more "one ingredient" stats you give me, the more complete the picture will be. When you combine a rigorous statistical analysis like that with the eyeball test, you can form a fairly complete picture of a QB's level of play.

Edited October 20, 2014 by Orton's Arm

[dave mcbride]

 

 

> If what Alphadog says is true -- that teams don't pay attention to collective passer rating stats -- then there a lot of dumb teams out there.

 

Prediction and measurement are two different things. Meaning, that the stat you described may be relatively good at predicting a team's success. But that does not necessarily mean it measures individual players' contribution to that success. For example, I agree that sacks are bad, and that any conglomerate measure which takes them into account will yield a higher r^2 (if winning is the dependent variable) than would have been the case had sacks not been included as one of the independent variables. But while all of that is true, it doesn't tell you whether those sacks should be blamed on the QB, the offensive line, or (in some cases) are coverage sacks blameable on the WRs.

 

If the stat you've described is being used as a tool to measure quarterbacks' performance, the underlying assumption is that quarterbacks are completely to blame for the sacks they take. An assumption like that isn't even accurate of Rob Johnson. (Mostly because Johnson played behind a terrible OL, and many of the sacks he took were the line's fault. But given that Rob Johnson's sack percentage was twice as high as any other QB in NFL history, there was plenty of blame to go around.)

 

My preference is typically for "one ingredient" stats over conglomerate stats. Don't tell me what a QB's collective passer rating is. Tell me his air yards per attempt, his YAC yards per attempt, his INT percentage, his sack percentage, etc. Give me the individual ingredients one by one, and let me put together my own analysis of what he is or isn't accomplishing. The more "one ingredient" stats you give me, the more complete the picture will be. When you combine a rigorous statistical analysis like that with the eyeball test, you can form a fairly complete picture of a QB's level of play.

I really think we're talking past each other.

[dave mcbride]

 

 

Guess what passer ratings don't count in the stat?...Fumbles and many other meaningful stats that determine the outcome of the game. ESPNs QBR is way better than the old passer rating and Ortons is not good. Orton did not play like a 94 rated QB, he lost a fumble that hurt this Bills a lot too.

 

Orton is the 31st ranked QB by using the QBR measure in the NFL at a pathetic 39.1 which is terrible. He should be 0-2 as a starter if it were not for no one telling the Detroit Kicker they won and all he had to do was make one kick.

 

His QB rating is a lie at 90, but his QBR shows his real effectiveness at a sad 39.1 that takes into consideration RELEVANT stats like first downs, all turnovers, etc etc. The traditional rating is a stat that not one GM, Scout, Analyst, etc ever used to measure a QB's performance. It was the least used stat in all of sports because it was meaningless. That is why the QBR was developed to have a better way to gauge how effective a QB is and its actually used quite a bit now as a barometer. No one stat is the be all, end all stat...but the old rating is as useless as they come and the modern day QBR is substantially more insightful. And according to that, he sucks...just like Manuel did who is only 2 spots behind Orton.

Alphadawg -- Chris Collinsworth just said on Sunday Night Football that he spoke with Dom Capers before the game, and that Capers said that there is one stat that he pays attention to: team quarterback rating differential. He quoted Capers as saying that when looking at who makes the final four every ear, it almost always comes down to team QB rating differential.

Edited October 27, 2014 by dave mcbride