No matter what you think the future holds for JP

[Orton's Arm]

Of course.

 

99/162 66.1% 1131 yds 6.98 yds/comp 5 int

 

Also, the Bills have only attempted to pass 10 times in 7 games with less than two yards to go.

817860[/snapback]

I was wondering how those stats compared to those for other quarterbacks. In the process of looking for aggregate QB stats, I came across this breakdown from Football Outsiders. Their system has Losman ranked 29th out of 36 QBs.

[Orton's Arm]

Wow, you just make up your own rules so the "converted" numbers can support you argument.

Lol, no, the 2005 numbers were ones I crunched before the season began. Going through the play-by-play is a pain in the neck, and I was too lazy to do it a second time.

 

I had to draw the line somewhere, and I figured 10 yards was enough for a first down. I didn't realize that there were more 10 - 20 passing yard drives last year than there've been this year. Good find.

[Wraith]

I was wondering how those stats compared to those for other quarterbacks. In the process of looking for aggregate QB stats, I came across this breakdown from Football Outsiders. Their system has Losman ranked 29th out of the 36 QBs with enough passes to be ranked.

817888[/snapback]

 

Their system is interesting, I will give them that. They attempt something very important, putting the data in context. However, I've read their book and their methods, while a huge step in the right direction, are still amateurish. Some of their comments in regards to statistics make me cringe. In addition, by their own admission, these rankings are much more for the passing game as a whole rather than the individual quarterback. They are unable to decouple the quarterback from the offensive line and wide receivers. It's a difficult issue to avoid.

 

Two weeks ago Losman was 14th in their rankings.

 

Now he is 29th.

 

Where he ends up at the end of the year will be interesting, and possibly somewhat valid, but not before then.

[syhuang]

Lol, no, the 2005 numbers were ones I crunched before the season began. Going through the play-by-play is a pain in the neck, and I was too lazy to do it a second time.

 

I had to draw the line somewhere, and I figured 10 yards was enough for a first down. I didn't realize that there were more 10 - 20 passing yard drives last year than there've been this year. Good find.

817893[/snapback]

You have to set a threshold, but it can not favor one side. The converted numbers tell a totally different story when using 10 yards and 15 yards as threshold.

 

So can we conclude the numbers you showed are not that meaningful on evaluating JP's progression?

[ganesh]

The Bills problem, and Losman's problem, has not been sustaining drives as you assert. It's scoring points. The Bills, and Losman, have been fairly succesful in between the 20s. Where both Losman and the Bills in general have suffered is in the red zone. But the blame does not lay solely with Losman, as Willis McGahee is averaging 1.53 yds/carry (43 yards on 28 carries) in the red zone and has 1 TD. Hell, Losman has 77.7% of the Bills TDs on offense (7/9) and McGahee has 11.1% (1/9).

817846[/snapback]

 

Bingo..Our redzone inefficiency has really killed us....Either we have had turnovers or come back with FGs...Some of the problems there should be

attributed to Losman for locking on to Evans, or fumbling the ball. At the same

time, if the linemen get called for false starts or illegal motion penalties by WRs then it puts the offense again on 1-20s which kills us from running the ball

down there. Also, this OL has failed to show any push when we needed to

get 4-6 yards to the endzone...

[Orton's Arm]

You have to set a threshold, but it can not favor one side. The converted numbers tell a totally different story when using 10 yards and 15 yards as threshold.

 

So can we conclude the numbers you showed are not that meaningful on evaluating JP's progression?

817901[/snapback]

Even using a 15 yard threshold, Losman's only increased his points per game by 7% this year versus his second stint from last year. That's not a big enough increase to demonstrate meaningful improvement.

 

Add to that the fact that his yards per attempt is slightly lower this year than it was last year, his TD/INT ratio is about the same, and you're looking at numbers which don't support the idea that he's improved. I guess you could use completion percentage to make the case he's improved, if you're willing to overlook the ways completion percentage can be inflated. Or you could use the eyeball test, at least if your eyes saw something different than mine. Or you could make the claim that this year's circumstances are worse than last year's. If that was the case, stable numbers would indicate improving performance. But this year the offensive line is, um, less bad than it was last year, and the playcalling is better. Eric Moulds is gone, but that's partially negated by the Robert Royal upgrade at TE.

[syhuang]

Even using a 15 yard threshold, Losman's only increased his points per game by 7% this year versus his second stint from last year. That's not a big enough increase to demonstrate meaningful improvement.

Define "Big enough". Have you showed any historical data to define "big enough"?

 

What is regarded as "big enough" for using 15 yards as threshold? how about 20 yards?

 

Add to that the fact that his yards per attempt is slightly lower this year than it was last year, his TD/INT ratio is about the same, and you're looking at numbers which don't support the idea that he's improved. I guess you could use completion percentage to make the case he's improved, if you're willing to overlook the ways completion percentage can be inflated. Or you could use the eyeball test, at least if your eyes saw something different than mine. Or you could make the claim that this year's circumstances are worse than last year's. If that was the case, stable numbers would indicate improving performance. But this year the offensive line is, um, less bad than it was last year, and the playcalling is better. Eric Moulds is gone, but that's partially negated by the Robert Royal upgrade at TE.

817936[/snapback]

You must be mistaken, you're discussing JP's improvement with other people here. All I said was to point out you manipulate numbers by making up your own rules so the converted numbers can support your argument.

 

Don't bring other stuff in to shift the focus, I'm only talking about your "converted points per game" numbers from your own rules. Admit your rules have at least one major flaw and the results are not consistent by using different thresholds.

[Orton's Arm]

Define "Big enough". Have you showed any historical data to define "big enough"?

By "big enough" I mean "likely to be statistically significant." Generally, small differences are likely to be due to noise. So the slight decrease in yards per attempt probably doesn't mean Losman is doing worse, just as the slight increase in adjusted points per game (using a 15 yard threshold) probably doesn't mean Losman is doing better. Either or both of these trends could easily be wiped out by his next game.

Don't bring other stuff in to shift to focus, I'm only talking about your "converted points per game" numbers from your own rules. Admit your rules have major flaw and the results are not consistent by using different thresholds.

817947[/snapback]

Whether you use a 10 yard threshold or a 15 yard threshold, it's very hard to use the adjusted points per game stat to make the case that Losman is doing significantly better this year than he was in his second stint last year.

[Wraith]

By "big enough" I mean "likely to be statistically significant." Generally, small differences are likely to be due to noise. So the slight decrease in yards per attempt probably doesn't mean Losman is doing worse, just as the slight increase in adjusted points per game (using a 15 yard threshold) probably doesn't mean Losman is doing better. Either or both of these trends could easily be wiped out by his next game.

817962[/snapback]

 

This is the most reasonable statement you've ever made on these boards in regards to statistics. You are probably right. But have you done the work to prove it or are you just assuming?

[syhuang]

By "big enough" I mean "likely to be statistically significant."

You still can not show why 7% or 15% is not "statistically significant" in your method. In some cases, 1% is statistically significant. In other cases, 20% is statistically significant.

 

Generally, small differences are likely to be due to noise. So the slight decrease in yards per attempt probably doesn't mean Losman is doing worse, just as the slight increase in adjusted points per game (using a 15 yard threshold) probably doesn't mean Losman is doing better. Either or both of these trends could easily be wiped out by his next game.

Again, you haven't defined "noise" here. The noise range is different in different studies. Some cases have higher noise range, some have lower.

 

Whether you use a 10 yard threshold or a 15 yard threshold, it's very hard to use the adjusted points per game stat to make the case that Losman is doing significantly better this year than he was in his second stint last year.

In this the third time to remind you, you never define "big enough" or "significantly better" by providing other qb's stats. Your method shows very inconsistent result when using 10 yards, 15 yards, or 20 yards as threshold.

[Orton's Arm]

This is the most reasonable statement you've ever made on these boards in regards to statistics. You are probably right. But have you done the work to prove it or are you just assuming?

817966[/snapback]

To test for statistical significance, I'd need to create a t-distribution for his adjusted points from last year (at a 15 yard threshold), and compare that to the t-distribution for his adjusted points from this year. I could go through all that, but I already know what answer it would give me. The difference won't be statistically significant.

 

But even if there was a statistically significant difference (which there wouldn't be), what, precisely, would that mean? Especially, what would it mean in light of the fact that his adjusted points per game at a ten yard ratio was actually higher in his second stint of 2005 than it's been this year?

[Dante]

When you present the numbers that way, it doesn't look like he's made that much progress at all. And when you watch him play, it just seems like there's something missing. Especially these last few weeks.

817432[/snapback]

IMHO that "something missing" is quick decision making. He really hesitates and his feet start getting that nervous shuffle and his head starts quickly scanning the field in a desperate way. He reminds me a little of Drew as far as the indecison goes. If he can commit quicker I think he would be much better off and Im sure the OLine would appreciate it as well.

[syhuang]

To test for statistical significance, I'd need to create a t-distribution for his adjusted points from last year (at a 15 yard threshold), and compare that to the t-distribution for his adjusted points from this year. I could go through all that, but I already know what answer it would give me. The difference won't be statistically significant.

817971[/snapback]

 

Stop bringing up some technical terms like t-distribution until you really know what it is. IN your method, you take the summations of JP's number last year and this year. You are comparing summations, the t-distribution of the same player is useless when comparing summations.

 

What are other quarterbacks' numbers you used to set the standard?

[Orton's Arm]

You still can not show why 7% or 15% is not "statistically significant" in your method. In some cases, 1% is statistically significant. In other cases, 20% is statistically significant.

In this case, the difference will be statistically insignificant due to a low sample size (4.5 games last year, 7 games this year) and due to the high level of variance in each sample. In 2006, for example, Losman helped the team score 17 points in the Jets game, but just 6 points in the second New England game. This creates a lot of uncertainty about how many points the Losman of 2006 will help the team score in the next game.

Again, you haven't defined "noise" here. The noise range is different in different studies. Some cases have higher noise range, some have lower.

In this the third time to remind you, you never define "big enough" or "significantly better" by providing other qb's stats. Your method shows very inconsistent result when using 10 yards, 15 yards, or 20 yards as threshold.

817967[/snapback]

The definition of "noise" is very broad, and includes differences in the quality of the defenses Losman faced, differences in playcalling, differences in the play of the supporting cast, differences in weather conditions, and everything else along those lines. Please believe me when I say that "noise" could very easily account for quite a bit more than the 7% change we're looking at.

 

As for other QBs' stats, it's not necessary to bring them into this particular discussion. The question at hand is whether Losman is playing better this year than he did in his second stint last year.

[syhuang]

In this case, the difference will be statistically insignificant due to a low sample size (4.5 games last year, 7 games this year) and due to the high level of variance in each sample. In 2006, for example, Losman helped the team score 17 points in the Jets game, but just 6 points in the second New England game. This creates a lot of uncertainty about how many points the Losman of 2006 will help the team score in the next game.

Again, where did you get your stastics degree? You're comparing summations between two seasons, and now, you are trying to use the individual game distribution of a season to set 'statistically insignificant'?

 

The definition of "noise" is very broad, and includes differences in the quality of the defenses Losman faced, differences in playcalling, differences in the play of the supporting cast, differences in weather conditions, and everything else along those lines. Please believe me when I say that "noise" could very easily account for quite a bit more than the 7% change we're looking at.

No, I don't believe you. Your study of statistics method is totally wrong. You can NOT use individual game variance to define statisitcally significnat noise on year-to-year comparison.

 

Give you a hint: look at year-to-year variance of bigger sample.

 

As for other QBs' stats, it's not necessary to bring them into this particular discussion. The question at hand is whether Losman is playing better this year than he did in his second stint last year.

817980[/snapback]

So you didn't take other qb's data into study and basically only compare JP's numbers to define "statistically insignificant"?

[Orton's Arm]

Where did you get your statistics degree? You only looked at one player to define 'statistical significance'. This is amazing.

The question was whether Losman's 7% improvement in adjusted points per game at a 15 yard threshold is statistically significant. Bringing the results of other players into the picture won't help answer that question, because the question is about whether what Losman is doing this year is better than what he was doing last year.

 

The fact of the matter is that the aforementioned 7% improvement is not statistically significant. But you can't read too much into that either. It's possible that Losman has in fact made improvements, but that these improvements have been masked by the noise factors I mentioned earlier.

 

What you're looking at is a null hypothesis that Losman hasn't improved. Then you're trying to disprove the null by looking for a statistically significant improvement in his adjusted points per game total for this year versus his second stint last year. In this case, you're going to fail to disprove the null. But failing to disprove the null is not the same as proving the null. It's possible that you couldn't disprove the null because the null was true. It's also possible the null is false, but that you couldn't disprove it because there weren't enough data points for you to do so.

[Wraith]

Add to that the fact that his yards per attempt is slightly lower this year than it was last year, his TD/INT ratio is about the same, and you're looking at numbers which don't support the idea that he's improved. I guess you could use completion percentage to make the case he's improved, if you're willing to overlook the ways completion percentage can be inflated. Or you could use the eyeball test, at least if your eyes saw something different than mine. Or you could make the claim that this year's circumstances are worse than last year's. If that was the case, stable numbers would indicate improving performance. But this year the offensive line is, um, less bad than it was last year, and the playcalling is better. Eric Moulds is gone, but that's partially negated by the Robert Royal upgrade at TE.

817936[/snapback]

 

It is funny hearing an argument from a guy named Holcomb's arm AGAINST completion percentage. However, you are right, completion percentage can be artificially inflated.

 

In the case of J.P. Losman, it hasn't.

 

Pass Thrown in Air Table, 2006

Distance______Att_______Comp______Comp%_______% of Total Plays

Behind LOS____46_______28_________60.9%__________23.4%

1-10 Yds______95_______70_________73.7%__________48.2%

11-20 Yds_____31_______17_________54.8%__________15.7%

21-30 Yds_____14_______6__________42.9%__________7.1%

31-40 Yds_____6________1__________16.7%__________3.0%

41+Yds_______5________0__________0.0%___________2.5%

 

Pass Thrown in Air Table, 2005

Distance______Att_______Comp______Comp%_______% of Total Plays

Behind LOS____46_______25_________54.0%__________20.4%

1-10 Yds______112______66_________58.9%__________49.5%

11-20 Yds_____42_______15_________35.7%__________18.5%

21-30 Yds_____10_______3__________30.0%__________4.4%

31-40 Yds_____9________2__________22.2%__________4.0%

41+Yds_______7________2__________28.6%__________3.0%

 

Completion PCT by Down, 2006

Down___Comp%____Att

1______70.1%______77

2______57.1%______63

3______56.4%______55

4______50.0%______2

 

Completion PCT by Down, 2005

Down___Comp%____Att

1______41.0%______83

2______63.2%______68

3______46.6%______73

4______50.0%______4

 

It's been argued in the past that Losman is inflating his completion % by throwing shorter passes(<10 yards), which clearly is not the case (only 2.0% more over last year). He's just MUCH better at it. In addition, the most significant improvement is coming from his improvement in the 11-20 yard pass range.

 

It has also been argued that Losman is inflating his completion % by throwing short on third and long (ala Mr. Holcomb). If that were the case, his completion percentage would have to be significantly better on 3rd down than the aggregate (to make up for the fewer attempts). From the table, that is also clearly not the case.

 

The real change is that Losman is much better this year at the short and intermediate passes and is really damn good at the first down pass.

 

This is the progress that I see. I still see a QB that loses the ball at odd moments and who should learn when to throw the ball away. But anyone who argues against progress is simply ignoring facts.

[syhuang]

The question was whether Losman's 7% improvement in adjusted points per game at a 15 yard threshold is statistically significant. Bringing the results of other players into the picture won't help answer that question, because the question is about whether what Losman is doing this year is better than what he was doing last year.

You are wrong again. By using 15 yards or 20 yards as threshold, JP is better than last year. The question is if 7% or 15% is statistically better. A standard needs to set based on a bigger sample. You only use one player and wrongfully use individual game distruction to define year-to-year standard.

 

The fact of the matter is that the aforementioned 7% improvement is not statistically significant. But you can't read too much into that either. It's possible that Losman has in fact made improvements, but that these improvements have been masked by the noise factors I mentioned earlier.

 

What you're looking at is a null hypothesis that Losman hasn't improved. Then you're trying to disprove the null by looking for a statistically significant improvement in his adjusted points per game total for this year versus his second stint last year. In this case, you're going to fail to disprove the null. But failing to disprove the null is not the same as proving the null. It's possible that you couldn't disprove the null because the null was true. It's also possible the null is false, but that you couldn't disprove it because there weren't enough data points for you to do so.

817988[/snapback]

This is irrevelant to study of "converted points per game". I only disapprove your "points per game" data and didn't say anything else. Stop shifting the focus.

[Wraith]

In this case, the difference will be statistically insignificant due to a low sample size (4.5 games last year, 7 games this year) and due to the high level of variance in each sample. In 2006, for example, Losman helped the team score 17 points in the Jets game, but just 6 points in the second New England game. This creates a lot of uncertainty about how many points the Losman of 2006 will help the team score in the next game.

 

The definition of "noise" is very broad, and includes differences in the quality of the defenses Losman faced, differences in playcalling, differences in the play of the supporting cast, differences in weather conditions, and everything else along those lines. Please believe me when I say that "noise" could very easily account for quite a bit more than the 7% change we're looking at.

 

As for other QBs' stats, it's not necessary to bring them into this particular discussion. The question at hand is whether Losman is playing better this year than he did in his second stint last year.

817980[/snapback]

 

Your definition of noise is fine, but the general explanation might be better to use here: Noise is variation that can be explained by the natural, random behavior of the system (I.E. cannot be explained by a special cause).

 

A t-test is the general method for determining differences in mean to a specified confidence. However, the T-Test is not nearly as sensitive at picking up process shifts when samples sizes are small (the process being quarter backing a football team in this example). Something as simple as control chart with run tests would pick up the change quicker.

[Orton's Arm]

Stop bringing up some technical terms like t-distribution until you really know what it is. IN your method, you take the summations of JP's number last year and this year. You are comparing summations, the t-distribution of the same player is useless when comparing summations.

This is about how many points Losman helped the offense score in an average game last year, versus how many he helped it score in an average game this year. In other words, we're dealing with two measured means, and we're testing to see whether the difference in those two measured means is statistically significant. The correct tool to test for statistical significance in differences between measured means is the t-distribution.