Re: Kirk Cousins pick 4.7
 

IMO, anyone determining a player's potential or even ability mainly based on past stats should be shot. I know every quarterbacks' situation is different, and really that's my point.

Here's an example of 2 nfl quarterbacks' career qb completion %. the first one's retired and the second one's still actively playing.
1) 56.9% after 16 seasons
2) 60.8% so far after 6 seasons.

The first one's John Elway and second is Jason Campbell. So tell me now, how useful are these stats (just on their own) when determining playing ability and potential?

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916784]IMO, anyone determining a player's potential or even ability mainly based on past stats should be shot. I know every quarterbacks' situation is different, and really that's my point.

Here's an example of 2 nfl quarterbacks' career qb completion %. the first one's retired and the second one's still actively playing.
1) 56.9% after 16 seasons
2) 60.8% so far after 6 seasons.

The first one's John Elway and second is Jason Campbell. So tell me now, how useful are these stats (just on their own) when determining playing ability and potential?[/quote]Very, if you know what you are doing and can use them responsibly.

So to you, they're basically useless.

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916787]Very, if you know what you are doing and can use them responsibly.

So to you, they're basically useless.[/quote]

So stats are basically useless unless you are responsible or skillful? So if you're not... then what they are useless?

Ok then in the case of RGIII and Kolb, they may be useful - used to support one's opinion or projection. And in my case, of Elway/Campbell - it's just useless.

OKAY.

Re: Kirk Cousins pick 4.7
 

Numbers don't lie, but the opinions that some of the stats seem to back-up - are not always the truth.

Re: Kirk Cousins pick 4.7
 

[quote=CRedskinsRule;916064]Interesting article SS, I wanted to ask College FB fans about one statement in it though:



Do those numbers really seem eerily similar? a 5% difference in completions, in a college season seems different, in fact each stat seems to be significantly better and in Griffins favor. But maybe the difference in opponents makes up for that? Just wanted to get some feel from the college afficianados on the Warpath.

edit: I was curious about rushing as well, which the author left off, in their senior years, Kolb had a 1.4 yd rushing avg, and Griffin had a 3.9yd rushing avg. if anyone else cared.[/quote]5% college completion is a lot, although about 2ish percent of the difference is basically accounted for in the differences in era between Kolb's college career (middle of the last decade) and the last four years when Griffin was at Baylor.

I would point out that we still really don't know what Kolb is in the pro game because he had to wait so long to play consistently that we still haven't really got to start a full season yet. A good amount of that is Kolb's fault for being hurt each of the last two years and not doing anything to define himself as a quarterback.

There has not been a huge difference between Kevin Kolb and Mark Sanchez as a pro thus far in performance, but Sanchez has managed to play consistently and Kolb hasn't.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916790]Numbers don't lie, but the opinions that some of the stats seem to back-up - are not the truth.[/quote]That's basically correct, but the opinions you are referring to would be wrong whether or not stats were used to defend them.

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916792]That's basically correct, but the opinions you are referring to would be wrong whether or not stats were used to defend them.[/quote]


so what about the elway/campbell completion% stats? how telling are they of the truth?

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916793]so what about the elway/campbell completion% stats? how telling are they of the truth?[/quote]Well, first of all (and most importantly), no one is making Elway/Campbell quarterback comparisons besides you. So there's that.

There are numbers that will tell you plenty about John Elway's career if you're interested in learning more about it. Stats tell us that Elway doesn't compare great to Joe Montana, Dan Marino, or Steve Young. He had kind of a fascinating statistical career for a guy who is regarded as an all time great. There's plenty to be learned about Elway with stats. The most important takeaway for you is that there was nothing wrong with a 56% completion percentage in 1987.

Comparing his completion numbers to Campbell's is an awful misapplication, unless the argument is something like "look, even a mid-tier QB from the modern game was more productive than John Elway 25 years ago!" It's an interesting concept, but one that obviously sells Elway short. You want to avoid grading retroactively if possible. That's not what rate stats were intended for.

Re: Kirk Cousins pick 4.7
 

I firmly believe if you stuck young (25 year old, for sake of argument) John Elway on the field in the modern passing environment, didn't let him go to school or come up through the development ranks with modern coaching, and made him play 1970-80's style football against modern pass defenses, he'd probably get benched a couple weeks into the season for a veteran like Matt Hasselbeck or something.

But you could say that about a lot of the all-time greats. The game is played entirely differently now.

I think it's important to keep greatness in context. Rex Grossman faces far more challenges as quarterback of the Redskins than Sammy Baugh did and handles them probably a lot better than Baugh would have. But Grossman also had way more access to coaching and resources coming up through the ranks as a kid. Baugh pretty much invented the pass as a weapon for attacking defenses. Big difference. Grossman wouldn't have resources to become a polished pro quarterback if it wasn't for all-time greats like Baugh who came before him. Without our predecessors, life would be one big experiment of trial and error. Emphasis on error.

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916794]Well, first of all (and most importantly), no one is making Elway/Campbell quarterback comparisons besides you. So there's that.

There are numbers that will tell you plenty about John Elway's career if you're interested in learning more about it. Stats tell us that Elway doesn't compare great to Joe Montana, Dan Marino, or Steve Young. He had kind of a fascinating statistical career for a guy who is regarded as an all time great. There's plenty to be learned about Elway with stats. The most important takeaway for you is that there was nothing wrong with a 56% completion percentage in 1987.

Comparing his completion numbers to Campbell's is an awful misapplication, unless the argument is something like "look, even a mid-tier QB from the modern game was more productive than John Elway 25 years ago!" It's an interesting concept, but one that obviously sells Elway short. You want to avoid grading retroactively if possible. That's not what rate stats were intended for.[/quote]

What my point is - you can't take stats out of context and use it in place of playing ability. Every qb's situation is different. And in the case of Elway's (and also, Campbell's case), you have to throw the completion% stat out of window. So when someone is supporting an opinion and uses a stat like that - to me - it's the lazy way to determine a qb's potential and ability -- in place of analyzing through watching and it's just plain B.S.

But it's just my opinion and I have the stats to (dis)prove it. haha.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916797][B]What my point is - you can't take stats out of context[/B] and use it in place of playing ability. Every qb's situation is different. And in the case of Elway's (and also, Campbell's case), you have to throw the completion% stat out of window. So when someone is supporting an opinion and uses a stat like that - to me - it's the lazy way to determine a qb's potential and ability -- in place of analyzing through watching and it's just plain B.S.[/quote]When you take statistics out of context, you've made them useless. Which is why I said if you don't know how to use them, they aren't useful. It has absolutely nothing to do with your ability to judge talent from watching a guy play. It is not an either or. They are two different evaluation skills. They are actually both at their most predictive when they are unrelated to each other.

Most fans don't know how to use statistics, and they don't know how to evaluate from watching. But I think a much higher percentage of fans overestimate their ability to evaluate by watching. Hardly anyone overvalues their ability to read statistics meaningfully.

And the truth is if you watch a team 16 games per year you will hear enough football jargon on broadcasts and in the papers to have some concept of who is good and who is bad, even if you have absolutely no football background. So even someone who never has seen a football game in their lives, if they watched every Redskins game between now and 2013, they would be able to tell you that Brian Orakpo is really good, and Jammal Brown is not.

Re: Kirk Cousins pick 4.7
 

Just to give you comparison qbs from their respective playing eras.
Steve Young: 64.3%
Aaron Rodgers: 65.4%

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916799]When you take statistics out of context, you've made them useless. Which is why I said if you don't know how to use them, they aren't useful. It has absolutely nothing to do with your ability to judge talent from watching a guy play. It is not an either or. They are two different evaluation skills. They are actually both at their most predictive when they are unrelated to each other.

Most fans don't know how to use statistics, and they don't know how to evaluate from watching. But I think a much higher percentage of fans overestimate their ability to evaluate by watching. Hardly anyone overvalues their ability to read statistics meaningfully.

And the truth is if you watch a team 16 games per year you will hear enough football jargon on broadcasts and in the papers to have some concept of who is good and who is bad, even if you have absolutely no football background. So even someone who never has seen a football game in their lives, if they watched every Redskins game between now and 2013, they would be able to tell you that Brian Orakpo is really good, and Jammal Brown is not.[/quote]


Exactly, but even in the Elway/Campbell case, completion% rate is completely useless. So in the case of Kolb and RGIII, they can be completely useless or completely accurate, but it just depends on how RGIII performs. And not because what the stats are. It's the other way around. And that's what the misconception is to me.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916797]What my point is - you can't take stats out of context and use it in place of playing ability. Every qb's situation is different. And in the case of Elway's (and also, Campbell's case), you have to throw the completion% stat out of window. So when someone is supporting an opinion and uses a stat like that - to me - it's the lazy way to determine a qb's potential and ability -- in place of analyzing through watching and it's just plain B.S.

But it's just my opinion and I have the stats to (dis)prove it. haha.[/quote]

Those who misuse stats are usually un-versed dumbasses who can't wrap their heads around certain concepts in statistics(the subject of statistics is not the same as "stats" on nfl.com, which are compilations of data which can be analyzed by statistical measures).

Those who reject the these dumbasses' conclusions and inferences...however reject them in an equally stupid manner, blindly eschewing the use of stats(or as I prefer to call them, data) altogether.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916800]Just to give you comparison qbs from their respective playing eras.
Steve Young: 64.3%
Aaron Rodgers: 65.4%[/quote]If you're trying to say the average completion percentage isn't higher in 2011 than it was in 1987 or 1994 or whenever...you're wrong. It's about four to four and a half percent higher on average.

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916803]If you're trying to say the average completion percentage isn't higher in 2011 than it was in 1987 or 1994 or whenever...you're wrong. It's about four to four and a half percent higher on average.[/quote]

Even with that difference, the comparison is even more telling. Because when you adjust accordingly with respective eras, their completion% is about even. 56% to 60% for Elway. So like I was saying, you can try to play with the stats with however you want, but sometimes it's just plain bs. So how do you determine when they are complete bs without just plain simple watching them? And even then, it's the eye of the beholder thing in some cases.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916801]Exactly, but even in the Elway/Campbell case, completion% rate is completely useless. So in the case of Kolb and RGIII, they can be completely useless or completely accurate, but it just depends on how RGIII performs. And not because what the stats are. It's the other way around. And that's what the misconception is to me.[/quote]What you've missed the whole time is that it's not the completion percentage rate that's useless in Elway and Campbell's case. It's that the comparison is being made at all with no concept of an era regression on their statistics.

Not that this is a widespread problem in this thread, because you were the only one to make the comparison. The takeaway is not that completion percentage is useless in this case. The takeaway is that irresponsibly comparing players on any grounds is useless.

Re: Kirk Cousins pick 4.7
 

Don't get me wrong, stats like any other kind of number can be used as guidelines. But the end-all to bad, good, or great... not for me.

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916805]What you've missed the whole time is that it's not the completion percentage rate that's useless in Elway and Campbell's case. It's that the comparison is being made at all with no concept of an era regression on their statistics.

Not that this is a widespread problem in this thread, because you were the only one to make the comparison. The takeaway is not that completion percentage is useless in this case. The takeaway is that irresponsibly comparing players on any grounds is useless.[/quote]

But I just made the adjustment to the Elway's %, is it still BS?

And how do you determine if it's BS?

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916804]Even with that difference, the comparison is even more telling. Because when you adjust accordingly with respective eras, their completion% is about even. 56% to 60% for Elway. So like I was saying, you can try to play with the stats with however you want, but sometimes it's just plain bs. [B]So how do you determine when they are complete bs without just plain simple watching them? And even then, it's the eye of the beholder thing in some cases.[/B][/quote]You don't just determine it unless you have some concept of how to responsibly use statistics to back a point. The point doesn't have to be right 100% of the time, but it has to be defensible 100% of the time.

Jason Campbell actually did complete a higher percentage of his passes over the course of his career (to date) than John Elway did. That is a real thing that happened. It is not a statistical illusion.

What it isn't is a fair thing to hold against the career of John Elway, because that would lack context and miss about 14 critical points in terms of football analysis.

Stats are just records of events. The bs'ers you refer to are typically more human.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916808]But I just made the adjustment to the Elway's %, is it still BS?

And how do you determine if it's BS?[/quote]For the last time, the statistics are not BSing me. Your argument is BSing me. That is the key difference that you are missing.

The statistics could not be more clear. Total completions divided by total attempts. No black box needed.

Re: Kirk Cousins pick 4.7
 

... going back to my point that stats alone can't determine sh*t. Potential, ability, or otherwise.

They can only be used as guidelines. And I don't know what you are saying with responsible usage. There's only different opinions and judgments. Like who was better- Elway or Marino? Brady or Manning? There's no absolute truth to any of these answers. Only opinions and judgments.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916800]Just to give you comparison qbs from their respective playing eras.
Steve Young: 64.3%
Aaron Rodgers: 65.4%[/quote]

Yeah, the obvious thing to concluded that that each quarterbacks' respective receivers caught a certain percentage of passes, with Rodgers having the higher percentage. That's the only fact that the data tell you.

Saying they Rodgers is better is an [I]inference[/I]; they are as good as the premises you use to justify them.

1. Aaron Rodgers had a higher completion percentage than Steve Young(true)
2. Having a higher completion % than someone else necessarily makes him a better player than someone else(likely not true)
Conclusion: Aaron Rodgers is better than Steve Young (when using the above premises, likely not true. You can use other premises to try and justify it though)

The truth value (true or false) of Premise 2 is not true, and for a multitude of reasons, none of which discredits the use of statistical analysis done properly. In fact, it can be disproven using statistical methods depending on how the data behaves. One can use a test of whether the means are equivalent to see if there is a statistically significant different between the two mean values(note: statistical significance does not necessarily signify significance in real life)

Re: Kirk Cousins pick 4.7
 

[quote=GTripp0012;916810]For the last time, the statistics are not BSing me. Your argument is BSing me. That is the key difference that you are missing.

The statistics could not be more clear. Total completions divided by total attempts. No black box needed.[/quote]

Stats are just numbers that can be used as guidelines and not the truth to playing ability. This is the one underlying statement that you can take from what I am saying.

Using the stats to project RGIII by comparing him to Kolb's is complete B.S. to me.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916808][B]But I just made the adjustment to the Elway's %[/B], is it still BS?[/quote]If you make the adjusted statement, it reads something like "when adjusted for the era they each played in, Jason Campbell is roughly as effective in terms of his rate of completing passes as John Elway was back it the day."

Which is pretty close to factual. That's a really INTERESTING statement, and there is plenty that can be learned from it. It probably says something about Campbell, and it likely says something about Elway as well. This is a great example of how using stats responsibly can help us understand football better.

Here's now to misuse statistics. An adjusted statement: "once adjusted for the era they play in, Jason Campbell is just as good of a player statistically as John Elway was."

Well, no, not really. In some ways yes, in most other ways no. Campbell only looks like Elway in terms of adjusted rate statistics. In terms of career statistics such as total touchdown passes, total attempts, games started, total completions, total passing yards, etc., statistics of a counting nature are highly unlikely to ever support the idea that Campbell is eventually likely to be John Elway. If we came back and looked at Campbell's stats in 2015, we would likely see a guy who had an undistinguished six year career as a productive starter, then settled in as a backup.

Which is why I think it would have been so interesting if Elway played today. Would he have been able to have had a 15 year career? We know for a fact it would have been tougher.

Re: Kirk Cousins pick 4.7
 

[quote=SirClintonPortis;916812]Yeah, the obvious thing to concluded that that each quarterbacks' respective receivers caught a certain percentage of passes, with Rodgers having the higher percentage. That's the only fact that the data tell you.

Saying they Rodgers is better is an [I]inference[/I]; they are as good as the premises you use to justify them.

1. Aaron Rodgers had a higher completion percentage than Steve Young(true)
2. Having a higher completion % than someone else necessarily makes him a better player than someone else(likely not true)
Conclusion: Aaron Rodgers is better than Steve Young (when using the above premises, likely not true. You can use other premises to try and justify it though)

The truth value (true or false) of Premise 2 is not true, and for a multitude of reasons, none of which discredits the use of statistical analysis done properly. In fact, it can be disproven using statistical methods depending on how the data behaves. One can use a test of whether the means are equivalent to see if there is a statistically significant different between the two mean values(note: statistical significance does not necessarily signify significance in real life)[/quote]

This was way too wordy and too much scientific/lawyer talk for me... you lost me at the word: concluded.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916813]Stats are just numbers that can be used as guidelines and not the truth to playing ability. This is the one underlying statement that you can take from what I am saying.

Using the stats to project RGIII by comparing him to Kolb's is complete B.S. to me.[/quote]Well, if you compare him to Kolb in college statistically, you get the idea that RG3 played at a higher level. Which is also what getting the Heisman trophy says.

The Kolb comparison is useful for one reason and one reason only: sometimes productive college players end up being undistinguished professionals for whatever reason.

But unproductive college players are FAR more likely to be undistinguished pros than productive college players.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916815]This was way too wordy and too much scientific/lawyer talk for me... you lost me at the word: concluded.[/quote]
Logic is simple subject. There are three things you must be able to comprehend:
1. Statements
2. Premises
3. Conclusions

There is no science, because no scientific principles are applied here, but lawyers do have to know this shit...and a lot more.
I don't bother proofreading, but since you insist, I will rectify that problem. So, will you now read it? ;)

[quote]Yeah, the obvious thing to conclude is that each quarterbacks' respective receivers caught a certain percentage of passes, with Rodgers having the higher percentage. That's the only fact that the data tell you.

Saying they Rodgers is better is an inference; they are as good as the premises you use to justify them.

1. Aaron Rodgers had a higher completion percentage than Steve Young(true)
2. Having a higher completion % than someone else necessarily makes him a better player than someone else(likely not true)
Conclusion: Aaron Rodgers is better than Steve Young (when using the above premises, likely not true. You can use other premises to try and justify it though)

The truth value (true or false) of Premise 2 is not true, and for a multitude of reasons, none of which discredits the use of statistical analysis done properly. In fact, it can be disproven using statistical methods depending on how the data behaves. One can use a test of whether the means are equivalent to see if there is a statistically significant different between the two mean values(note: statistical significance does not necessarily signify significance in real life)[/quote]

Re: Kirk Cousins pick 4.7
 

[quote=SirClintonPortis;916817]Logic is simple subject. There are three things you must be able to comprehend:
1. Statements
2. Premises
3. Conclusions

There is no science, because no scientific principles are applied here, but lawyers do have to know this shit...and a lot more.
I don't bother proofreading, but since you insist, I will rectify that problem. So, will you now read it? ;)[/quote]

You must've just finished a college course in writing.

Re: Kirk Cousins pick 4.7
 

[quote=warriorzpath;916822]You must've just finished a college course in writing.[/quote]

Over a year ago. The logic class was taken 5 years ago.
Just as a note: I'm also one of those who hate people misusing stats.

The hopefully shorter and less nuanced version:

Conclusions based off statistics are only as sound as the assumptions behind them. As assumption that a higher career completion percentage by itself, no matter how small the difference, is not a good assumption for variety of reasons, including a few statistical ones.

What usually passes for "analysis" of statistics in usual football conversations is not really statistical analysis at all, and the fellas doing it would get slaughtered by those "in the know".

Re: Kirk Cousins pick 4.7
 

Has anyone heard that the Redskins picked Cousins, because the Broncos took Bolden? I

Re: Kirk Cousins pick 4.7
 

^ yes there is some truth to that.


Omar Bolden, the CB out of ASU was considered a late 1 early 2nd rounder but he got injured in practice before the season started, and never came back. The Skins worked him out and liked him at that first 4th rounder, but he was gone. Him being gone, and Cousins being that highly ranked made it an easier decision for shanahan.

I'm just glad that this Kirk Cousins talk is dying down.

Re: Kirk Cousins pick 4.7
 

All the so called experts that ripped the Skins for drafting Cousins don't know shit. Now a days and in the past you need 2 good QB's on your team, Grossnan isn't one of them. Cousins is insurance in case RGIII gets hurt.

Re: Kirk Cousins pick 4.7
 

[quote=CultBrennan59;916870]^ yes there is some truth to that.


Omar Bolden, the CB out of ASU was considered a late 1 early 2nd rounder but he got injured in practice before the season started, and never came back. The Skins worked him out and liked him at that first 4th rounder, but he was gone. Him being gone, and Cousins being that highly ranked made it a more easier decision for shanahan.

I'm just glad that this Kirk Cousins talk is dying down.[/quote]
Yeah, a friend of mine told me the next day that that is who they wanted. He was also under the opinion that if we could not get Griffin we would have drafted Cousins as early as the second round. When I realized that he was available, in the fourth that is the gut I hoped for. It was a great pick. I think we will get a minimum of third round compensatory pick for him when its all said and done.

Re: Kirk Cousins pick 4.7
 

i don't think he will be traded, but a 4th at best, if he is

Re: Kirk Cousins pick 4.7
 

[quote=SFREDSKIN;916871]All the so called experts that ripped the Skins for drafting Cousins don't know shit. Now a days and in the past you need 2 good QB's on your team, [B][I][U]Grossnan[/U][/I][/B] isn't one of them. Cousins is insurance in case RGIII gets hurt.[/quote]

Another new word for Ruh.

Re: Kirk Cousins pick 4.7
 

[quote=CultBrennan59;916870]^ yes there is some truth to that.


Omar Bolden, the CB out of ASU was considered a late 1 early 2nd rounder but he got injured in practice before the season started, and never came back. The Skins worked him out and liked him at that first 4th rounder, but he was gone. Him being gone, and Cousins being that highly ranked made it a[B][I] more easier[/I][/B] decision for shanahan.

I'm just glad that this Kirk Cousins talk is dying down.[/quote]

Come on man....really?????

Re: Kirk Cousins pick 4.7
 

Robert Griffin and Kevin Kolb are separated by a gulf of physical talent.

Re: Kirk Cousins pick 4.7
 

Dissertation! Dissertation! Dissertation! Dissertation!

Re: Kirk Cousins pick 4.7
 

[quote=SFREDSKIN;916871]All the so called experts that ripped the Skins for drafting Cousins don't know shit. Now a days and in the past you need 2 good QB's on your team, Grossnan isn't one of them. Cousins is insurance in case RGIII gets hurt.[/quote]

I concur. A lot of these NFL analysts don't know anything. If they were smart they would observe closely what the Redskins quarterback situation was and came to the conclusion that the Redskins not only desperately needed a quality starter to man the position, but equally a quality back up. Rex Grossman is only the back up because he knows the offense. The drafting of Kirk Cousins could prove to be one of the most valuable picks by the Redskins in years.