Draft article charting where the starters are chosen

[San Jose Bills Fan]

Very interesting article. Thanks for posting.

[San Jose Bills Fan]

Also, this topic has been hilarious, both intentionally and otherwise.

 

But I'm not getting how a few of the earnest responses are not understanding this article.

 

This is actual data. It is quantitative analysis.

 

Now some may disagree with the author's conclusions. But they'd be wrong.

 

Of course there are exceptions. LeGarrette Blount and Chris Ivory were undrafted free agents. So was Arian Foster. But that doesn't change the fact that half of the top running backs in the NFL were drafted in the first and second rounds. It's pretty simple stuff.

 

Yes you can find players like Jabbari Greer off the street, but it's a fact that half the starting cornerbacks in the league were drafted in the first or second round.

 

The exceptions don't invalidate the study…because the study provides for the exceptions…every player came from somewhere and all those categories are provided for in the study.

 

My only quibble with the study is that they didn't break down the offensive line positions.

 

I would guess that tackles, guards, and centers would have different draft characteristics.

[K Gun Special]

This is probably the dumbest draft article I have ever seen.

 

>>>>> Get talent at CB, RB, TE, and WR early or you’ll be waiting for them to develop into contributors.<<<<<

 

If this philosophy was true, the Bills would have won every Superbowl for the last 15 or more years.

 

 

The patriots have done this with great success.

[ieatcrayonz]

By the way, what was your grandma's phone number again? She needs to return the Gordon Lightfoot CD to me.

Very funny. My Grandma died a long time ago. And she burned all of her Gordon Lightfoot records when he changed his name and became an Islamic terrorist.

[Bill from NYC]

My only quibble with the study is that they didn't break down the offensive line positions.

 

I would guess that tackles, guards, and centers would have different draft characteristics.

 

That is a pretty big quibble.

 

Another is the implication that this is the way to build a football team. Sure, the Bills have drafted 100 first round corners in last 20 or so years. The thing is, they are losing. And I DO know that there is more to it. When they drafted 1st round DEs they were either too small, sucked, or both. When they took an OT at #4 he was a fat RT (they NEVER go this early), with a chronic ankle injury.

Wide receivers? Here ya go....

http://coldhardfootballfacts.com/Articles/11_3677_A_CHFF_theory_elevated_to_Man_Law.html

 

People love to point out to me how the league has "evolved," and I am well aware of this. However, there will never be a time where QB and line play are not the most important aspects of this sport. That's why cbs and safeties are referred to as "secondary."

 

I have no problem on early picks on positions that traditionally go later, such as Guard or LB. The thing is, you REALLY don't want to miss on these guys. For instance, Patrick Willis went at #11, and Branden Albert was taken as a Guard at 15. These are the players I wanted the Bills to select in those seasons. When you miss at a player like this, it is that much harder to fill positions such as DE, QB and LT in later rounds. These guys almost always go early.

 

And to the OP, I think that the article was stupid, and this was not a personal referrence. I'm sorry if I came off as hostile.

 

The patriots have done this with great success.

 

The Patriots didn't really start taking defensive backs in the first round until they were set (and deep) at QB and both lines. Look back to 1999....

http://www.drafthistory.com/teams/patriots.html

 

If the Bills were set in those areas, I would be fine with the "BPA." Besides, when your team is set as such you will be picking late.

Edited March 1, 2011 by Bill from NYC

[ieatcrayonz]

 

 

The Patriots didn't really start taking defensive backs in the first round until they were set (and deep) at QB and both lines. Look back to 1999....

http://www.drafthistory.com/teams/patriots.html

 

If the Bills were set in those areas, I would be fine with the "BPA." Besides, when your team is set as such you will be picking late.

1999 is a long time ago. Think of it this way:

 

Building a team around an offensive line is like driving a DeSoto. It worked great for Richie's father on Happy Days and it was the right thing to do at the time. There was a lot of wear and tear but it was a rough and tumble time. There is also a reason why your local DeSoto dealer is out of business and it is not Obama's fault so don't start in with me.

 

Ferarris on the other hand are thriving and sort of like cornerbacks in today's turn on a dime NFL. Sure you might not be able to plow your driveway with it, but who cares when you can just zip down to Florida or some other warm locale where all the chicks are anyway?

[Beerball]

1999 is a long time ago. Think of it this way:

 

Building a team around an offensive line is like driving a DeSoto. It worked great for Richie's father on Happy Days and it was the right thing to do at the time. There was a lot of wear and tear but it was a rough and tumble time. There is also a reason why your local DeSoto dealer is out of business and it is not Obama's fault so don't start in with me.

 

Ferarris on the other hand are thriving and sort of like cornerbacks in today's turn on a dime NFL. Sure you might not be able to plow your driveway with it, but who cares when you can just zip down to Florida or some other warm locale where all the chicks are anyway?

What happened to the Yugo?

[Mr. WEO]

Also, this topic has been hilarious, both intentionally and otherwise.

 

But I'm not getting how a few of the earnest responses are not understanding this article.

 

This is actual data. It is quantitative analysis.

 

Now some may disagree with the author's conclusions. But they'd be wrong.

 

Of course there are exceptions. LeGarrette Blount and Chris Ivory were undrafted free agents. So was Arian Foster. But that doesn't change the fact that half of the top running backs in the NFL were drafted in the first and second rounds. It's pretty simple stuff.

 

Yes you can find players like Jabbari Greer off the street, but it's a fact that half the starting cornerbacks in the league were drafted in the first or second round.

 

The exceptions don't invalidate the study…because the study provides for the exceptions…every player came from somewhere and all those categories are provided for in the study.

 

My only quibble with the study is that they didn't break down the offensive line positions.

 

I would guess that tackles, guards, and centers would have different draft characteristics.

The author assumed that the Lions in 2008's top priority was drafting a top CB. He couldn't have been more wrong.

 

Really bad teams never should take a CB in the first round. The fact that many have, including our own and this has led to a lot of starting CBs (not "stars") coming from the 1st 2 rounds is hardly a revelation.

 

Next we'll see an article describing the CB as the next most important position after QB because of the huge contracts they sign as FA's.

[San Jose Bills Fan]

The author assumed that the Lions in 2008's top priority was drafting a top CB. He couldn't have been more wrong.

 

Really bad teams never should take a CB in the first round. The fact that many have, including our own and this has led to a lot of starting CBs (not "stars") coming from the 1st 2 rounds is hardly a revelation.

 

Next we'll see an article describing the CB as the next most important position after QB because of the huge contracts they sign as FA's.

I read the article differently than you do.

 

To me, the author is using Detroit as an example of a team which hasn't fulfilled the need for a cornerback because they're trying to find one in the bargain bin. I interpret the author as saying that IF the Lions feel the need to address cornerback in the 2008 draft that they should do so in the first two rounds.

 

In 2008 they drafted tackle Gosder Cherilus. The fact that he's been a flop doesn't really affect the argument one way or the other.

 

As others have said, it's not about what position you draft so much as what player you draft.

 

 

[Geno Smith's Arm]

I read the article differently than you do.

 

To me, the author is using Detroit as an example of a team which hasn't fulfilled the need for a cornerback because they're trying to find one in the bargain bin. I interpret the author as saying that IF the Lions feel the need to address cornerback in the 2008 draft that they should do so in the first two rounds.

 

In 2008 they drafted tackle Gosder Cherilus. The fact that he's been a flop doesn't really affect the argument one way or the other.

 

As others have said, it's not about what position you draft so much as what player you draft.

 

 

I think it is clear that the intention of the article is suggest where you can get the best value for each position. It definitely doesn't encourage drafting Cornerbacks when you have the worst rush defense in the league or any such nonsense, and I don't see how anyone could interpret anything along those lines.

 

It doesn't suggest what positions to draft when building a team, only where teams have generally gotten the best value for each position/ or where teams choose to find different positional players.

 

The article isn't as interesting to me as the chart.

[ieatcrayonz]

What happened to the Yugo?

That's only in basketball. I don't want to get into it.

[Sisyphean Bills]

This is actual data. It is quantitative analysis.

 

Now some may disagree with the author's conclusions. But they'd be wrong.

The chart is the 2-deep average for every NFL team. Since it includes K, LS and P, that's 47 players out of a 53 man roster on 32 teams. There is no differentiation given for good teams, good players, starters vs. backups, replacement players that moved into the 2-deep because of injuries (this was compiled later in the season), no consideration of the salary cap, the breakdown by position lumps some positions together (OL, LB, WR) and differentiates others (RB vs. FB), it doesn't consider that there are 32 1st round picks per year and ~2000 roster positions in the NFL.

 

So, yeah, some could disagree with any conclusions based on such data and they'd have ample reasons to blow holes in the study. About all this really says is that NFL teams (good and/or bad) value certain positions more and draft those positions higher than others (with good and/or bad results) -- this is hardly news as any draftnik would tell you the same thing. With the salary cap, short careers, and the small constant number of annual 1st round picks, it's a statistical certainty that depth in the NFL is going to come via cheaper talent via the pool of UDFAs and since this chart lumps depth with All-Pros, it provides little other meaningful information.

[San Jose Bills Fan]

The chart is the 2-deep average for every NFL team. Since it includes K, LS and P, that's 47 players out of a 53 man roster on 32 teams. There is no differentiation given for good teams, good players, starters vs. backups, replacement players that moved into the 2-deep because of injuries (this was compiled later in the season), no consideration of the salary cap, the breakdown by position lumps some positions together (OL, LB, WR) and differentiates others (RB vs. FB), it doesn't consider that there are 32 1st round picks per year and ~2000 roster positions in the NFL.

 

So, yeah, some could disagree with any conclusions based on such data and they'd have ample reasons to blow holes in the study. About all this really says is that NFL teams (good and/or bad) value certain positions more and draft those positions higher than others (with good and/or bad results) -- this is hardly news as any draftnik would tell you the same thing. With the salary cap, short careers, and the small constant number of annual 1st round picks, it's a statistical certainty that depth in the NFL is going to come via cheaper talent via the pool of UDFAs and since this chart lumps depth with All-Pros, it provides little other meaningful information.

We're talking about an article on a "scouting" website, not a postgraduate thesis or doctoral dissertation.

 

You seem to want this study to be more than it is. To be truly comprehensive it would have to be a much larger study. I was the first to point out that it lumped offensive tackles with centers and guards so I am fully aware of the limitations the article has.

 

You have to look at the information it gives you and appreciate the value it offers. Glass half full. That's the way I am.

 

I find that one of the author's findings in particular is compelling: Of the 14 positional groups, the second largest source for players in 10 of those groups is undrafted free agency. Regardless of the fact that only 250 or so players are drafted each year, it is still very interesting to consider the reason why undrafted free agents have a better chance of landing meaningful roster spots than drafted players in so many cases. The implications regarding scouting talent are fairly enormous.

 

As for the bolded above, I partially disagree with your conclusion of the study. The study states where these players were drafted…which is not exactly the same thing as stating that "teams value certain positions more and draft those positions more highly." Certainly that's part of what's going on in the study but I don't think it explains the entirety of the data.

 

Ultimately, I don't need to explain the study to you nor you to me. I think it's a good piece. You don't. And that's fine.

 

[Sisyphean Bills]

We're talking about an article on a "scouting" website, not a postgraduate thesis or doctoral dissertation.

Obviously.

You seem to want this study to be more than it is... so I am fully aware of the limitations the article has.

Nope. You clearly wanted it to be more than what it is though, as you tried to use it as ammo in an argument to imply other posters were wrong in their criticism of it.

You have to look at the information it gives you and appreciate the value it offers. Glass half full. That's the way I am.

So, there's an "optimistic way" to read articles? OK.

 

Seriously, I have no problem with the article or the author. I'm sure he spent a lot of time on it and it's well-written. Doesn't mean I have to gullibly accept it at face value and unplug my head, though.

I find that one of the author's findings in particular is compelling: Of the 14 positional groups, the second largest source for players in 10 of those groups is undrafted free agency. Regardless of the fact that only 250 or so players are drafted each year, it is still very interesting to consider the reason why undrafted free agents have a better chance of landing meaningful roster spots than drafted players in so many cases. The implications regarding scouting talent are fairly enormous.

I thought it was kind of obvious, but I'll run some simple numbers for you. He said the table is a 2 deep of the entire NFL. There are 11 players on offense, 11 on defense, and 3 STs positions listed. That's roughly 90% of an NFL roster. There are roughly 2000 players in the NFL, so that comes to 1800 players in his chart. Now, factor in that teams average 7 picks a year, the average NFL career is 2 or 3 years, and even if we skew things towards the drafted players optimistically, there is a significant percentage of roster positions that have to be filled by somebody -- these happen to be UDFAs. Let's say every team keeps all 7 of their picks for 5 years (which the data does not support), there would still be 25% of the positions in the chart left to be filled by somebody. Now, if you factor in the pressures of the salary cap and free agency and actually do look at the data over time, it's obvious that NFL teams have been trending towards depth that is cheaper and younger, precisely UDFAs. Again, it's pretty clear why this is the case, but if there is 50 dollars to split 50 ways and you start off giving the star players 15 dollars, then there is going to be much less than a dollar left over for the backups on the team. If the CBA guarantees veterans at least 85 cents, then there is going to be a tipping point where the cost of the star players makes it impossible to keep the veterans.

Ultimately, I don't need to explain the study to you nor you to me. I think it's a good piece. You don't. And that's fine.

Whatever.

[Geno Smith's Arm]

Obviously.

 

Nope. You clearly wanted it to be more than what it is though, as you tried to use it as ammo in an argument to imply other posters were wrong in their criticism of it.

 

So, there's an "optimistic way" to read articles? OK.

 

Seriously, I have no problem with the article or the author. I'm sure he spent a lot of time on it and it's well-written. Doesn't mean I have to gullibly accept it at face value and unplug my head, though.

 

I thought it was kind of obvious, but I'll run some simple numbers for you. He said the table is a 2 deep of the entire NFL. There are 11 players on offense, 11 on defense, and 3 STs positions listed. That's roughly 90% of an NFL roster. There are roughly 2000 players in the NFL, so that comes to 1800 players in his chart. Now, factor in that teams average 7 picks a year, the average NFL career is 2 or 3 years, and even if we skew things towards the drafted players optimistically, there is a significant percentage of roster positions that have to be filled by somebody -- these happen to be UDFAs. Let's say every team keeps all 7 of their picks for 5 years (which the data does not support), there would still be 25% of the positions in the chart left to be filled by somebody. Now, if you factor in the pressures of the salary cap and free agency and actually do look at the data over time, it's obvious that NFL teams have been trending towards depth that is cheaper and younger, precisely UDFAs. Again, it's pretty clear why this is the case, but if there is 50 dollars to split 50 ways and you start off giving the star players 15 dollars, then there is going to be much less than a dollar left over for the backups on the team. If the CBA guarantees veterans at least 85 cents, then there is going to be a tipping point where the cost of the star players makes it impossible to keep the veterans.

 

Whatever.

 

 

There are way too many !@#$s on this board

[San Jose Bills Fan]

There are way too many !@#$s on this board

Some of the people here I just don't get. It's really strange. Some days they're nice and other days they're really pissy.

 

Anyways, thanks for posting the article.

Edited March 4, 2011 by San Jose Bills Fan

[Orton's Arm]

The chart is the 2-deep average for every NFL team. Since it includes K, LS and P, that's 47 players out of a 53 man roster on 32 teams. There is no differentiation given for good teams, good players, starters vs. backups, replacement players that moved into the 2-deep because of injuries (this was compiled later in the season), no consideration of the salary cap, the breakdown by position lumps some positions together (OL, LB, WR) and differentiates others (RB vs. FB), it doesn't consider that there are 32 1st round picks per year and ~2000 roster positions in the NFL.

 

So, yeah, some could disagree with any conclusions based on such data and they'd have ample reasons to blow holes in the study. About all this really says is that NFL teams (good and/or bad) value certain positions more and draft those positions higher than others (with good and/or bad results) -- this is hardly news as any draftnik would tell you the same thing. With the salary cap, short careers, and the small constant number of annual 1st round picks, it's a statistical certainty that depth in the NFL is going to come via cheaper talent via the pool of UDFAs and since this chart lumps depth with All-Pros, it provides little other meaningful information.

I agree with the core of your criticism of the study.

 

For example, I've seen evidence which shows that most of the good QBs in the NFL are taken in the first round. That's the sort of thing that can easily be obscured by the flaws in the study you pointed out. For example, Peyton Manning and his backup count equally in the author's two deep system. As the Colts' QBs get engulfed in the maw of that study, Manning is treated as having exactly equal value as his backup. Which is ridiculous. That flaw is repeated across the entire breadth of the study, a fact which clearly shows that the study can't be used to show where the good players at any given position are to be found. All the study indicates is where "backups + starters lumped together" are found. While the latter data are not useless, we must be aware of those data's (very severe) limitations.

 

On an unrelated note, those who believe that games are won by running and by stopping the run are incorrect. The New York Times performed a statistical analysis which showed that a 1 SD improvement in your passing game was four times as important as a 1 SD improvement in your running game. They observed the same ratio in passing defense versus rushing defense.

 

I think a big reason for that is the difference between average yards per rushing attempt versus average yards per passing attempt. You'd expect a standard-issue RB to get about 4 yards per rushing attempt. A good QB can achieve 7.5 yards per pass attempt. Those extra 3.5 yards per play that your passing game provides are pivotal in sustaining drives. You could say, "if I got 4 yards a carry every carry, I'd score a touchdown every drive without a single pass play." The problem with that is the random variation. On some plays you may get no gain, or only a couple of yards or something. Other running plays will give you more than 4 yards a carry to bring the overall average up to 4.

 

Suppose your plan is to run the ball on first, second, and third downs, get four yards each time, and get a first down. So you run the ball on first and second down, but things don't go your way. Now you're in third and eight. At this point, the onus is wholly on your passing game to bail you out and allow you to sustain the drive. If your passing attack is good enough to consistently bail you out of 3rd and 8 situations, it's also (presumably) good enough to do a lot of damage on first and second downs too. You need a good passing game to allow you to sustain your running game. But once you have that good passing game, you no longer need a great running game. Sure, it'd be nice to have both. But the Patriots' offense did just fine with Antowain Smith at RB and Tom Brady at QB.

 

A good passing attack relies on a good quarterback to throw the ball, a good offensive line to give him time to throw, and a good receiving corps to catch the passes. All three elements are important, but you could make the argument that the receivers are less important than the other two components. Also, an offense will probably achieve more with a great QB and a merely competent OL than it can with a great OL and a merely competent QB. That being said, a great OL can and will magnify the impact of a great QB. Kurt Warner will give you a very good passing game. Kurt Warner + Orlando Pace will give you a passing game that's frightening!

Edited March 4, 2011 by Edwards' Arm

[Orton's Arm]

One other thing: I found some parts of the WR article to be convincing, and other parts less so. As an example of the latter, they mentioned that the Patriots' offense improved after they released Moss. But the fact that Moss was playing badly enough for the Patriots to release him illustrates that he's washed-up. The fact that the Patriots weren't harmed by releasing a washed-up WR does not support the argument that WRs are of secondary importance.

 

Neither, for that matter, do some of the final sentences of that article. "Boldin let what might have been a game-winning, 6-yard touchdown pass bounce off his chest with four minutes left and with the Ravens trailing 24-21." Using that example to try to show that good WRs aren't important is a lot like using Scott Norwood's miss to show that having a good kicker is unimportant. It's just not a logical argument to make. I think there's some element of truth to the article's conclusion, but I feel that conclusion was incorrectly supported.

 

If you have a great QB and a solid OL, you can get by with Stevie Johnson or Deion Branch-type WRs. But if you have a Tom Brady of a QB and a Patriots-like OL, adding a non-washed-up Randy Moss to the mix will help significantly improve your offense.

[Sisyphean Bills]

One other thing: I found some parts of the WR article to be convincing, and other parts less so. As an example of the latter, they mentioned that the Patriots' offense improved after they released Moss. But the fact that Moss was playing badly enough for the Patriots to release him illustrates that he's washed-up. The fact that the Patriots weren't harmed by releasing a washed-up WR does not support the argument that WRs are of secondary importance.

 

Neither, for that matter, do some of the final sentences of that article. "Boldin let what might have been a game-winning, 6-yard touchdown pass bounce off his chest with four minutes left and with the Ravens trailing 24-21." Using that example to try to show that good WRs aren't important is a lot like using Scott Norwood's miss to show that having a good kicker is unimportant. It's just not a logical argument to make. I think there's some element of truth to the article's conclusion, but I feel that conclusion was incorrectly supported.

 

If you have a great QB and a solid OL, you can get by with Stevie Johnson or Deion Branch-type WRs. But if you have a Tom Brady of a QB and a Patriots-like OL, adding a non-washed-up Randy Moss to the mix will help significantly improve your offense.

Riffing on your general direction of thought... yes, the NFL has become more of a passing game. Part of that is by design, with rules changes to favor the passing game, to increase explosive exciting plays, and produce more scoring, and a more marketable, exciting product. Bill Walsh was ahead of the curve there, embracing those changes and transforming offensive football into the "pass to set up the pass" game it has become. Other coaches like Mouse Davis started introducing the spread formations, option routes, etc. into the professional level. It's rather ironic that it is Marv Levy who is so often quoted with his "run and stop the run" comments. It could be argued that his successful Bills teams were anything but built that way and actually were ahead of the curve as far as running a spread look offense and using a smaller, faster defense against the air forces of the AFC East, the look of most NFL offenses of today. Jim Kelly brought the run-n-shoot, spread concepts from the USFL to Buffalo (and would later in his career butt heads with Bresnahan and Shofner on an annual basis); and, while the smaller, quicker defense was ideal against the AFC teams of the day (Dan Marino, Dan Fouts, etc.), it was not built to stop the run as the mammoth OL of the NFC East demonstrated over and over in the biggest games of the season.

[shrader]

I don't like the way the undrafted free agents are included into that analysis. Comparing each of the rounds is fine since there are roughly an equal amount of picks in each round. There are far more undrafted free agent signings though, so that's going to inflate the number of starters out of that group a bit. That needs to be weighted in some fashion.