Writers of Pro Football Prospectus 2008

Most Recent FO Features

KellyChi13.jpg

» The Week In Quotes: September 19, 2014

Yes, people are still talking about Adrian Peterson, Greg Hardy, and Ray McDonald. But hey, we also get Chip Kelly exchanging pleasantries with reporters.

18 May 2007

Familiar Words

In Pro Football Prospectus 2006, we ran an article by David Lewin that introduced a new system for projecting rookie quarterbacks based on college statistics. As far as we know, this research was the first of its kind. We've talked a lot about the college quarterback projection system since then. We feel it is one of the signature pieces of Football Outsiders research, on the same level as DVOA ratings, the third-down rebound theory, and the inconsistency of fumble recovery.

As some of you may know, three weeks ago Gil Brandt wrote an article on NFL.com previewing quarterbacks for the draft. This article contained a section about projecting rookie quarterbacks based on college games started and completion percentage that resembled David Lewin's research.

Since this article ran on April 26, FO has not commented on the Brandt article publicly. However, it is now three weeks since this article ran. Although our editors at ESPN The Magazine are still trying to settle this matter with NFL.com, we feel that we need to say something before our research is permanently identified with another writer.

We hope that two article excerpts will help people understand why we feel so strongly about this issue, and why there is more to it than coincidence or similarity in concept.

First, these are the first two paragraphs in an article written by Aaron Schatz for the ESPN NFL Draft 2007 magazine. You can find it on page 31, at the start of the section on quarterbacks:

In 1998, the Indianapolis Colts faced one of the most important decisions in franchise history. With the No. 1 overall pick, would they take Manning or Leaf? Peyton proponents argued that he was more mature and accomplished. Leaf backers argued he had the stronger arm and more growth potential. But all the Colts really needed to know was four numbers: Manning started 45 college games, completing 63% of his passes. Leaf started just 24 games, completing 54% of his passes. The future couldn't be more clear.

It's true: College statistics really can predict NFL performance. For our 2006 Pro Football Prospectus, we studied 10 years' worth of drafts and discovered that the greatest indicator of NFL success for QBs taken in the first two rounds is the number of college games they started. Philip Rivers, for one, started 51 games at NC State. Donovan McNabb started 49 college games and Carson Palmer started 45. On the flip side, busts-to-be Joey Harrington (28), Jim Druckenmiller (24), and Akili Smith (19) had relatively little starting experience.

Second, this is the fourth paragraph in an article written by Gil Brandt on NFL.com, posted April 26:

There seem to be two important predictors for success when drafting quarterbacks. One is games started. Peyton Manning started 45 games and completed 63 percent of his passes. Ryan Leaf started 24 games and completed 54 percent of his passes. Philip Rivers started 51, Donovan McNabb 49 and Carson Palmer 45; all three have quarterbacked teams to playoff games. Joey Harrington started 28 games, Jim Druckenmiller 24 and Akili Smith 19. All were first-round picks in the past 10 drafts, two are no longer in the league and Harrington will be playing for his third team in three years.

Readers can come to their own conclusion about what is going on here, but we would like to point out two oddities:

  • The third paragraph of the ESPN Draft article, not printed above, talks about college completion percentage as the second predictor of quarterback success. The Brandt article introduces two important predictors for success, and then only addresses one.
  • The original ESPN rough draft listed McNabb with 45 starts, because Lewin's research only used regular-season games. A fact-checker at ESPN added McNabb's four bowl games, but did not add bowl games to the total for any other quarterback. In the Brandt article, Donovan McNabb is the only quarterback listed with bowl games included as collegiate starts.

Posted by: admin on 18 May 2007

285 comments, Last at 22 Apr 2009, 5:55pm by andy3887

Comments

1
by PackerNation (not verified) :: Fri, 05/18/2007 - 1:03pm

People can say or not say whatever they wish, but it's pretty clear what transpired here.

2
by pawnking (not verified) :: Fri, 05/18/2007 - 1:05pm

So, what you're saying is, a major news outlet plagurized and furthermore did a sloppy job of it. Water is also wet. Good luck becuase I think it's clear what's going on here, and even if it's "just" sportswriting, it's important to properly credit work that's not your own.

3
by coltrane23 (not verified) :: Fri, 05/18/2007 - 1:07pm

If this were a college course, Brandt would have received an automatic "F". Guess he must have gotten some tips from Ron Borges.

4
by McGayTrain (not verified) :: Fri, 05/18/2007 - 1:10pm

Pretty damning stuff.

5
by MCS (not verified) :: Fri, 05/18/2007 - 1:13pm

So, when the Outsiders are being plagerized by the insiders, does this mean you guys have made it?

Good luck in your fight.

6
by jimmo (not verified) :: Fri, 05/18/2007 - 1:13pm

damn, I read that article when it was posted and was suspicious without thinking too much about it. Now, side by side, and rereading Brandt, wow! It can hardly be more obvious. Go get 'em FO!!

(of course, on a selfish note, more publicity about the projection system means that more usually clueless dynasty Fantasy players will have some of the same info that I do!)

7
by Crushinator (not verified) :: Fri, 05/18/2007 - 1:15pm

This is flagrant, and it's frankly sad that the Outsiders have to go through all this crap just because they do their own research and make their own conclusions instead of all just parroting one another.

My respect for big sports media just dropped another notch, if that is even possible.

I just hope this all works out in the end and good luck guys.

8
by robopunterjoe (not verified) :: Fri, 05/18/2007 - 1:22pm

I read the Brandt article and thought that he might have borrowed the idea, but I had no idea that it was ripped off so blatantly until now. Good luck getting the credit you deserve for this.

9
by Seth (not verified) :: Fri, 05/18/2007 - 1:22pm

#2 - At the very least. He'd probably get placed on some sort of academic probation.

10
by White Rose Duelist (not verified) :: Fri, 05/18/2007 - 1:24pm

My apologies to ROBO-PUNTER for failing to change my username back before commenting. I had no intention to...

***CARRIER LOST***

11
by asg (not verified) :: Fri, 05/18/2007 - 1:26pm

This is not the only piece of FO research that has been plagiarized, although it is certainly the highest-profile (the other plagiarists did not write for NFL.com!). I have seen paraphrases-without-attribution of Aaron's work on predicting fantasy performance of team defenses on a fairly well-known fantasy football site, and I've also seen some plagiarism of the numbers regarding 390+ carry seasons for RBs and the likelihood of decline in the ensuing seasons.

12
by MJK (not verified) :: Fri, 05/18/2007 - 1:53pm

I read the Brandt article and my first thought was "this guy has been reading FO". Then when he didn't acknowledge FO, I was pretty annoyed. I know because FO is connected with FOX, and NFL.com has ties to CBS Sportsline, they might be considered "opponents", but even rival researchers are required to give credit.

On the other hand, I'm not sure what legal action FO could take. College statistics are in the public domain, and it's difficult to patent or otherwise protect a mathematical method of analyzing and reducing publically available data (that may change as IP law evolves, but currently algorithms and formulas are unprotectable).

Brandt and NFL.com seem to have clearly violated the implied (and possible explicit) copyright on FO's work, which protects the expression of ideas, not the ideas themselves, but even pursuing legal action on this front could be difficult because there are no "damages" per se--ESPN might have a case that the Brandt article reduced sales of their magazine, but it seems tenuous to me.

I'm not an IP law expert (I've taken some basic courses in it, but am no lawer)...maybe someone more knowledgable could weigh in?

The best possible outcome would be for NFL.com to print a retraction/apology acknowledging FO. Don't hold your breath, though. NFL has more lawers than ESPN or FO put together.

Glad this was brought to light, at least in this forum, though.

13
by zip (not verified) :: Fri, 05/18/2007 - 1:58pm

I knew this was going to end badly for Gil as soon as I read his article and noticed that he mentioned two predictors and then only wrote about one. I was like, "wow, that sounds like he took the FO article and didn't edit it down properly."

Good luck guys.

14
by Andy (not verified) :: Fri, 05/18/2007 - 2:02pm

My unsolicited advice: Punch 'em in the mouth as hard as you can.

15
by Nathan (not verified) :: Fri, 05/18/2007 - 2:02pm

So very rude, and disingenuous.

16
by Seth (not verified) :: Fri, 05/18/2007 - 2:07pm

#11 - While I wouldn't presume to speak for the Outsiders, I'd be personally surprised if they sought some kind of legal recourse. It's on NFL.com to do the right thing, as the Boston Globe did in the wake of the Ron Borges plagiarism incident: Print a public apology as you suggest, and suspend Brandt. Could they really be so arrogant as to *not* do this?

17
by The 42nd president (not verified) :: Fri, 05/18/2007 - 2:10pm

If you can't prosper honestly, you should at least cheat well. What a dope.

18
by Rich Conley (not verified) :: Fri, 05/18/2007 - 2:11pm

On a related note, Ron Borges is now working for the Globe again.

19
by Pacifist Viking (not verified) :: Fri, 05/18/2007 - 2:16pm

Really interesting. In my line of work, I occasionally have to go through articles and papers finding matches like this, and when it is this obvious, it is pretty clearly plagiarism.

And considering all Brandt would have to do is write something like "According to Aaron Schatz" before that paragraph, it's either sloppy or shady.

20
by Ryan Mc (not verified) :: Fri, 05/18/2007 - 2:21pm

I used this research last season to successfully predict a division title by San Diego, whilst others in my betting pool were turned off by Rivers' inexperience. I did, however, explain the source of my confidence in Rivers' pro prospects, and hopefully succeeded in directing a little more traffic this way.

Show no mercy lads! Brandt's actions are unacceptable.

21
by James, London (not verified) :: Fri, 05/18/2007 - 2:25pm

TMQ flagged this in his post-draft column.

He credited this to GIl Brandt initially, and then cam this addendum:

"Addendum: After this item ran, I learned that material similar to parts of Brandt's column previously appeared on the independent Web site FootballOutsiders.com, in the book "Pro Football Prospectus 2006," and in the publication ESPN NFL Draft. Original research into the relationship between college starts and pro quarterback success was conducted by David Lewin for "Pro Football Prospectus 2006" and FootballOutsiders.com; analysis was written for ESPN NFL Draft by Aaron Schatz, founder of Football Outsiders. I called Brandt, who said he had not seen their work. In his first career as general manager of the Dallas Cowboys, and in his second career as a sportswriter, Gil Brandt has compiled voluminous statistical research into football and its players, and become an important voice in helping the public understand college and professional gridiron. I can tell you Gil is a kind and generous guy, and repeatedly in his professional life has gone out of his way to help others. Many little-known college players from modest personal backgrounds never would have gotten their chance in the NFL without Brandt, while he has long made himself and his stats files available to writers and broadcasters at their convenience rather than his. But comparison makes clear Brandt should have credited the research in his column to David Lewin and the analysis to Aaron Schatz..

TMQ has it spot on. If Borges (or any other journalist) can't do this, niether should Gil Brandt.

22
by john (not verified) :: Fri, 05/18/2007 - 2:32pm

Re #11 - This is NOT legal advice. First, FO does have a copyright on content, regardless of formalities, at the moment the expression is fixed in a tangible medium, either on a website or on paper. Generally speaking, copyright law does not protect ideas or statistics. But creative arrangements and presentations of ideas can be protected. I like to think of a cookbook as an analogy; you can't copyright the list of ingredients and steps in a recipe, but you can copyright the pictures of the food on the same page.

FO has every right to be morally outraged by having their novel and innovative research lifted without proper attribution. But the numbers are public domain, and the ideas can't be copyrighted.

On the plus side, recognition from NFL.com may give FO some free publicity and well-deserved recognition. And the obviousness of Brandt's copying, combined with FO's vocal unhappiness with it, reduce NFL.com's reputation among the people who visit these kinds of sites regularly. Like us.

Keep on keepin' on, FO. Imitation is the sincerest form of flattery.

23
by Joe Rowles (not verified) :: Fri, 05/18/2007 - 2:39pm

I think it is rather sad that Brandt had to stoop to something so low to add to his own work. To be quite honest I have lost a lot of respect for him due to this.

24
by Joe Rowles (not verified) :: Fri, 05/18/2007 - 2:43pm

I also use the quarterback article when I talk on Denver's official board and use it to talk about what I think of draft prospects, but I also credit FO when I talk about it. Me and another poster on the bronco boards have pretty much earned a reputation as being FO guys because we credit you guys in our posts so much.

25
by ChrisS (not verified) :: Fri, 05/18/2007 - 2:52pm

That sort of behavior is very low. I hope FO gets the appropriate retraction/apology from Brandt/NFL. I do want to congratulate you on the non-confrontational low key approach to solving the problem (waiting three weeks to publicize the dispute), very professional.

26
by wrmjr (not verified) :: Fri, 05/18/2007 - 2:59pm

It seems to me the best thing we can do as readers of this site and as fans of the NFL is to let the league know we are displeased with Brandt's actions. At the bottom of the help page on NFL.com (linked at my name), they have a "Contact Us" section. There may be a more appropriate address as well. I'd think some respectfully-phrased displeasure from fans might be a worthwhile effort. What do others think?

27
by Chris G (not verified) :: Fri, 05/18/2007 - 3:13pm

I read Brandt's piece and thought it was very un-original, but I had no idea! As I was reading I kept expecting FO or DL to be credited at some point. When it didn't happen I thought Brandt was a hack. Now it looks like he's a theif...and a hack. Hopefully you'll be able to get some sort of appology from nfl.com. Good luck guys.

28
by Oldcat (not verified) :: Fri, 05/18/2007 - 3:21pm

This kind of thing happened to Bill James in the 80s in baseball. The 'official' guys copied lots of his work and techniques without attribution. Hope it works out better for you.

29
by Oldcat (not verified) :: Fri, 05/18/2007 - 3:22pm

If he starts writing about "AUTO_PUNTER" then we will know for sure!

30
by Fergasun (not verified) :: Fri, 05/18/2007 - 3:23pm

It'd be interesting to see the number of people willing to defend NFL.com and Brandt. At least any publicity is good publicity, I think "NFL.com writer Gil Brandt steals work from Pro Football Prospectus 2006" and "Pro Football Prospectus 2007 for sale now!" is a good marketing juxtaposition.

There's a reason I read Football Outsiders and never saw this nfl.com article...

31
by Sophandros (not verified) :: Fri, 05/18/2007 - 3:24pm

11: I read the Brandt article and my first thought was “this guy has been reading FO�. Then when he didn’t acknowledge FO, I was pretty annoyed. I know because FO is connected with FOX, and NFL.com has ties to CBS Sportsline, they might be considered “opponents�, but even rival researchers are required to give credit.

FO comments area whipping boy Gregg Easterbrook gives credit whenever he gets a concept from elsewhere.

32
by Will Allen (not verified) :: Fri, 05/18/2007 - 3:31pm

Frankly, I've had a bad feeling about Brandt for a long, long, time, going way back to his days as a Cowboys front office guy, and I was never part of the anti-Dallas contingent until Jerry Jones split with Jimmy Johnson, and I left that contingent with the hiring of Parcells.

In other words, I generally had a favorable impression of the organization in the Landry days, but I always got the feeling that Brandt marketed himself as a guru to a greater degree than what his actual contribution was. This incident confirms the feeling. Really shameful, if one assumes that Brandt is capable of shame. The fact that an attribution and apology has not yet been issued indicates to me that Brandt lacks such capacity, and thus it is entirely fair for me to conclude that Brandt is a dishonest jerk without anything approaching a sense of integrity.

Shame on the thief.

33
by Rex Hudler (not verified) :: Fri, 05/18/2007 - 3:34pm

Here's the terms for settlement of this matter: the NFL lets FO put the NFL logo on the new Prospectus.

34
by mmm... sacrilicious (not verified) :: Fri, 05/18/2007 - 3:43pm

#32: But if they keep the name abstracted as "Pro Football..." then they can make jokes about Ricky Williams in the CFL.

Congrats for keeping it low-key, FO. It shows a lot of class to get dicked over and not throw a fit.

35
by SJM (not verified) :: Fri, 05/18/2007 - 3:59pm

Wait wait wait...

Are bowl game starts supposed to count in the formula or not? If not, why not?

(Uh, Gil Brandt is evil, etc.)

36
by NewsToTom (not verified) :: Fri, 05/18/2007 - 4:01pm

I'd like to second john's comments in #21.

"Morally", I share the general belief that Brandt's article was likely derivative and the copying rather shameless. I'm touched by the apparent outrage shown by some commenters, and their belief in the rapidity of harsh punishment that will surely befall this incident. Such has not necessarily been my experience, though perhaps those students all reached the exact same wording as was on the answer key on their own. I also would not rank this as bad as the recent Zaslow-Couric affair on the severity scale.

I was (mildly) surprised to see comments open on this post.

37
by Richie (not verified) :: Fri, 05/18/2007 - 4:11pm

I can't find the Brandt article on NFL.com. I guess they pulled it?

38
by David Mazzotta (not verified) :: Fri, 05/18/2007 - 4:23pm

You know, the pathetic thing is that all he really had to do was insert a line like:

"As discovered by the researchers at Football Outsiders..." and maybe include a link.

It would have done nothing to denigrate his article and it would have given credit where due.

39
by SocioJoe (not verified) :: Fri, 05/18/2007 - 4:37pm

http://www.nfl.com/help/emailtech

....is where you can send an angry message.

"Fire the plagiarist", etc...

40
by zip (not verified) :: Fri, 05/18/2007 - 4:38pm

#37

Since his article was mainly a condensed summary of FO's work, citing it would have made his article look very weak. It's not like he was adding any value.

41
by Richie (not verified) :: Fri, 05/18/2007 - 4:44pm

#39, isn't that standard? A writer is writing something, and uses a piece of a larger work to add to HIS comment, not to add to the original work.

42
by DrewTS (not verified) :: Fri, 05/18/2007 - 4:46pm

Re 37

I often wonder why so many writers get themselves in trouble this way. It's not illegal, or even looked down upon, to borrow stuff from somewhere else. All you have to do is say where you got it and no one will care. FO probably would have been happy to be cited in an NFL.com article.

What I think is even worse was this bit from the TMQ article --

I called Brandt, who said he had not seen their work.

In other words, he didn't just have a brainfart and forget to give credit. He's choosing not to. Someone explain that to me.

43
by JT (not verified) :: Fri, 05/18/2007 - 4:52pm

This is pretty shameful, good luck in getting some justice on this one FO. My pre-order of the next PFP will come as soon as you start selling it.

44
by JT (not verified) :: Fri, 05/18/2007 - 4:55pm

By the way, please keep us posted on how this is resolved.

45
by J (not verified) :: Fri, 05/18/2007 - 5:01pm

Wow... this is deeply ironic coming on the heels of the NFL's placing heavy restrictions on how third-party websites are allowed to use NFL audio and video content.

46
by CA (not verified) :: Fri, 05/18/2007 - 5:03pm

Re: 20 / Easterbrook

I called Brandt, who said he had not seen their work.

If this is true, I am quite disturbed by the fact that a professional football writer is ignoring Football Outsiders, who are doing the most important work in football analysis that is widely publicly available. If you want to understand football, and, even more so, if you claim to understand football, you should be reading the work of the Outsiders. Period.

Let me suggest one possibility that would cast Brandt in a slightly less negative light: Perhaps Lewin's ideas have become so pervasive in NFL circles that Brandt obtained the ideas originally developed by Lewin through another source. Then, perhaps a research intern pulled the specific data from Aaron's article. If that were the case, it's still lazy. It's still plagiarism. It's still bad journalism. But it's not necessarily as callous as most of us are assuming. Just throwing that out there as a possibility before we all bust out the torches.

By the way, see the link in my name for a brief apology by Brandt to Rick Gosselin for supposedly accidentally borrowing his work. That was published on NFL.com today.

47
by MadDogExtra (not verified) :: Fri, 05/18/2007 - 5:14pm

#45. It still does not explain why he only list one out of 2 predictors. Either he recognized that it would be too obvious if he lists both or somebody had borrowed his PFP 2006 and he could remember what the last one is all about.

48
by MadDogExtra (not verified) :: Fri, 05/18/2007 - 5:16pm

he could -> "he couldn't"

49
by Richie (not verified) :: Fri, 05/18/2007 - 5:20pm

By the way, see the link in my name for a brief apology by Brandt to Rick Gosselin for supposedly accidentally borrowing his work. That was published on NFL.com today.

Wow. That's bad. That means he published two stories within two days that were based on plagiarism.

Yeah, he "accidentally" used Gosselin's list. How does that work? He couldn't tell the difference between his own writing, and a list printed in a newspaper?

50
by zip (not verified) :: Fri, 05/18/2007 - 5:20pm

#40

I'm saying that he didn't do anything except regurgitate their work. There is no "HIS Comment" except a summary of FO's stuff, which is why the cite would make his uselessness apparently to anyone who followed it.

51
by Wes M (not verified) :: Fri, 05/18/2007 - 5:21pm

Article is missing, (or very hard to find,) and interesting enough, there is an "oops, I'm sorry" article (posted today,) in regards to Brandt accidentaly using someone else's Top 100 list for a draft article...

52
by Richie (not verified) :: Fri, 05/18/2007 - 5:21pm

I used to like listening to Brandt on Sirius radio, but now all I'm ever going to think of when I hear him is that everything he says might be parroting somebody else's original thoughts.

53
by Pat (not verified) :: Fri, 05/18/2007 - 5:30pm

How is the Gil Brandt article hard to find? It's linked in the FO story above.

54
by Wes M (not verified) :: Fri, 05/18/2007 - 5:44pm

"How is the Gil Brandt article hard to find? It’s linked in the FO story above. "

True, but it isn't linked on Brandt's NFL.com index page. You have to go through links along the Draft -> Features -> Analysis -> Brandt (QB)

55
by philly bill (not verified) :: Fri, 05/18/2007 - 5:44pm

And we should be surprised at the disingenuous behavior of a longtime Cowboys employee... why?

Once a Cowboy, always a might-makes-right crook.

56
by NewsToTom (not verified) :: Fri, 05/18/2007 - 5:47pm

Another possibility, suggested by the Zaslow-Couric affair I noted in #35, is that Brandt didn't actually produce the content published under his name, but it was produced by someone else, and Brandt then reviewed it and it was posted on NFL.com. In and of itself, this doesn't strike me as a big deal, though that may be because I frequently for work do similar things. This is a classic agency problem, and the risk/reward tradeoff for the producing party may not seem that great.

The Gosselin confusion is annoying, but it's just carelessness and thoughtlessness (I'd hope you could remember that yes, this person was on my Top 100 list, or no, he was not). This is a different deal.

57
by Will Allen (not verified) :: Fri, 05/18/2007 - 5:55pm

It's one thing to do a soft news commentary piece, and have it published under the byline of a famous "journalist" although the plagiarism is never excusable. It is quite another to put out research-based analysis under the byline of a supposed expert who did not do the analysis. Either way, Brandt is exposed as a fraud.

58
by Karl Cuba (not verified) :: Fri, 05/18/2007 - 5:59pm

Isn't Brandt quite an old man? I personally don't feel right laying into an OAP. He made a mistake and should make some form of atonement but a few of the comments here are a little extreme.

59
by Will Allen (not verified) :: Fri, 05/18/2007 - 6:10pm

Like I said, Karl, given that more than three weeks have elapsed, and Brandt has been asked about it, and he still hasn't made an attribution, I think it is entirely fair to draw some pretty harsh conclusions.

60
by Scot K (not verified) :: Fri, 05/18/2007 - 6:13pm

Oh, COME ON!!!

I can appreciate (to a certain extent) the issue of Brandt tightly paraphrasing the previous article.

But is anyone seriously contending that the determination that games started and completion percentage is some miraculously original metric that requires attribution? I don't think it is. It's an interesting CORRELLATION. There is no CAUSATION here. I don't think FO can reasonably have any property right (moral, intellectual, or otherwise) in pointing out a correllation.

I understand that in today's blogosphers, everyone is very protective when they turn a good phrase or come up with an idea they THINK is highly original. Maybe Brandt was lazy, and we should recognize that and take his analysis accordingly.

But the idea that FO has some claim to moral indignation is untenable, as far as I'm concerned.

61
by Will Allen (not verified) :: Fri, 05/18/2007 - 6:26pm

Tell ya' what, Scot; show me somebody prior to Lewin who specifically identified those two particular stats as being the critical two in predicting pro performance, and then it will be fair to say that Brandt was just using ideas that have been part of the common body of knowledge. Until then it is entirely fair to conclude that he is presenting other people's analytical work as his own. As for indignation, well it is just football, but if I was trying to figure out a way to make a living doing football analysis, and somebody used my work in the way that Brandt had, it'd make me mad. Nothing like dipping into your own wallet to change the intensity of feeling.

62
by Plagiarism is Bad. (not verified) :: Fri, 05/18/2007 - 6:30pm

Oh jeez!!

I can acknowledge (kind of) the problem with Brandt closely quoting the earlier article.

However is anyone sincerely arguing that the conclusion that games started and completion % is some amazingly original standard that demands citation? I don’t think so. It’s a compelling CORRELLATION. There is no CAUSE->EFFECT here. I don’t think Outsiders can seriously have any property right (intellectual, moral or otherwise) in showing a correllation.

I "get" that in today’s tubenets, all are protective when they turn a great point or come up with an idea they BELIEVE is highly innovative. Perhaps Brandt was comatose, and we ought to recognize that and take his analysis as such.

But the concept that FO has claim to moral indignation is ludicrous, as far as someone else is concerned.

63
by QB projection systemjoe (not verified) :: Fri, 05/18/2007 - 6:32pm

#60: "But is anyone seriously contending that the determination that games started and completion percentage is some miraculously original metric that requires attribution? I don’t think it is. It’s an interesting CORRELLATION. There is no CAUSATION here. I don’t think FO can reasonably have any property right (moral, intellectual, or otherwise) in pointing out a correllation."

Why the hell not?! It doesn't matter whether it's an example of causation or correlation. The fact that success, starts, and completion percentage are connected in such a way that allows prediction of the first from the other two is definitely original, if only because of its precision. Gil Brandt would not have known about that correlation, or about how strong it is, if not for David Lewin's research. That means he should've cited that research if he was going to use it in an article.

64
by Alex (not verified) :: Fri, 05/18/2007 - 6:36pm

Oops, forgot to change my name back. Ignore that last post. Well, the name anyway.

65
by Horacio (not verified) :: Fri, 05/18/2007 - 6:36pm

The answer to this is easy: we should all send angry mail to NFL.com until they do something. I don't care if I'm sending a daily angry missive to them. The damn article is exactly the same as FO's and, even if Brandt says he hadn't seen FO's article, it was published before his own and, therefore, he should be crediting FO for creating this analysis.

Of course, it's impossible because Brandt doesn't have A DAMN MAILBOX. Yeah, pretty easy, Gil, stealing stuff and then not having to answer about it.

In ANY university IN THE WORLD in which I would have presented a thesis statement so similar to another one, I would have been dubbed for plagiarism and expelled immediately. Then again, it's a common practice these days. Len Pasquarelli has been claiming through his famous phrase "ESPN has learned..." to break out news which have been previously reported by other media outlets (particularly, Sirius Radio has been jobbed several times by him). Ron Borges is another example.

Too sad football journalists can't be professional journalists.

66
by Sergio (not verified) :: Fri, 05/18/2007 - 6:38pm

re: 59

The fact that there's correlation or causation in the phenomenon has NOTHING to do with the fact being discussed - namely, that Gil Brandt most likely pulled his article off the ESPN one.

As far as property rights go, it has already been discussed in the thread (see posts 11, 21) that the Outsiders might not have a LEGAL case against the NFL/Brandt. However, it's still in poor taste (and it shows a lack of journalistic integrity) to so obviously take something out of someone else's work, and not even mention them as a reference. You can make the case that the variables in question (games started/completion %) and their correlation with NFL success are so simple, in hindsight... yet, the first time we heard about it was when David Lewin brought his research to FO. There's actual evidence to that (publishing dates of Lewin's first article, PFP 2006, etc.), so it's highly careless to dismiss it simply as something that "they [FO] THINK is highly original", and stating that "the idea that FO has some claim to moral indignation is untenable".

This is not a matter of IP (at least legally), IMO. This is a matter of integrity, respect, and most of all, good manners. Feel free to think it's not worth of your attention, but as someone that has published original work of any kind, I am compelled to feel dissapointed of Mr. Brandt, and I hope he does the right thing and gives credit where credit is due (i.e. Aaron Schatz, ESPN NFL Draft 2007, Football Outsiders and/or David Lewin), as he should've in the first place.

67
by NewsToTom (not verified) :: Fri, 05/18/2007 - 6:40pm

Oh, Come On, to you, too.

As noted previously in the thread, legally FO cannot have a copyright in facts or in the idea. What they have is a copyright in the expression of those facts. Your personal feelings (and mine) aside, there is a definite legal issue with Brandt's chosen expression of the idea.

Perhaps the most interest thing about Lewin's research is not that experience and completion percentage matter, but within the bounds of the study (1st/2nd round), they are the only things that matter. Given the few variables, this is the more reason to be vigilant when one in particular is in some sense attacked.

And on causation, who cares? Given the nature of the analysis, I don't see how it could matter.

68
by Insancipitory (not verified) :: Fri, 05/18/2007 - 6:45pm

I think no matter how this particular incident turns out FootballOutsiders is destined to prevail, why? As pointed out, that's one of an array of useful predictions, and there may well be new predictions coming as well as incrimental improvements to the older ones. Gil Brandt's already revealed his inability to produce them. Quite frankly, there just aren't a lot of options for this kind of penetrating insight into the game we love. So, I guess, Thanks.

69
by Scot K (not verified) :: Fri, 05/18/2007 - 6:47pm

RE #60

Even IF I conceded that Lewin was the first to PUBLISH the correllation, that wouldn't justify the implict and explicit condemnation that seems to be so prevalent amongst bloggers.

FO didn't create the stats from which it's correllation arose. Rather, FO just opened up a stat book and made an observation. Anyone was free to make the same observation. FO doesn't have any ownership interest in statistics.

I'm not saying that FO's indignation is out of touch with the general attitude in the blogosphere. I'm just saying that the general attitude in the blogosphere is pretty ridiculous.

Take a fantasy football example. Someone, somewhere, first observed the correllation in the "Third-Year WR" breakout. Someone was first, and they got it from looking at stats. But you never see fantasy writers "attributing" it to the all knowing guy who first observed it (or rather, wrote about it).

At what point does the "No Sh*t factor" come into play? There is NO logical distinction between my example and the correllation pointed out by Brandt.

70
by NewsToTom (not verified) :: Fri, 05/18/2007 - 6:49pm

Re #64
In ANY university IN THE WORLD in which I would have presented a thesis statement so similar to another one, I would have been dubbed for plagiarism and expelled immediately.

I wish people would stop reading university press releases without thinking. Yes, some schools are willing to expel students for plagiarism violations, but not all of them, and many offenses more serious than Brandt's result in nothing worse than a failing grade in the class and/or being placed on academic probation. Judge Posner's book on this subject is apparently rather decent, and his identification of moral shaming as the proper solution well-chosen in my opinion.

71
by JT (not verified) :: Fri, 05/18/2007 - 6:56pm

#61: brilliant.

The stealing of what seems to have been a previously undiscovered idea isn't the only give away. The extremely similar article phrasings (same order lists of same QBs) and the bowl-game anomaly are dead give-aways. He used FO research without attribution. Shouldn't be thrown in jail, but definitely should apologize with link to FO. Wonder if someone can get PFT on this, they love bashing "real" media. Wait, doesn't one FO writer have a connection there...? Hmm.

72
by Sidewards (not verified) :: Fri, 05/18/2007 - 7:02pm

RE: 67
I think it's alright to be indignant when the error has been pointed out and not corrected or paid attention to.

Especially when Brandt uses the exact same quarterbacks as examples, as well as several similar phrasings.

And if it's such a commonplace thing for everyone to have figured out the facts presented in the articles on FO, why haven't we seen it more?

73
by john (not verified) :: Fri, 05/18/2007 - 7:07pm

I don't think you can dismiss possible causation so easily. The more starts a quarterback has in college, the more faith that player's coach had in that quarterback from an early stage. That means the quarterback got it, and got it early. And he had the physical skills to start at age 18 or 19. Don't you think that that kind of aptitude and physical maturity correlates well to success in the NFL?

Completion percentage is another indicator that the quarterback knows what he's doing, and what he's supposed to be doing. Being able to communicate with a coach, understand his job, execute it faithfully and consistently ... that sounds an awful lot like Tom Brady, Peyton Manning, etc., to me.

The fact that the kid played a whole bunch just means you have more data to work with, and more data is always better.

Oh, and I should be clearer about my point earlier. Even if an idea IS novel, it will not be copyrightable. You might, however, think about a patent ...

74
by Scot K (not verified) :: Fri, 05/18/2007 - 7:13pm

I just did some research.

1. No QB under four feet tall has ever won the superbowl.

2. No QB over three hundred pounds has ever won more than 50% of his starts in the NFL.

No one has ever done this research in this way before. Clearly, THE critical indicator for success in the NFL is that a QB must be over 4 feet tall and under three hundred pounds.

FIRST!!!!

Thank you. I expect to see attribution in any future publication that suggests height and weight are critical indicators of success for an NFL Quarterback.

75
by Richie (not verified) :: Fri, 05/18/2007 - 7:17pm

Take a fantasy football example. Someone, somewhere, first observed the correllation in the “Third-Year WR� breakout. Someone was first, and they got it from looking at stats. But you never see fantasy writers “attributing� it to the all knowing guy who first observed it (or rather, wrote about it).

And it may not even be true. Click my name for article.
http://www.pro-football-reference.com/articles/wr3.htm

76
by MJK (not verified) :: Fri, 05/18/2007 - 7:19pm

Scot,

If Brandt had just written an article about how QB completion percentage and college starts point to pro success, then things would be fine and I'd buy your "No Sh*t factor" explanation. FO and David Lewin would be entitled to feel smug about noticing, and publishing an article describing, this same correlation a year ahead of the "mainstream media" and nothing more. (Something similar happened when Troy Aikman began publishing his "efficiency scores", which are fairly similar to, although not as sophisticated as, DVOA. However, this was not a problem because it was clear that "efficiency scores" were NOT derived directly from DVOA, but rather were based on some of the same good ideas but were independently derived and uniquely expressed. Aaron was "smug" that a mainstream media source was following in his footsteps, but there was no moral outrage that Brandt is getting). As several posters have commented, the data is not protected IP, nor is the idea of the manner in which to use the data.

However, what is protected, both legally by copyright law and ethically by journalistic standards regarding plaigarism, is the manner in which an idea about some data is expressed. The fact that Brandt cited the exact same eight QB statistics in the exact same order to elucidate the point, and the fact that the journalistic structure of his prose was eerily similar to FO's works, indicates a very clear case of plaigarism and a violation of copyright law. In short, he was expressing the same idea that FO had expressed, using the exact same examples in the exact same order, but just with a couple of words changed around. That's plaigarism and is sufficient to merit failure or explusion at every high school and college in the country.

The fact that the statistics that Brandt cites contain the same randomly introduced irregularity (bowl starts for McNabb) as FO's statistics serves as evidence that the author of the article (who I will concede is not necessarily Brandt) almost certainly was aware of FO's work, and yet passed the article off as an original work, which supports the supposition that the plaigarism was intentional, which is even worse (plus it indicates poor and lazy journalism).

And, legally, it is the burden of the infringer to establish prior art. If Brandt was citing an idea that is considered to be well established and worthy of the "No sh*t" clause that you define, then the tone of his article should have so indicated, even if he did not explicitly say something like "it has long been known that completion percentage and college starts point to pro success". He did nothing of the sort...the tone of his article implies that he is presenting new and original ideas.

77
by NewsToTom (not verified) :: Fri, 05/18/2007 - 7:26pm

Just for the record, Tom Brady only started 25 games at Michigan and had a completion percentage of 62.3% (thanks, UM website). Then again, he was a 6th round pick, so Lewin's system doesn't even apply to him.

I think what 30 games correlates to that is the closest thing to causation we have is that the player (i) was good enough to start for the better part of 3 years and (ii) has the physical talent to be drafted highly. The question for me, then, is whether there's anything magical about 30 games, or whether "started 6 games+ in at least 3 seasons" would be any better. I suspect the answer is that those are not interestingly different, but am too lazy to confirm that suspicion.

78
by MJK (not verified) :: Fri, 05/18/2007 - 7:30pm

Scot,

If I write an article that says

"I've figured something novel out. Because no QB under four feet tall has ever won the superbowl, and no QB over three hundred pounds has ever won more than 50% of his starts in the NFL, height and weight are important for QB success".

Then you can feel free to feel morally outraged at me. That's essentially what Brandt did. On the other hand, if I wrote:

"The majority of successful NFL QB's fall within a relatively narrow spectrum of physical characteristics. While there are outliers, such as Doug Flutie succeeding despite his less-than-prototypical-QB height, the QB's height and weight apparently count for something."

Well, you could criticize my bad grammar and poor sentence construction, but you would have no case to feel outraged at me stealing your idea. Of course, you are obviously not the first person to consider height and weight as indicators of QB success (I can cite hundreds of scouting reports discussing QB physical characteristics predating your comment, but can you cite an article pre-dating FO's talking about using starts+completion percentage to predict a QB's pro career? David Lewin may well be the first media analyst to do so in print). But if you genuinely were the first person to suggest considering height and weight of a QB, and if you had published that material in a publically accessible manner, then I probably would be required to acknowledge you with something like "considering height and weight as indicators of QB success was first proposed by Scot K; however, the conclusions he draws are so general as to be practically useless. However, we can refine this method ..." and so forth.

79
by SJM (not verified) :: Fri, 05/18/2007 - 7:30pm

Scot K,

That's a ridiculous argument and you know it. Your example is completely unlike Lewin's research. Lewin discovered a high correlation between certain stats and certain other stats which may be a big help in identifying successes and failures in the draft.

You made something up which is probably true, but is utterly useless because it has no practical use. You also did not discover that height and weight are important, since scouts have known that from time immemorial.

Scouts have not known that starts are important (or they wouldn't keep drafting guys with

80
by Scot K (not verified) :: Fri, 05/18/2007 - 7:32pm

MJK;

Yeah, I see what you mean. I think you are probably right that this was lazy writing by either Brandt or an NFL.com flunky who works for him.

But on the other hand, FO has even less of a moral claim to the particular "phrasing" of its article than it does to the "no sh*t" methodology. There is nothing particularly creative, unique, or original about FO's use of those examples, particularly to the extent that they are undoubtedly the examples that best illustrate the correllation. We aren't talking about Shakespeare here. There are only so many ways to express an idea in a cogent sentence. No one owns a copyright or any other moral claim on the sentence "The American Flag is Red, White, and Blue," and I wouldn't get kicked out of any high school or college in the Country for writing that sentence, despite the fact that it has undoubtedly been published previously.

81
by SJM (not verified) :: Fri, 05/18/2007 - 7:35pm

[post got cut off]

less that 30 starts so high), and if they did, they did not know that that can be combined with completion % to predict success accurately, and even if they knew all that, they didn't make it public. Lewin was possibly the first person to discover these things and definitely the first to publish them (and being the first to publish is very important. Ask any researcher).

Brandt "discovered" that starts are a good predictor, and published that well after Lewin and Schatz. That's fine, except that there are indications that he actually stole their presentation, not just their content. He didn't coincidentally use the same quarterbacks in the same order with the same copy-edited inconsistency, he copied it.

What I'm saying is that if some jackass wants to write an article claiming that starts are a good predictor for success without giving a citation, that's fine, but he better not copy the exact style of presentation also.

The only good to come from this is that it will become news which will cause publicity for FO.

82
by SJM (not verified) :: Fri, 05/18/2007 - 7:38pm

Scot K,

You're still not getting it.

Writing an article about how starts are important, and citing Philip Rivers and Michael Vick, is OK. Writing that same article, and citing the exact same list of QBs in the exact same order is plagiarism if it is not cited.

The McNabb bowl game anomaly is just further proof that the material was lifted directly from someone else's work. That is the very definition of plagiarism.

83
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Fri, 05/18/2007 - 7:41pm

Also, I will henceforth be known as "The McNabb Bowl Game Anomaly."

84
by andy (not verified) :: Fri, 05/18/2007 - 7:43pm

someone talked about how gil brandt's plagiarism is ironic considering the nfl's strictness regarding their own content usage. going off on a tangent on that subject - it's all about the money baby! there's an article i read on either espn.com or si.com/nfl (forgot which, is that plagiarism? hehehe) stated that high round draft picks are required to attend a reebok/nfl marketing event even if they miss an OTA. tony dungy called it 'troubling.' typical nfl.

85
by Randy S. (not verified) :: Fri, 05/18/2007 - 7:51pm

80 - That's a great name.

I wonder why ProFootballTalk isn't all over this. They trash other writers for plagiarism, why not Brandt?

86
by Scot K (not verified) :: Fri, 05/18/2007 - 7:53pm

SJM;

No, I do get it.

In your post, you cite a "standard" that plagiarism is proven because Brandt cited the exact same QB's in the exact same order. But you weren't the first to point out the standard, and you aren't attributing it. But its okay, because there are certain "ideas" that are well accepted or "obvious" enough to someone that they don't need attribution. I'm okay with that. You aren't.

87
by Scot K (not verified) :: Fri, 05/18/2007 - 8:00pm

MJK;

See, now we are just engaged in line drawing, as in "When is an idea so original that it requires attribution, and when is it an obvious correllation that is in the public domain.

Once you start down that road, you are acknowledging that reasonable minds could differ on where the line is drawn. Once you acknowledge that, it takes it out of the realm of the explicit moral outrage that is being expressed on this thread, and the implicit outrage in the FO "acknowledgment" of the issue.

I read Easterbrooks "mea culpa" on this, and his gentle scolding of Gil Brandt, and kinda thought Easterbrook was off base. But I was quite impressed that FO , until today, hadn't been wringing its hands over this highly debatable instance of "plagiarism."

88
by asg (not verified) :: Fri, 05/18/2007 - 8:06pm

... so, Scot K, if someone were to photocopy the entirety of, say, FO Prospectus 2007 and sell it as their own work, you'd have no problem with that? You wouldn't call that plagiarism? It's all just looking at stats and making observations, right?

89
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Fri, 05/18/2007 - 8:10pm

Scot K

The standard definition of plagiarism is "common knowledge" and is a good example of something that does not need to be cited. An exact list and order of QBs used to demonstrate a theory about starts cannot be common knowledge. How can everyone obviously know a certain list in a certain order, with one odd statistical anomaly, used to demonstrate a theory about starts? The concept doesn't make sense.

You can argue that the importance of starts is common knowledge (though I would dispute it, I admit it's a defensible position). But you can't argue that a list and order of QBs is common knowlege.

Again, if Brandt had used a different list and not included the odd McNabb anomaly, then it would not be plagiarism (it would just be bad journalism and unethical). But he made it apparent that his material was lifted.

If I write an article about how the Redskins are poorly managed, then that's common knowledge. If I write a similar article, but every example I use is the same and in the same order as a similar, previously published Washington Post article by Howard Bryant, then that's plagiarism. Do you see the difference?

90
by Insancipitory (not verified) :: Fri, 05/18/2007 - 8:21pm

83, Scot K, you're illustrating the difference between pedantic and erudite. The summery at the top points it out, everyone reading this thread knows that, it's perfectly reasonable to expect that to be common knowledge of this audiance. In Brandt's swiping, it's not perfectly reasonable to assume his audiance was familiar with the source material, in his case, he was might have been counting on the fact that they wouldn't be.

As for not original or too 'facty', google "statistical method" or "statistical process control" on google's patent search. Be in awe of all those who disagree with you. Of all the variables that consipire to describe a college QB, height, weight, body mass, arm strength, accuracy, surrounding talent, 40 times, pass distribution, likely dozens if not hundreds more, 2 were found that correlate well with NFL success and failure. More over one of them was overlooked and given lesser significance by the world at large. That combination to create a statistical tool, IS new and non-obvious. It's probably on the order of 1 in 10,000 of possible 2 variable combinations.

Then if that weren't enough, Brandt didn't just pick a good idea, a better idea than he'd had, from a less widely known peer, he took that exact same expression of it, adding nothing. It's the kind of thing that's just unclassy enough to get one kicked out of San Deigo.

91
by ardak (not verified) :: Fri, 05/18/2007 - 8:29pm

Anyone have any idea what email addresses we should be firing off angry mail to about this issue? Can't seem to find any email addresses on the nfl.com website.

92
by Scot K (not verified) :: Fri, 05/18/2007 - 8:36pm

SJM;

I hear what your saying, but it all depends on context. If Howard Bryant writes an article about how poorly the Redskins are managed, he doesn't have any proprietary or creative interest in the order in which his examples are listed. Even if you blatantly and intentionally listed the examples in the same order, it's nothing Bryant should get in a twist over.

With respect to the "examples" you might use to demonstrate poor management, whether the use of the examples constitute's plagiarism is all a matter of context. If the examples are well known, then Bryant shouldn't get a moral interest in those examples just because he was the first to publish.

I know alot of people get in a twist when ESPN does it's "Len Pasquerelli has learned" spiel. But as a Non-Blogger, Non-Journalist, I just don't care. Trust me - my estimation of Len goes NOWHERE based upon that stuff.

This is not to say that a reporter who gets a "scoop" shouldn't be given credit, but the endpoint at which that credit must no longer be attributed varies in relation to the difficulty and/or novelty of the information. For example, I think it was right for Woodward and Bernstein to get attribution for facts they discovered ... back in 1973.

But at this point, does a journalist still need to attribute them whenever they write about fact that those guys were the first to find? I don't think anyone can seriously say that's the case. If it is the case, then ALOT of journalists plagiarize every day.

I just don't think Lewins research is that so "original" that it deserves a high degree of protection. It probably took him about five minutes to do that research.

Now, if someone tried to pass the DVOA and DPAR methods, and passed them off as their own, FO should call a lawyer. But that is a self-contained methodology that FO spent years creating. It's not just a compilation of stats, its a particular way of looking at stats, that took years of research and tweaking.

You and I both agree that there is line here, and that the line can be murky. We both agree that Brandt's work is close to the line. You see it on one side and I see it on the other.

93
by Trogdor (not verified) :: Fri, 05/18/2007 - 8:38pm

You could always put this quote on the cover of PFP '07:

"The groundbreaking research of the Football Outsiders is simply tremendous. Whenever I want to seem creative or insightful, they're the first ones I rip off." - Gil Brandt, NFL.com

If he objects that he didn't actually say that, you can respond that he didn't actually make any of the discoveries he presented as his own, either.

In other news, I've decided that I'll be reprinting the game charting stats in PFP, and re-packaging them as my own work. Why not? Granted, I didn't actually do any of the work, but the possibility existed that I could have. Really, anyone could have - all the games were available to anyone. The fact that I didn't develop those stats is absolutely no reason that I shouldn't present them as if I did. After all, there is a certain "no sh*t" factor involved, if you just watch close enough, right?

94
by Richie (not verified) :: Fri, 05/18/2007 - 8:44pm

ardak, I used this link to send my complaint:

http://www.nfl.com/help/emailtech

95
by Sid (not verified) :: Fri, 05/18/2007 - 8:46pm

When I saw TMQ bring it up (Brandt's "innovative" system), I was highly suspicious. Now I'm even moreso.

96
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Fri, 05/18/2007 - 9:06pm

Scot K,

You are right that after a while, "exclusive" information becomes part of the common domain and does not need to be cited.

However, I think David Lewin would be extremely offended by your suggestion that he cobbled his system together in five minutes. That is simply not the case, and I don't think there's a question that nobody besides him, the outsiders, and their readers even knew that it existed prior to the ESPN magazine article in question. (Certainly there is no way that anyone knows his exact formula, because he has never published it. Even if he had or does, that will never become public domain because that aspect of it is clearly his original idea. But that's just a side point.)

Further, I don't like your suggestion that the plagiarism line is murky and that I think Brandt is close. Lifting another person's material is plagiarism, flat and simple. That includes both original research and presentation, and Brandt apparently stole both. I don't consider Brandt's article to be "barely plagiarism" or "just over the line." It just is, and I don't think there can be and question.

97
by Scott de B. (not verified) :: Fri, 05/18/2007 - 9:11pm

But at this point, does a journalist still need to attribute them whenever they write about fact that those guys were the first to find? I don’t think anyone can seriously say that’s the case.

Absolutely. In my line of work, I am careful to properly attribute ideas and thoughts first written in the 19th century, and would be censured if I did not.

98
by ABW (not verified) :: Fri, 05/18/2007 - 9:14pm

Scot, you don't seem to have a good grasp of what plagiarism is.

If Howard Bryant writes an article about how poorly the Redskins are managed, he doesn’t have any proprietary or creative interest in the order in which his examples are listed. Even if you blatantly and intentionally listed the examples in the same order, it’s nothing Bryant should get in a twist over

Yes, he does have a proprietary and creative interest in the order in which his examples are listed, and he should get in a twist over it if someone blatantly and intentionally rips them off. This is the very definition of plagiarism. Listing it as an example of something that is not plagiarism is just wrong. Not a difference of opinion, wrong.

I've linked to an article that was on Slate in 1999 about plagiarism that lists a number of examples of Monica Crowley plagiarizing another article in the Wall Street Journal. The WSJ published an apology for this. I think if you look at this confirmed example of plagiarism, and compare it to the example above, you'll see that there is simply no way this is not plagiarism.

99
by Pat (not verified) :: Fri, 05/18/2007 - 9:24pm

I just don’t think Lewins research is that so “original� that it deserves a high degree of protection. It probably took him about five minutes to do that research.

"High degree of protection?" The fact that Brandt used the same list of QBs, and had the same number of games started, which, given the mistake with McNabb, is a signature of the ESPN Draft article), is basic plagiarism.

It would've been fine if he had redone the research on his own. But he clearly didn't. The fact that it's "simple" research (in your mind) doesn't change the fact that Brandt didn't do it.

Look, everyone who's ever done a study like Lewin's knows that you make little mistakes, here and there. It happens. One of the nice things, though, is that it ends up being a signature. When you see that signature elsewhere, it's a clear sign they just flat copied.

I don't think David, or the guys at FO, would care if Gil had seen the article, said "hey, that's interesting," done a brief bit of research himself, and then written an article, but he didn't. Sports journalists are getting worse and worse about not doing any research/reporting on their own. When crap like this happens, they have to be called on it.

It's their job to investigate the validity of sources and to be able to provide proof for their statements.

100
by Pat (not verified) :: Fri, 05/18/2007 - 9:34pm

I hear what your saying, but it all depends on context. If Howard Bryant writes an article about how poorly the Redskins are managed, he doesn’t have any proprietary or creative interest in the order in which his examples are listed.

You're holding plagiarism up to the standards of copyright infringement.

From Wikipedia (and a bazillion other places):

Plagiarism is different from copyright infringement. While both terms may apply to a particular act, they emphasize different aspects of the transgression. Copyright infringement is a violation of the rights of the copyright holder, which involves the loss of income and artistic control of the material when it is used without the copyright holder's consent. On the other hand, plagiarism is concerned with the unearned increment to the plagiarizing author's reputation.

Just because something's uncopyrightable doesn't mean you can't plagiarize it.

101
by Pacifist Viking (not verified) :: Fri, 05/18/2007 - 9:40pm

41. Exactly. I hate when columnists write columns without any research, reporting, inside information, or special insight. If I see a writer (like Brandt) citing somebody else's research, I know he's done some work and is trying to present us with something useful. As I tell my students, it's a good thing to use borrowed information, as long as you cite it properly.

Scott K: Brandt used the same examples, in the same order. If a student turned in a paper doing this, I would be hard-pressed to be convinced that is coincidence. The ideas may be out there, but they are not usually presented in such a similar and exact order. You really don't have a reasonable explanation for the specific examples and order used. A teacher might accept that a student comes up with the same conclusions as somebody else; however, if those conclusions are presented in similar ways, the teacher will (rightly) suspect plagiarism and confront the student on it.

64: Probably not expelled, but certainly punished.

102
by Will Allen (not verified) :: Fri, 05/18/2007 - 9:45pm

Yes, scot, the correlation between starts, completion percentage, exclusively, without any other variables worth factoring, and eventual NFL performance, was so obvious that Brandt was the only guy to write of it, except for Lewin. Or something.

103
by Andy (not verified) :: Fri, 05/18/2007 - 9:55pm

I bet Brandt reclines his seat on an airplane. Jerk!

104
by Will Allen (not verified) :: Fri, 05/18/2007 - 10:02pm

I wouldn't be surprised, Andy. I have more tolerance for inconsiderate twits than I do thieves, however.

105
by jimmo (not verified) :: Fri, 05/18/2007 - 10:13pm

I can say from experience that Lewin's research did not take "five minutes." (credit Scot K.) It takes about that long (ok, maybe not) to even find a single decent NCAA statistics website, decent being one that lists games started. If that stat alone was such a "no Sh*t factor" (credit Scot K.), every NCAA stat site would show it. They don't.

106
by jimmo (not verified) :: Fri, 05/18/2007 - 10:23pm

Linked in my name, a bit down the page, is an excellent way Brandt should have done things...

107
by brandtjoe (not verified) :: Fri, 05/18/2007 - 10:30pm

GilBrandt most original of all NFL writers. Gil speaks the Truth and doesn't need to kiss butt of FO. Brandt on his way to Phoenix in 08. Bet on it!

108
by jimmo (not verified) :: Fri, 05/18/2007 - 10:36pm

sorry for the excessive posting, but i just can't let this go.
Linked in my name is a re-posting of Gil Brandt's discussion of the 2003 QBs. Not a single mention of starts and/or % being a deciding, "no-sh*t factor" (credit Scot K.). Brandt's top quality to look for? "the most important quality is to pass exceptionally well."
Damn, a genius in our midst.

109
by Will Allen (not verified) :: Fri, 05/18/2007 - 10:47pm

Re 104; that's pretty funny....

110
by Sergio (not verified) :: Fri, 05/18/2007 - 10:49pm

The most definitive conclusion I draw about Scot K's argument comes from this bit on post 90:

"[...]But as a Non-Blogger, Non-Journalist, I just don’t care."

That's exactly the point here, I believe. It's not that you don't believe this is plagiarism; it's that you simply don't care if it is. Which is fine; it's your prerrogative. I find it confusing, however, that you try and convince us that this *isn't* plagiarism, when it's obvious Brandt took (at the very least) the paragraph mentioned in this page pretty much verbatim. Or can you seriously make a case he didn't? I would dare to say, even, that you simply can't discuss whether this is plagiarism or not if you can't grasp that basic concept. If you can't accept that Brandt did that (which I, based on your posts, believe you don't) then the whole discussion is moot.

Just curious, in what line of work are you?

111
by Pat (not verified) :: Fri, 05/18/2007 - 10:57pm

Linked in my name, a bit down the page, is an excellent way Brandt should have done things…

That link also is an example of why what Brandt did is wrong. The linked poster credited Brandt for the research.

112
by admin :: Fri, 05/18/2007 - 10:59pm

A brief apology: My last name was tagged as a moderation word because we had a problem recently with people posting using my name. I just un-moderated those comments, which explains why references to past posts are now one or two posts off. Back to the book...

113
by jimmo (not verified) :: Fri, 05/18/2007 - 11:06pm

re: Pat 110, to be fair, he attributed Lewin extensively also. Try googling (sorry, that's probably copyright infringement) ... try searching some combination of "quarterback projection", Gil Brandt, College starts.... 90% of the results are crediting Brandt. That's terrible, and further proof that FO has a at least a case for "moral indignation."

For added emphasis, my name is linked to a copy of Brandt's 2002 QB preview, another year with a top-of-the-draft debate, and another year where starts or % failed to be featured as importants numbers by Brandt.

114
by jimmo (not verified) :: Fri, 05/18/2007 - 11:08pm

yeah, *important above, minus the "s". Too many posts

115
by Glenn (not verified) :: Fri, 05/18/2007 - 11:18pm

106: Ha!
On a side note, I've always thought that Brandt's fantasy analysis on NFL.com was totally worthless, written by a guy who had no clue how to gauge fantasy players or give insightful advice beyond "Start Tomlinson this week". It looked to me like the website couldn't figure out what to do with him, so they gave him the fantasy column. Fortunately, they now have some more insightful analysis from Fabiano, but for awhile, the offical NFL website had dreadful Brandt fantasy advice.

116
by Glenn (not verified) :: Fri, 05/18/2007 - 11:31pm

In more plagiarist news, RON BORGES HAS JUST RETIRED FROM THE GLOBE.

Link here.

117
by Richie (not verified) :: Fri, 05/18/2007 - 11:32pm

When I saw TMQ bring it up (Brandt’s “innovative� system), I was highly suspicious. Now I’m even moreso.

Does this mean that TMQ did not even read the book that he is quoted as endorsing on the back cover?

118
by Pat (not verified) :: Fri, 05/18/2007 - 11:44pm

#112: Yes, he did, but if you read it, the author actually is implying that David took Gil's idea ("40+ starts") and went further. Which is just a load of crap.

119
by Dr. Evil (not verified) :: Fri, 05/18/2007 - 11:51pm

All he had to do is cite Lewin properly in the text? Didn't any of these so called "experts" go to college?

120
by jimmo (not verified) :: Sat, 05/19/2007 - 12:01am

yep, right again Pat, #117. I read a little fast the first time. Subsequently, I fired off an email to Terry Pluto (author of the linked piece), complete with FO links, asking him to call out Brandt and properly credit FO/David Lewin.
I encourage all to also use the link Richie posted a couple of times, most recently #93, to send something to NFL.com.

121
by CA (not verified) :: Sat, 05/19/2007 - 12:02am

In #45, I suggested one possibility on how this apparent plagiarism may have occurred somewhat unintentionally. Let me suggest another: Let's for a moment take Brandt at his word in his apology to Rick Gosselin in the link in my name. Perhaps, similarly, Brandt mistook Lewin's work and Aaron's writings for his own. He might have taken virtually word for word notes on Aaron's ESPN NFL Draft 2007 magazine piece but failed to cite the source in his notes. Later, when preparing his own article, he found the notes and thought that they reflected his own work. If, through speaking with his contacts or independently, he had developed similar, if less specific, ideas already, maybe he could have made such a mistake. If something along those lines happened, then this could be a relatively innocent mistake by an old man whose work has gotten sloppy. I'm just trying my best to give the guy the benefit of the doubt here. Perhaps I'm trying too hard.

122
by CA (not verified) :: Sat, 05/19/2007 - 12:11am

Following up on #120 above, I know that Brandt denies having seen the Outsiders' work on this topic (which, as I wrote before, if true is a shame in its own right), but I am suggesting that he may be dense enough that he honestly doesn't realize that he has. Or, as I speculated earlier, somebody else pulled David and Aaron's info for him and he was unaware of the source.

123
by asg (not verified) :: Sat, 05/19/2007 - 12:13am

>I just don’t think Lewins research is that so “original� that it deserves a high degree of protection. It probably took him about five minutes to do that research.

Really? Have you read the original article in PFP 2006? You think that could have been done in five minutes? Having a clue in terms of what you're talking about isn't really that high on your list of priorities, is it?

124
by CA (not verified) :: Sat, 05/19/2007 - 12:25am

One more thing, and sorry for the triple post:

Considering that his work is important enough to have been appropriated by the NFL's official website, by the power vested in me by nobody in particular, I hereby nominate David Lewin for full official Football Outsider status, complete with a Jason Beattie caricature and brown font color in the comments sections. He's earned it!

125
by Pat (not verified) :: Sat, 05/19/2007 - 12:27am

#120: CA: Honestly, if that's true - and it could be - Brandt should either retire or hire an intern. That might sound harsh, but if he's at a stage where he can't properly attribute sources, he's no longer fit to be a journalist. It's a basic fundamental point of journalistic ethics.

The part I really don't like about all of this is Brandt denying he saw the work elsewhere. He clearly did.

126
by Skeptic (not verified) :: Sat, 05/19/2007 - 12:29am

It would probably help to send anyone who you run across that quotes Brandt a polite note explaining what happend.

127
by jimmo (not verified) :: Sat, 05/19/2007 - 12:46am

Oh Skeptic, #125, electronically I now belong to a Xavier University BB, posting in a thread titled "The Mormons" (it was in response, believe me), some sort of International political Forum called "Its all Politics" which I'm afraid has now put me under government surveillance, various team sites, DraftSharks, etc... trying to spread the word.

And you know what? Its still taken me less time than it took to find out how many starts this year's draft class QBs had, prior to Lewin's 07 update here.

128
by Pacifist Viking (not verified) :: Sat, 05/19/2007 - 2:09am

I think the projection system is pretty original, but lay that point aside. Even if the data is available for anybody to analyze, if you use somebody else's interpretation of that data, you still must cite it (it has clearly not reached the point of common knowledge). If you do come to that conclusion on your own, then obviously you don't have to cite it (though if you don't, you are showing a lack of awareness of other commentary on the subject). However, the presentation clearly signals use of borrowed data.

Compare it to a poem. The poem is there, the words are unchanging, and thus anybody can make any interpretation of the text. However, if you see somebody else making a particular interpretation, you don't get to make that interpretation and claim it as your own. You can agree with the interpretation, but you must recognize the person who came up with it.

The issue isn't whether Lewin's system is original (which it is, and wonderful), or whether somebody else could come up with the same interpretation of the data on his or her own without consulting Lewin's system (which is possible). The issue is: was the writer aware of Lewin's interpretation when presenting it as his interpretation? The way the material is presented makes it quite clear that Brandt (or whatever ghost writer does his columns) was aware of the other material. The names and order are so exact it is hard to believe it could just be an inadvertent oversight (Brandt read the material but forgot that he read it somewhere else).

By the way, Brandt's presentation is not as well written, either; the original passage is clear and easy to read, while Brandt's choppy and rough. It's something I often see when an inferior writer is rewording the ideas of a superior writer.

129
by QB projection systemjoe (not verified) :: Sat, 05/19/2007 - 2:53am

#106: "GilBrandt most original of all NFL writers. Gil speaks the Truth and doesn’t need to kiss butt of FO. Brandt on his way to Phoenix in 08. Bet on it!"

Brandt plaigarist hack. Will go 0-16 next season, need FO's projection system to decide on QB to draft #1 overall. Lewin and Schatz will crush Brandt, win division. FO going to Phoenix in '08. Buy tickets now!

130
by coyotl666 (not verified) :: Sat, 05/19/2007 - 4:31am

i see two conflicting things going on here. first it does seem obvious that branch or one of his minions lifted research and analysis without attribution. in academia, and apparently now online, he deserves to be put in the docks and pelted with rotting organic matter. but not for using the others' work, rather for not attributing it to them. he does have a fair use right to replicate the actual data and even analysis. knowledge should be free, and intellectual property law, just as general property law is, shall always be an artificial power construct which has a simple definining term: theft. secondly, and largely overlooked in the maelstrom of outrage: it's highly liely that, given the ridiculously small sample size, that both the outsiders and brandts conclusions are, to put it delicately, erroneous. if i remember correctly, some serious fudging had to happen to bring vince young into the fold. and what about a quarterback like charlie ward, who i believe started many games in florida only to end up frustrating knicks fans rather than jets ones? personally, while i do think brandt was wrong, i think the greater wrong is pretending there's any validity to this pseudo-scientific predictive metric.

131
by Tom (not verified) :: Sat, 05/19/2007 - 4:55am

Re 112:

That would be trademark violation, not copyright. However, Google was denied that trademark because it has become so common.

132
by Seth (not verified) :: Sat, 05/19/2007 - 5:36am

129: The efficacy of the QB projection system has not been much discussed in this thread because it is largely irrelevant to the discussion. If I publish an article showing why I believe that the two best predictors of success for first-round QBs are shoe size and mother's maiden name, and then Gil Brandt comes along and says the same thing in the same way, sans citation... That's still plagiarism. And since we're talking about journalism here, there are few greater wrongs than that.

133
by James G (not verified) :: Sat, 05/19/2007 - 7:09am

Scot K - It's plagiarism, no doubt about it. And FO/ESPN have the right to be upset.

As for the obviousness, Lewin looked at many other factors (yards/attempt, INTS, etc.) and discovered that they did NOT correlate. Basically to discover that only 2 statistics mattered, while others didn't, was very novel. And completely non-obvious.

134
by Gerry (not verified) :: Sat, 05/19/2007 - 8:08am

"knowledge should be free, and intellectual property law, just as general property law is, shall always be an artificial power construct which has a simple definining term: theft."

Good lord, who let the commie in?

If knowledge were 'free', there would be less motive for people to produce it, resulting in there being less knowledge.

Same with goods. If goods are free, then one cannot profit from making them; if one cannot profit from making them, the incentive to make them goes down*. If the incentive goes down, there is less.

*Unless, of course, another incentive is added, such as the jack-boot of the state. There is a reason why communist states have been heavily authoritarian.

We now return everyone to their normal football arguments.

135
by jimmo (not verified) :: Sat, 05/19/2007 - 8:43am

re:130- noted, thanks.

I'm now going to use some kleenex to stop my bleeding while I look for a band-aid, or at least a q-tip. Scotch tape should come in handy as well.

136
by Karl Cuba (not verified) :: Sat, 05/19/2007 - 9:53am

Aren't the results of the projection system intuitive anyway? It basically says that quarterbacks should be accurate and that young Qbs improve by playing. Neither point is ground breaking. Just look at the number of quarterbacks that became decent NFL starters after getting playing time in NFL Europe: Kitna, Warner, Johnson, Delhomme etc.

137
by Zac (not verified) :: Sat, 05/19/2007 - 10:34am

coyotl666, you obviously haven't read the original article. If you had, you'd know that David Lewin said that only two college stats correlated with NFL success, among first and second round draft picks. David said (correctly) that it's hard to tell how good a backup is, because many of them rarely get the chance. Charlie Ward wasn't a 1st or 2nd round pick, so he is irrelevant to the discussion.

And I believe David had only two statistical adjustments: one for quarterbacks playing on an expansion team (which only counts for David Carr and Tim Couch) and an adjustment for QBs who played a year or more in I-AA (Culpepper, Tarvaris Jackson, Giovanni Carmazzi).

But David was also smart enough to know that the statistics are just a tool. Completion percentage can be artifically inflated by some of the schemes teams run. David's method was to compare a QB's completion percentage with the completion percentage of other QBs at that school. Brett Favre played at a bad school in Southern Mississippi, so his completion percentage looks low, but it was still higher than the average completion percentage at Southern Mississippi since he left.

All in all it's a much deeper system than you give it credit for.

138
by Pat (not verified) :: Sat, 05/19/2007 - 11:11am

if i remember correctly, some serious fudging had to happen to bring vince young into the fold.

You don't remember correctly. Young was projected as "not a good passer." He wasn't one.

139
by Karl Cuba (not verified) :: Sat, 05/19/2007 - 11:22am

137: Give the kid a chance. He plays one year with bobbins receivers and you're already binning him, that's a little harsh.

140
by Pacifist Viking (not verified) :: Sat, 05/19/2007 - 11:59am

133: He's right: the knowledge is free. Most ideas fall under "fair use," and to advance in any field, ideas have to be available for use. The error is not in using borrowed material, but in failing to identify the borrowed material as borrowed material. It is absolutely free to use somebody else's ideas or conclusions; if we were required to "pay" for those ideas and conclusions, all fields of study would become stagnant. In any field, researchers build off each other's work. We can use each others ideas--that's good for intellectual growth and development in a particular field. But we have to credit the source.

141
by Fergasun (not verified) :: Sat, 05/19/2007 - 12:26pm

Well if Brandt wants to do this... we'll just have to start referring to the draft value chart as the David Lewin Draft Value Chart.

142
by Adam, VA (not verified) :: Sat, 05/19/2007 - 1:05pm

Anyone notice this little Brandt retraction I've linked to? Whoops!

143
by BillWallace (not verified) :: Sat, 05/19/2007 - 1:33pm

I have no comment on this issue, but I wanted to note that we're not currently making fun of the Bengals. I thought that was odd.

144
by bubba (not verified) :: Sat, 05/19/2007 - 1:50pm

#136 - I wonder if Brandt was such a hack that when he blatantly plagiarizing DL's system, that he didn't realize that completion % is a relative statistic? That he started going down the route of researching completion % and couldn't explain how Texas Tech. QB (Kingsbury et. al) weren't #1 draft picks. So he just dropped completion %, but inadvertently left in a reference without explaining the concept. This guy must have been under a huge deadline crunch if he was this sloppy (and that's all of his making).

145
by throughthelookingglass (not verified) :: Sat, 05/19/2007 - 1:50pm

not really much to add, I just appreciate the way FO has dealt with this.

146
by Fergasun (not verified) :: Sat, 05/19/2007 - 1:53pm

So David is neither a good friend nor a respected writer amongst this circle jerk of petty jerks... guess he doesn't deserve a retraction.

147
by rashreflection (not verified) :: Sat, 05/19/2007 - 3:35pm

You know, I agree that Brandt majorly f'ed up here, and am impressed with FO's class in handling this.

However, I'm still not sold on this projection system. In a matter this large, I will never in a million years be convinced that there are only two meaningful variables. Never. Most likely, we're just not examining other factors from the right angle.

I would say this is sort of like how sabermetrics always used to say fielding doesn't matter - turns out, they just didn't look at it in the right ways. Over the years we have learned all sorts of odd little facts in this area, such as removing strikeouts from catchers' fielding percentages, and now it is accepted that fielding is approximately 25% of the game. I think we still have much of this type of knowledge to figure out with college performance - especially with specific systems and how they relate to NFL success.

I can definitely understand why you guys are pimping this like it's the next DVOA - you got lucky with Rivers and look like geniuses now. But I think it's a good start, not a complete system, and you are going to get hoodwinked by some system QB who gets first-round hype; hell, if you came up with this a decade ago, it would've already happened with Tim Couch. But as of now, my money's on Colt Brennan...

148
by Israel (not verified) :: Sat, 05/19/2007 - 3:52pm

The McNabb anomaly is the killer. You should make sure you have one of those in all such analyses, just to keep the idea thieves honest. (Or at least make them work a bit.)

149
by Ilanin (not verified) :: Sat, 05/19/2007 - 3:54pm

There are *three* variables to the system, and one of them is a vast meshing of all the other unexamined ones together - draft status. There's a reason the system doesn't work for later round QBs, and it's because at a college level a QB can have a significantly flawed game and still produce high GS/comp % if he's in the right system or similar. The projection system relies on NFL scouts to weed those guys out, and so doesn't assign first- or second-round grades to them.

150
by asg (not verified) :: Sat, 05/19/2007 - 4:06pm

However, I’m still not sold on this projection system. In a matter this large, I will never in a million years be convinced that there are only two meaningful variables. Never.

In addition to what #148 said, in the original article, Lewin says, for QBs taken in rounds 1 and 2 of the last 10 drafts, that college starts + completion percentage explain about 65% of all variation in pro performance. So, Lewin didn't say that there were only 2 meaningful variables -- just that there were 2 variables that do the heavy lifting, plus a whole slew of others that, taken together, explain the other 35%. But these other variables individually matter much less than the big 2.

151
by Pacifist Viking (not verified) :: Sat, 05/19/2007 - 4:08pm

An interesting bit from U.S. copyright law:

"In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work."

That seems to fit perfectly here. Indeed, ideas are free, and this provision specifically allows advancement in fields of research. But in an ethical sense, you get to use this material free of charge, but you are supposed to cite the author. Failure to do so comes with consequences in academia and journalism.

(follow my name for link).

152
by Alex (not verified) :: Sat, 05/19/2007 - 4:08pm

#129: "secondly, and largely overlooked in the maelstrom of outrage: it’s highly liely that, given the ridiculously small sample size, that both the outsiders and brandts conclusions are, to put it delicately, erroneous."

That's putting it delicately?! I'm guessing you're not familiar with the concept of putting something delicately, then.

"if i remember correctly, some serious fudging had to happen to bring vince young into the fold."

No, as mentioned previously, he wasn't projected to be great passing, and he wasn't great at passing. He was just so good at running that it didn't matter. The QB projection system wasn't designed to incorporate the running ability of scrambling QBs, IIRC. So maybe it should've been called the "passer projection system". :)

"and what about a quarterback like charlie ward, who i believe started many games in florida only to end up frustrating knicks fans rather than jets ones?"

Charlie Ward started 25 games, so even if the system applied to him, it wouldn't say good things about his potential as a pro passer. But, since he wasn't drafted in the first two rounds, it doesn't apply to him at all.

If you don't understand the way the system works, don't be so quick to criticize it.

#135: "Aren’t the results of the projection system intuitive anyway? It basically says that quarterbacks should be accurate and that young Qbs improve by playing."

Careful, there, it doesn't really say that young QBs improve by playing more games, it says that QBs who start more games while in college are more likely to become good pro passers, all else being equal. In other words, it says that a guy who wins (and keeps) the starting QB job earlier in his college career is more likely to be the better player. The projection for a QB wouldn't become better if he just spent some time playing in Europe, because it wouldn't change the fact that he wasn't able to win the starting job early in his college career.

153
by Pat (not verified) :: Sat, 05/19/2007 - 5:05pm

#150: Plagiarism doesn't always fall under copyright infringement. It frequently can fall under defamation of character as well. But, to be honest, if this would ever get to a legal level, the NFL would be stupider than I could ever imagine.

#138: I said Young was bad, not that he will always be bad. I don't have to give someone a break to state a fact. :)

154
by Pacifist Viking (not verified) :: Sat, 05/19/2007 - 5:26pm

152: I never said it wasn't plagiarism just because it's not copyright infringement; I posted that excerpt in response to #133's objection that knowledge isn't free. "Any idea, procedure, process, system, method of operation, concept, principle, or discovery" can be borrowed, but if one fails to identify the source of the borrowed material, it is plagiarism.

155
by jimmo (not verified) :: Sat, 05/19/2007 - 5:51pm

PFT is on the case, thanks MDS

156
by Will Allen (not verified) :: Sat, 05/19/2007 - 6:15pm

What bugs me about this episode is that Lewin, I presume, is really just beginning to attempt to make a living analyzing football player performance, whereas Brandt has had a long career, received a lot of recognition, and has made a pretty good living already. For Brandt to use techniques and ideas that Lewin developed, without attribution, really reveals a sense of entitlement, and actively harms Lewin's career, if it goes unnoticed. It is really awful behavior. Luckily, it's looks like this episode is going to get attention, and thus Lewin might actually benefit in the end from Brandt's unethical behavior.

157
by Pacifist Viking (not verified) :: Sat, 05/19/2007 - 6:29pm

155: And if Brandt had done all that he's supposed to do and simply identified the source of the material, he might have helped an up-and-coming writer (though based on the passages presented above, he'd be identifying Schatz, not Lewin). It's a good thing to use the thinking of other researchers and to promote it; it's a bad thing to pass it off as your own.

158
by kleph (not verified) :: Sat, 05/19/2007 - 7:26pm

having been a FO regular for several years now i have had the pleasure of watching their hard work slowly ease out of the fringe and into the mainstream. their approach to football analysis is clearly on the right track and the fact more and more of the conventional outlets for information on the sport are picking up on the work they have started here is a testament to that.

it is almost inevitable that something like this would occur as the process from outsider to insider transpired and i regret that it has finally come to pass. it is my hope that the situation can be resolved with mr. lewin and the site getting the credit they deserve - but i'm not holding my breath.

yet, in another way, the outcome of this particular is somewhat irrelevant in the long run. the value of the work done on this site is becoming obvious to the point of absurdity. while one lazy hack or another will swipe it, nobody can replicate what this site produces consistantly year in and year out. eventually, that will become recognized by everyone but the most adamant die-hards.

i'll admit. the math and often the analysis i find on this site often goes over my head. i love reading it anyway because i've discovered trying to understand it has given me a much greater appreciation of the game.

i have each and every copy of the book because, in ten years when everyone recognizes what has been done here, i'll have the whole set on my bookshelf already.

159
by Richie (not verified) :: Sat, 05/19/2007 - 8:02pm

Well, here's the response I got from CBS Sportsline about my complaint:

"Response (Tom C.) - 05/19/2007 06:59 PM
Gil Brandt is employed by NFL.com, not sportsline.com. We run the NFL.com fantasy leagues but not the fantasy writers and have no bearing on the editorial content."

160
by MarkB (not verified) :: Sat, 05/19/2007 - 8:45pm

Ron Borges has resigned from the Boston Globe, so there's a job opening for Gil Brandt

161
by Ilanin (not verified) :: Sat, 05/19/2007 - 10:42pm

On an entirely irrelevant aside, I fished out my PFP 2006 to see what precisely it was Lewin said about Vince Young. Turns out he pretty much nailed it:

"Young's projection is lower than Cutler and Leinart's but that does not mean the system thinks he will be the worst pro of the three. If Young meets this passing projection, he will be the best quarterback of this year's crop because he can combine that with his running skills."

Something else he nailed (well, at least halfway - "vastly worse" we'll have to wait on), which amused me as I encountered it on the way through, was in the paragraph on Byron Leftwich:

"In the next couple of years expect scouts to fall in love with the similarly talented, but vastly worse, JaMarcus Russell of LSU."

162
by skippyx (not verified) :: Sat, 05/19/2007 - 11:20pm

The Mcnabb starts is the key. It sounds like something the CIA would use to track a leak in a spy novel. If he did not steal the research then it was given to him by someone who did.

End of story.

163
by pbmax (not verified) :: Sat, 05/19/2007 - 11:25pm

The solution: Give Gil Brandt 390 carries, er, columns this year, and he'll be toast after that.

164
by Lee Davis (not verified) :: Sun, 05/20/2007 - 3:35am

I wrote to NFL.com and as I did it occurred to me that an additional component of this offense is the effort the commissioner is putting on the personal conduct policy for players and employees in the NFL. This new emphasis should also be applied to Mr. Brandt (whom I have read for years with respect). At the very least the commissioner needs to require an apology from Mr. Brandt similar to the one to Mr. Gossage. Perhaps additional emails to NFL.com can cite this hypocritical stand of silence about the plagerism on NFL.com from the NFL.

#84 - actually ProFootballTalk has an article and a link to this article. Mr. Florio is indeed a supporter of this outrage and has brought me here (although I read Football Outsiders nearly every week I read ProFootballTalk every day).

165
by Sid (not verified) :: Sun, 05/20/2007 - 3:53am

RE: 17

On a related note, Ron Borges is now working for the Globe again.

and now he retired.

166
by sam (not verified) :: Sun, 05/20/2007 - 10:20am

163:

Not to be nit-picky, but when Randy S. (#84) posted that PFT wasn't "all over this" Florio had yet to acknowledge this on the Rumor Mill. I've been checking this thread and PFT a few times a day :)

167
by eagletom (not verified) :: Sun, 05/20/2007 - 5:33pm

To be honest, I find myself wondering if Brandt was too busy round the time of the draft and didn't hand some stuff in. Maybe NFL.com then got some staffer to cobble stuff together so they had something. Looks like a case of deadline-itis to me, and the kind of thing a first-year student does in desparation.

I still think that's a bit reprehensible, of course. My point to NFL.com when I sent a complaint in was 'what are you paying him for? Maybe you should check the rest of his work out.'

168
by jmparker (not verified) :: Sun, 05/20/2007 - 5:55pm

1.FO readers are outraged mostly out of loyalty. Dlew is a favorite around here and for good reason. He opened our eyes with his projection system and we are all very greatful for his contribution to this site.

2. None of want to see this guy's hard work go down the drain when Gil Brandt starts getting credit for Dlew's research.

3. Brandt may not be the party to blame. However I think we all will feel cheated if DLew is not rewarded with some type of compensation for this ridiculous mistake. It doesn't have to be monetary but if someone else starts getting paid because his research was not properly cited in the Brandt's article then we all are going to be some pretty pissed off outsiders. The Chuck Liddell kind of pissed off.

4. It is making me angry that some posters who do not even understand the system have found their way to come over here and bash it. Some guy had the brilliant notion that a system qb who starts for 4 years and has a high completion pct might come along and destroy the projection system. Well "No shyte"(citation scot k). That why Lewin has created the 1st or 2nd round component of the system!!!

I've come up with stuff like this for basketball/wagering etc and shared it with my friends but I would never be ballsy enough to come out and print it for all the greedy bastards out there to steal my idea and profit from it. Its my idea and I would like to be the one in control of that idea thank you very much.

In this sense I may be projecting a bit but I feel badly for Lewin. If something like this happened to me I would be uncontrollably p.o'd.

I hope some sort of solution to this matter is agreed upon quickly. If I was David I would be losing my temper.

169
by Steven (not verified) :: Sun, 05/20/2007 - 7:58pm

How do we know that footballoutsiders.com didn't steal the information from Gil Brandt's files?

170
by Theo (not verified) :: Sun, 05/20/2007 - 8:06pm

150 Pacifist Viking:
It's the theft of an article. Plagiarism. Brandt re-wrote it... poorly.
Oh and he IS the one to blame, he put his name under it, so he is responsible.

171
by asg (not verified) :: Sun, 05/20/2007 - 8:39pm

168: Because it's usually the wronged party that brings charges of plagiarism, since plagiarism thrives on inattention. If FO is the copying party here, surely Brandt would be the first to complain. Moreover, if Brandt did this research first and it's in his files somewhere, it would be straightforward for him to produce a file with the allegedly copied passage in it (and, of course, to show how FO could have had access to it).

172
by Alex (not verified) :: Sun, 05/20/2007 - 8:53pm

#168: "How do we know that footballoutsiders.com didn’t steal the information from Gil Brandt’s files?"

You're joking, right? They published their article about a year before Brandt, and their explanation is actually a complete one, instead of leaving out the part about completion percentage being important. The burden of proof is on Brandt, because he published his article after FO.

173
by Starshatterer (not verified) :: Sun, 05/20/2007 - 9:01pm

Steven (#168 )--

First to publish gets the call, unless there's actual proof of theft.

174
by Pacifist Viking (not verified) :: Sun, 05/20/2007 - 11:06pm

169: Of course it's plagiarism--I was just noting that "any idea, procedure, process, system, method of operation, concept, principle, or discovery" can be borrowed, AS LONG AS THE SOURCE IS CITED. Failure to cite the source--especially when in this case so deliberately taking from the specific presentation--is what makes it theft or plagiarism.

175
by Pacifist Viking (not verified) :: Sun, 05/20/2007 - 11:13pm

169: I see what you are saying though: there is a difference between borrowing a system and borrowing from a specific article. In either case, though, the author just needs to cite the source.

176
by MJK (not verified) :: Mon, 05/21/2007 - 12:36am

Pacifist,

Not to nit-pick, but an idea, procedure, process, system, method of operation, concept, principle, or discover cannot be copyrighted, but it CAN be patented. Different protection methods for different things.

Not that that applies here--David Lewin's work probably qualifies as a "mathematical formula" or an "algorithm", which are currently not patentable (but give the software industry's lawers time...)

177
by Sergio (not verified) :: Mon, 05/21/2007 - 1:06am

re: 175

That'd be a sad day.

178
by Pacifist Viking (not verified) :: Mon, 05/21/2007 - 1:24am

175: Does a patent merely mean somebody else couldn't sell or profit from the system, or can't reference or use it? I would think even a patented system could still be discussed "for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research" (the "fair use" uses of copyright law).

I think you could also argue that Lewin's system is actually a "discovery"--the data is there for everybody to look at, and it is mainly the conclusion (that games started and completion percentage predict future performance) that Lewin deserves credit for.

(This is, of course, a side discussion separate from whether plagiarism occurred, which is pretty obvious).

179
by Ch V Kalyan (not verified) :: Mon, 05/21/2007 - 3:56am

let me be a devil's advocate here. the system for college QB success in NFL as defined by FO is very simple. You look at the no. of starts and percentage completion.

Someone whoz been in the NFL business could theoretically consider this as very rudimentary and view it as a "back of the hand" model. Esp. someone with the experience of Gil Brandt in NFL and scouting could have imbibed this over his vast experience

Just a thought ..

That said, i personally think his assistants lifted the data off FO ...

180
by Will Allen (not verified) :: Mon, 05/21/2007 - 4:37am

#178, yeah, except for the small matter that Brandt, when writing previously about qb evaluations, has never stated that these were the two key measures, and in fact stated that it was simply critical that the qb be an excellent passer, without any reference whatsoever to number of starts being important. He stole this from Lewin, plain and simple, or he had somebody steal it for him.

181
by Theo, Holland (not verified) :: Mon, 05/21/2007 - 4:48am

Re 136 Zac:

Would it be fair to say that JaMarcus Russell will be playing on an expansion caliber offense next year?

182
by Theo, Holland (not verified) :: Mon, 05/21/2007 - 4:49am

Re 136 Zac:
adjustments: one for quarterbacks playing on an expansion team

Would it be fair to say that JaMarcus Russell will be playing on an expansion caliber offense next year?

183
by Theo, Holland (not verified) :: Mon, 05/21/2007 - 4:54am

Re 136 - Zac:
...edit...
adjustments: one for quarterbacks playing on an expansion team

Would it be fair to say that JaMarcus Russell will be playing on an expansion caliber offense next year?

184
by Gerry (not verified) :: Mon, 05/21/2007 - 7:53am

"He’s right: the knowledge is free."

You're wrong; he didn't claim the knowledge is free, he claimed all knowledge should be free, and further that private property rights equate to theft.

The latter just highlighted the intellectual vacuum from which the former comes.

If there is no personal profit to be made from the development of knowledge (or even if personal profit is minimized), then there will be less development of knowledge. It is that simple.

185
by Gerry (not verified) :: Mon, 05/21/2007 - 8:10am

"I posted that excerpt in response to #133’s objection that knowledge isn’t free."

While noting you have misstated my position, let me also point out your fallacy. The original poster said that all knowledge (and everything else, for that matter) should be free.

I said bullocks.

You responded saying knowledge is free, with this from copyright law:

“In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work.�

The logical fallacy here is that saying that some knowledge is 'free' according to US copyright law is not the same as saying knowledge is free, period. Copyright law is not the only show in town; there are other branches of law, and further not everything that is legal is completely unfettered. The whims of tradition and culture apply costs. A perfect example of this is how society treats plagiarism.

So while I was mainly objecting to that guy's comment about the role of property rights (physical and intellectual), I think you took my reply as applying directly to this situation. Although my post did not say so, I believe that copyright law does not back up Brandt here. So let's go to the US Copyright Office website, right below where they talk about section 102.

"What Is Protected By Copyright

Copyright protection extends to a description, explanation, or illustration of an idea or system, assuming that the requirements of the copyright law are met. Copyright in such a case protects the particular literary or pictorial expression chosen by the author."

Anyhow, peace.

186
by Mr Shush (not verified) :: Mon, 05/21/2007 - 8:58am

"the system for college QB success in NFL as defined by FO is very simple. You look at the no. of starts and percentage completion"

No, Kalyan, it's not. You don't just "look" at the starts and completion percentage: you use them as inputs for a complex mathematical formula which no-one except David Lewin (and conceivably Aaron and other Outsiders - that I don't know about) knows. The formula then outputs a number, which is the quarterback's projected per game DPAR for (if I remember correctly) his 9th season in the NFL, and then use a table which I believe was published in last year's PFP to arrive at the equivalent number for other seasons.

And as has been stated above, please show me any article prior to Lewin's original piece which claimed that college starts were an important indicator of NFL success and that, for example, TD/int ratio was irrelevant.

187
by brown fonts for Lewin (not verified) :: Mon, 05/21/2007 - 10:34am

RE 123:
....David Lewin for full official Football Outsider status, complete with a Jason Beattie caricature and brown font color....

Seconded! Brown fonts for Lewin party!

188
by Chris (not verified) :: Mon, 05/21/2007 - 10:50am

I think the System is more suited to decide Leaf or Manning, Mcnabb or Smith, Young or Leinart, Russell or Quinn. The system is not to rank every player and predict if Chris Leak or Jordan Palmer will be better. Also, don't be suprised if the system isn't right 100% of the time.

It could have been that GB was just using the same examples because the fact that the sample size isn't teribly large, but the Mcnabb failure is the straw that broke the camels back. The guy straight up copied the research, and then lied about it. Shame on you GB.

189
by Chris (not verified) :: Mon, 05/21/2007 - 10:52am

So what is worse, Peter King talking about how many starbucks he passed on his road trip, or GB copying other DLews work and lying about it?

190
by glabrous3 (not verified) :: Mon, 05/21/2007 - 11:02am

I don't know if any of you have checked, but as of Monday morning Brandt's April 26th article does not appear under "More by this writer" on his nfl.com page (http://www.nfl.com/writers/brandt), though the article can still be accessed through the link on this site. Also, his May 18th article is quite interesting...

191
by Tarrant (not verified) :: Mon, 05/21/2007 - 11:09am

While I've for the most part stayed out of this thread, I wanted to say it's quite disappointing that NFL.com hasn't been willing to own up to this - it's not like this is research that was posted on some small blog and then never seen again. The article was in PFP for cryin' out loud! There's obvious and uncontestable evidence of prior publication via the book.

That said, the thing that disappoints me more in this thread is various people that have come in and said things like "Oh come on, Lewin's stuff isn't that great, anyone could have done it", or the corollary, "If it was that good, someone else would have done it first."

As someone who works in R&D, very little goes right to stabbing your core being more than those two statements, or variants thereof. At least in the case of plagiarism (while terrible), one can take to heart the fact that others found the work to be worthwhile - worthwhile enough to want to take credit for it themselves!

But saying that "Anyone could have done it", or "If it was that useful, someone else would have done it already", that's cruel. Because "anyone" didn't do it, and someone else didn't do it already (that we know of).

Sure, anyone could have come up with this system, but DL is the one that did, to the best of anyone's knowledge. Sure, anyone could have invented the wheel, but someone had to be the one to do it. Anyone could have formulated the Theory of Relativity, but anyone didn't.

One can debate the legalities or moralities of this situation, or whether or not it was plagiarism or whatnot, and such debates are welcome to this threat, but I won't get into them here - I have my own opinions. But were I DL, I think I'd be more disheartened at the people that have come here and simply said "Who cares if GB plagiarized, because the work itself is pedantry."

192
by Yaguar (not verified) :: Mon, 05/21/2007 - 11:36am

182: Yea, I think Russell is basically playing for an expansion-quality offense.

I believe that somewhere the same theory was postulated about Alex Smith, whose terrible rookie year obviously underperformed his projection.

193
by krugerindustrialsmoothing (not verified) :: Mon, 05/21/2007 - 11:44am

189. wow, maybe tne NFL is in the process of clearing up the backlog of brandt apologies. We should see one for FO coming up in a few weeks.

194
by Chris (not verified) :: Mon, 05/21/2007 - 12:07pm

190- Your right. Anyone can play charlatain ( or Monday Morning quarterback) and say " oh it's so simple, it's the wheel!".

I'd also agree that by the nature of him plagurising the work, he's accepting it as worthwhile... Then again, we don't need him to verify that for us.

195
by Gril Bandt (not verified) :: Mon, 05/21/2007 - 12:13pm

Clearing up a mistake

On the day of April 26, a few days before the the 2007 NFL Draft, I wrote a quarterback review for the upcoming NFL Draft. In that column, I referred to the idea that 2 main statistics predict NFL succes of college quarterbacks. Regretfully, with all the paperwork I had at my disposal that night, I forgot to mention where I got the idea from. It was actually the research by Football Outsiders columnist David Lewin. David is a friend with whom I share information, which is why I happened to have his analysis among all my own information.

I apologize to David for the misunderstanding, and I apologize for any misinterpretation of the circumstances. David is a good friend, and he also is one of the most respected NFL writers in the country. It is a credit to David's football research that FootballOutsiders.com is such an informative and insightfull website, much better than all the cliché loaded crap we write here at NFL.com.

I hope this helps clear up any confusion regarding this incident. Now we can begin focusing on the 2007 season!

Gril Bandt

196
by Pacifist Viking (not verified) :: Mon, 05/21/2007 - 12:27pm

183: Researchers in any field build off fellow researchers' knowledge. If that knowledge is not available for use, research in a field will become stagnant. I'm disagreeing with your statement that "If knowledge were ‘free’, there would be less motive for people to produce it, resulting in there being less knowledge." In fact, if you require people to pay for any knowledge discovered, there would be less knowledge, because researchers wouldn't have such open access to each others' findings. I think it was Newton that said, "If I see further, it is because I stand on the shoulders of giants." If you take away the premise that knowledge--knowledge itself--is free, you will have less knowledge.

I don't think you're premise that if you take away financial incentive, there will be less knowledge, is correct (besides, there already is incentive in the forms of promotions, tenure, grants, job offers, etc.).

But I'd still say the knowledge itself is free, and might even go so far to say that all knowledge is free. Can you give examples in which it is not? I'm talking about the knowledge, the discovery, the finding itself--not a particular system of discovery of the knowledge, or the presentation of the discovery.

I do recognize, though, that we were originally arguing at different points. I think that there is a difference between property and sale of goods and property and sale of knowledge, but you were arguing at somebody who seemed to equate the two. Understood, and peace.

197
by Chris (not verified) :: Mon, 05/21/2007 - 12:43pm

How much do you think the Oakland A's pay to collect and interpret their statistics? For example, from what I remember, they track every single ball in play in baseball and compare the outcome to that same hit judged by speed, location that has taken place in X number of years.

It goes along with the routine that " the best players make it look easy". Their sabermetrics say that Adam Everit of the Astros is the best fielding short stop in the game. He might not have the best dives and highlights, but they say that he makes more plays and has a higher pct. than everybody else.

For example, what if everybody loves those diving catches from a Tori Hunter or Jim Edmonds, but Griffey Jr. was a better fielder because he made it look easier?

There is a cost to gathering information. That information is worth money too.

What if there was a ROBO GM that always made the right calls? The guy would draft the best player at every position ( and of need) at every draft.

198
by mmm... sacrilicious (not verified) :: Mon, 05/21/2007 - 12:46pm

#196: ROBO-GM would be fairly easy to program. Just feed in data on everything Matt Millen has done so far, and tell it to do the exact opposite.

199
by Pacifist Viking (not verified) :: Mon, 05/21/2007 - 1:04pm

196: That's true: if a party comes by some knowledge, that party can keep it secret and choose to sell it to another party. So there are cases where knowledge costs money.

It is once knowledge goes public in some way that it becomes free. If politician X writes a book about his/her experiences behind the scenes, a reporter could freely cover what politician X said. I could make copies of the book or a news story about the book and distribute it to my classes. An historian could check the book out of a library and freely cite it in his/her own book. Of course there's a limit: you can talk about the knowledge in the book, and you can quote excerpts from the book, but you couldn't republish the book and sell it without permission. It's the knowledge in the book that's now free.

So yes, knowledge can be a commodity. Once you go public with that knowledge, however, it becomes free.

200
by azibuck (not verified) :: Mon, 05/21/2007 - 1:12pm

I didn't (have time to) read this whole comment thread, but I actually wrote about completion percentage as an indicator of NFL success before David Lewin. It was on my blog, and I emailed Aaron thinking it was the greatest idea ever and asked if he wanted to use it. He said they had someone doing something similar, but that we had reached different conclusions regarding Jay Cutler. I still have Aaron's (dated) email. A few weeks later, Lewin's article appeared at FO.
I note this only to say that the IDEA could have been thought of independently. But clearly Brandt plagiarized Lewin's work.

201
by Pat (not verified) :: Mon, 05/21/2007 - 1:19pm

#196: You're mixing knowledge and information. Information isn't free. Gaining information takes energy. That's basic thermodynamics. Energy is basically money, so yeah, gaining information is absolutely not free.

Knowledge, however, is the ability to process information, and that is free. Sometimes it can take effort to figure out - years and years of research, etc. - sometimes it can come in a single stroke of inspiration. But fundamentally, knowledge is just knowledge - once it's there, it takes no effort to maintain - fundamentally, at least.

The restriction of the flow of knowledge is artificial - and in some sense, if the knowledge is used, impossible. To take the Oakland A's example, they can try to keep the way they evaluate players secret, but given enough time, the other teams could just reverse engineer it by looking at the players the As covet and the players they release.

The only reason that people pay to process information is because human beings aren't infinitely smart.

202
by jmparker (not verified) :: Mon, 05/21/2007 - 1:21pm

Free knowledge is one of the key components of life as we know it. Hence, Guttenberg is one of the most important inventors of all time. Stating that research would become stagnant if not for potential monetary gain is like saying potential monetary gain is the reason why Aaron started FO. Since we all visit the site regularly without paying we know that ascertion is false.Its not monetary gain that motivates people to share their research. It's finding a platform to share it with many people. I'm sure Aaron would not have continued with FO if it was only receiving 10 hits a day.

203
by Gerry (not verified) :: Mon, 05/21/2007 - 1:22pm

"In fact, if you require people to pay for any knowledge discovered"

Pacifist, because I disagree with the assertion that all knowledge must be free, it does not follow that I believe that people must pay for any knowledge discovered.

204
by Theo (not verified) :: Mon, 05/21/2007 - 1:35pm

You guys ever heard of Google or Wikipedia?
Knowlegde is free.

205
by Gerry (not verified) :: Mon, 05/21/2007 - 1:42pm

"Stating that research would become stagnant if not for potential monetary gain is like saying potential monetary gain is the reason why Aaron started FO. Since we all visit the site regularly without paying we know that ascertion is false.Its not monetary gain that motivates people to share their research. It’s finding a platform to share it with many people. I’m sure Aaron would not have continued with FO if it was only receiving 10 hits a day."

If you take away incentives, it might not stagnate but it will not be as much as it could be.

As for Aaron, if FO was not providing him with some profit, then he would not have the time to do the analysis he does. FO as it is would not exist.

206
by Pacifist Viking (not verified) :: Mon, 05/21/2007 - 1:47pm

Gerry,

Re: 202, Understood, my mistake.

Re: 204. Again, the knowledge can be free, but there are still incentives for the people who come by the knowledge (promotions, grants, book deals, job offers, prestige, etc.). In my field, researchers have very little chance to make a profit on their research, but they keep plugging out books and articles (partly because that's what people in my field love to do, partly because they benefit in their career by doing so).

I'm just saying that open access to others' knowledge leads to there being more knowledge. That is more important to bringing about more knowledge than providing financial incentive for the knowledge.

207
by Gerry (not verified) :: Mon, 05/21/2007 - 1:56pm

"I’m just saying that open access to others’ knowledge leads to there being more knowledge. That is more important to bringing about more knowledge than providing financial incentive for the knowledge."

I think there is an intellectual equivalent to the Laugher curve. If taxes were zero, there would be no tax revenue. If taxes were 100%, there would be no tax revenue because production would fall to zero. There is a sweet spot in between where tax revenues are maximized.

208
by Chris (not verified) :: Mon, 05/21/2007 - 3:39pm

Vikings fan, are you a teacher and researcher at a University?

209
by Mark (not verified) :: Mon, 05/21/2007 - 3:46pm

Who cares? Get over it already stat-geeks

210
by AlexDL (not verified) :: Mon, 05/21/2007 - 4:02pm

Thanks Mark, I never looked at it that way before. Your deep and insigtful comment made me feel better. Much better.

211
by jmm (not verified) :: Mon, 05/21/2007 - 4:10pm

As a father looking at paying for years 9 through 12 of his 3 childrens' college educations, I can assure everyone neither knowledge nor information are free.

212
by Yaguar (not verified) :: Mon, 05/21/2007 - 4:20pm

196: ROBO-GM would probably be the most valuable person ever for any franchise. I'd assume that after a while, ROBO-GM would always win the Superbowl. Even without using trades, he could put together an astonishingly good team.

2000: Marcus Washington, Laveranues Coles, Dante Hall, Adalius Thomas, Tom Brady, Mark Tauscher, Mike Green
2001: Chad Johnson, Adrian Wilson, Rudi Johnson, Russ Hochstein, TJ Houshmandzadeh, Rick DeMulling, Marlon McCree
2002: LeCharles Bentley, Deion Branch, Alex Brown, Aaron Kampman, Raheem Brock, Chester Taylor, David Givens
2003: Rashean Mathis, Lance Briggs, Asante Samuel, Mike Scifres, Cato June, Josh Brown, Antonio Gates
2004: Bob Sanders, Nick Hardwick, Nathan Vasher, Michael Turner, Jim Sorgi, Shane Olivea, Bobby McCray
2005: Lofa Tatupu, Frank Gore, Kerry Rhodes, Trent Cole, Bo Scaife, Matt Cassel, Nate Washington
2006: Marcus McNeill, Elvis Dumervil, Dawan Landry, Mark Anderson, Antoine Bethea, Marques Colston, Hank Baskett

His team would be totally stacked. Consider the receivers alone: Chad Johnson, TJ Houshmandzadeh, Laveranues Coles, Deion Branch, Marques Colston, Hank Baskett, and Dante Hall all on the same team with Tom Brady throwing to them. Sure, you might say, but he'll lose some of those guys in free agency because he won't be able to afford them all. Maybe so, but for every good player he loses in free agency, he gets a shiny new 3rd or 4th round compensatory pick. And because he's ROBO-GM, he turns those compensatory picks into guys like Robert Mathis or Marc Bulger.

213
by Pacifist Viking (not verified) :: Mon, 05/21/2007 - 4:42pm

207: yes.

214
by Duane (not verified) :: Mon, 05/21/2007 - 5:11pm

Just for the fun of it: Information without understanding is not knowledge.

215
by coyotl666 (not verified) :: Mon, 05/21/2007 - 7:29pm

bad writers borrow, good writers steal.

and, gerry, i'm no commie - it's the black star that refused marx that'll get you to the unhderstanding that increasing wealth through ownership is morally reprehensible. excuse me now, i've got to go pay the rent to this sorry bastard that can afford to eat tonight as he's making this world a better place collecting money for a roof he's never slept under.

216
by Theo (not verified) :: Mon, 05/21/2007 - 7:30pm

Just for the fun of it: Information without knowledge is data.

Although the 'data' page doesn't agree.

OT;
Aaron.
How's the lawsuit going? Isn't it disrupting the production of THE BOOK?
Just curious.

217
by Tally (not verified) :: Mon, 05/21/2007 - 7:37pm

214:

I believe the saying is: Stealing the work of one person is called plagiarism. Stealing the work of many is called research.

218
by coyotl666 (not verified) :: Mon, 05/21/2007 - 8:05pm

>>I believe the saying is: Stealing the work of one person is called plagiarism. Stealing the work of many is called research.>>

well, i dunno. are the outsiders many? you'd be a bad writer to borrow directly without attribution, but if you just did a snatch and grab of me thoughts and ideas, you'd be golden. and i'd even give you a blurb on the bsack of the book.

and, yes, indeed peace (a phrase and concept that certainly deserves patent protection if only for its novelty today)

219
by Oily Harry (not verified) :: Mon, 05/21/2007 - 9:02pm

What happened to my Raiderjoe Porky Pig template post of yesterday afternoon? When I posted it, I believe it was #167 or #168.
It contained zero explicit words.

Are all _____joe posts being deleted?

Either way, I'm done with the ____joes. I would just like an explanation, however. Thanks.

220
by Theo (not verified) :: Mon, 05/21/2007 - 9:55pm

People please!
Knowledge is free! Everyone knows there's 60 minutes in an hour. It's free to know.

At hand is plagiarism.
the practice of claiming, or implying, original authorship or incorporating material from someone else's written or creative work, in whole or in part, into ones own without adequate acknowledgment. (wikipedia)

Good lord. Crash course lawschool:
Was Lewin's research new? Yes. No proof it was not.

Could anyone (Brandt in this case) write an article, with the same oddities and same order of names and same strange mistakes? Yes. Most unlikely, but can you prove it was stolen? No. That's the thing. Any normal human being knows it. Gil Brandt is a douchebag. Period.
Compare this, it's true.
But laws aren't common sense.

221
by sn0mm1s (not verified) :: Mon, 05/21/2007 - 9:59pm

Not defending Brandt here... but really, it isn't like the idea is revolutionary. Completion percentage and game starts? Christ, I have been saying that for years. Brandt clearly plagiarized and FO should get the credit for the article.

For ground breaking research why don't you guys try to combine your OLine research with the RB dvoa and dpar?

222
by Alex (not verified) :: Mon, 05/21/2007 - 11:20pm

#220: "Not defending Brandt here… but really, it isn’t like the idea is revolutionary. Completion percentage and game starts? Christ, I have been saying that for years."

Yes, it is revolutionary. The part that's revolutionary isn't that completion percentage and game starts are important, it's that those are the only things that are significant in predicting pro performance, and that they are only accurate when applied to quarterbacks taken in the first two rounds. Also, the fact that his system provides a quantitative prediction for future performance tells us how important these two factors are, which nobody knew before Lewin did his research.

If you'd asked anyone about this subject before Lewin's research, they would've probably agreed that having more starts and a higher completion percentage was better, but they also probably would've thought that TD:INT ratio was important, or yards/attempt, or QB rating, or any of a number of other factors. And I doubt anyone would've known that those factors lose their predictive value when they are applied to quarterbacks taken in rounds 3 through 7. This was not common knowledge before Lewin did his research, and even if someone had reached these conclusions earlier, nobody proved that they were true before Lewin.

223
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Mon, 05/21/2007 - 11:29pm

sn0mm1s,

The groudbreaking part, as people have already mentioned, is not only that those two stats are very important for 1st and 2nd round picks, but that all other stats are irrelevant.

So if you hear someone talking about a 1st or 2nd round pick and they mention touchdown passes or wins or interceptions or yards per pass attempt or hight and weight or big-game performance, ignore them. Once you've been drafted in the first two rounds, all those stats don't predict anything. I'd say that's pretty freaking groundbreaking.

Not only that, but the formula is good for 74% of quarterback variation, meaning that about three quarters of the factors which determine success of failure of quarterbacks are accounted for by the system. That number is insanely high for anything related to NFL analysis, and for that reason also I'd say it's pretty groundbreaking. It doesn't just predict success and failure better than anything else, it does so to an unprecedented degree of accuracy.

Theo,

In this country the standard for a criminal conviction is "beyond a reasonable doubt." If the doubt about whether the man might have been blacked out was reasonable, then he'd be aquitted here. From your description, perhaps you imply that you think the doubt was unreasonable. In that case, he'd be convicted.

Gil Brandt is guilty beyond a reasonable doubt. It would take an unreasonable doubt to believe that the parallels were coincidental. (And being on medication is no excuse. He's a journalist! He shouldn't be writing under the influence if he can't keep his sources straight.) I realize that your post was in jest, but I couldn't help clarifying.

224
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Mon, 05/21/2007 - 11:31pm

Oops, Alex got there first.

225
by sn0mm1s (not verified) :: Tue, 05/22/2007 - 12:50am

The very fact that FO is qualifying around 1st and 2nd round picks makes it far less than ground breaking. The whole reason these guys even get drafted in the 1st and 2nd round is due to all the other factors that are stated to be irrelevant.

For instance, Ty Detmer was not drafted in the 1st or 2nd round despite having started 3 years with 60%+ completion percentage. Why? Because he was just 6 ft tall, weighed less than 200 lbs, wasn't very fast, had a fairly weak arm, played in the WAC etc. etc. THAT is why he fell out of the first two rounds.

The "revolutionary" system really says
IF the QB has an NFL body
IF the QB played against quality competition
IF the QB has a strong enough arm
IF the QB has the decent stats

Essentially, IF the QB has all the qualifications to be a 1st or 2nd round pick as evaluated by NFL GMs THEN the most important stats are games started and completion percentage. Those stats deemed "irrelevant" are NOT because if they were then Ty Detmer would have been drafted in the 1st or 2nd round based on starts and completion percentage.

I am pretty confident that I could say that in the last 20 years, RBs drafted in ANY round that are 6 ft or shorter perform better in the NFL than RBs 6' 1'' or taller drafted in the same round. That isn't ground breaking any more than saying if a QB has the overall talent to be drafted in 1st or 2nd round then predict their success on starts and completion percentage.

226
by David (not verified) :: Tue, 05/22/2007 - 1:13am

Except that there is very obviously a wide range of aptitudes among all the QBs you just described - some will settle into the pro game with ease, while others will never be able to adapt. If that weren't true, Ryan Leaf would still be an NFL player. If Lewin's system is nothing new then why were Leaf, Kyle Boller, Tim Couch, Joey Harrington et al drafted high in the first round, when it would have screamed "bust" for every one of them?

227
by Alex (not verified) :: Tue, 05/22/2007 - 1:46am

#225: "The very fact that FO is qualifying around 1st and 2nd round picks makes it far less than ground breaking. The whole reason these guys even get drafted in the 1st and 2nd round is due to all the other factors that are stated to be irrelevant."

The ground breaking part is that it tells you who is likely to bust in the first two rounds, who is likely to become a star, and how good they are likely to be. If the system weren't revolutionary, Ryan Leaf, Akili Smith, and Joey Harrington wouldn't have been drafted with top 3 overall picks, because everyone would have known that they were likely to bust, despite being highly regarded by scouts.

There is something revolutionary about a system that isolates two extremely important factors that NFL scouts don't accurately account for in their analysis of QBs. If a group of highly paid experts, who spend a great deal of time and effort trying to identify the best prospects, is routinely failing to correctly separate the busts from the star players, then a system that shows why they are failing, and how to correct the problem, is pretty revolutionary.

"That isn’t ground breaking any more than saying if a QB has the overall talent to be drafted in 1st or 2nd round then predict their success on starts and completion percentage."

The system isn't just saying that you should predict the success of 1st or 2nd round QBs with starts and completion percentage. It's saying that once you've determined that a QB has first or second round talent, the only things you should consider are starts and completion percentage, and that you should ignore all other factors, like arm strength (Ryan Leaf, Akili Smith, Mike Vick, Jamarcus Russell?), instead of thinking that an insanely strong arm means they will be a better pro. The fact that so many NFL scouts, GMs, coaches, and teams so often fail to do this, and that Lewin's system succeeds, is proof that it's revolutionary. It doesn't replace scouting, but it fixes an absolutely crucial problem with the current scouting process.

228
by sn0mm1s (not verified) :: Tue, 05/22/2007 - 1:49am

David, you are not understanding what I am saying. More goes into a QB getting drafted in the 1st or 2nd round than just completion % and games started. Therefore, those other factors do play a signficant role in QB success. If they did NOT play a significant role then you would just draft on completion % and games started.

By limiting the sample size to only 1st and 2nd rounds then you are basically saying if a QB is good enough to be drafted in those rounds (meaning ALL aspects of the QB are being considered) then completion % and games are the deciding factor.

229
by Alex (not verified) :: Tue, 05/22/2007 - 1:56am

Sorry, I meant #224.

230
by Alex (not verified) :: Tue, 05/22/2007 - 2:14am

#227: "By limiting the sample size to only 1st and 2nd rounds then you are basically saying if a QB is good enough to be drafted in those rounds (meaning ALL aspects of the QB are being considered) then completion % and games are the deciding factor."

Nobody is denying that scouting will still be important. Lewin's system isn't meant to completely replace NFL scouts. But it doesn't have to do that to still be a revolutionary system.

It's revolutionary because GMs, coaches, and NFL teams currently fail to recognize that once their scouts give the 1st or 2nd round stamp of approval to a group of QBs, the best thing to do is to ignore all the things that the scouts say about them, ignore all the amazing physical abilities and unforgettable highlights, and choose solely based on starts and completion percentage.

What's revolutionary is that NFL teams don't know that they should ignore the other factors, so they routinely draft QBs that bust despite (seemingly) 1st or 2nd round talent. This system could correct a huge problem with the way teams draft their most important players, and save them huge amounts of money and make them much better teams. Maybe that doesn't qualify as revolutionary to you, but I think if I showed some NFL teams how they could avoid drafting a QB bust with their first round pick, they'd consider it pretty revolutionary.

231
by sn0mm1s (not verified) :: Tue, 05/22/2007 - 2:27am

"The ground breaking part is that it tells you who is likely to bust in the first two rounds, who is likely to become a star, and how good they are likely to be. If the system weren’t revolutionary, Ryan Leaf, Akili Smith, and Joey Harrington wouldn’t have been drafted with top 3 overall picks, because everyone would have known that they were likely to bust, despite being highly regarded by scouts."

But if the system was truly predictive on college performance then you wouldn't need to limit the system to only the first two rounds of the draft. Your system says absolutely nothing about Zabransky, Leak,Clint Marks, or Troy Smith despite the fact that their completion percentage and games started compare favorably to Quinn and Russell. I hardly call that ground breaking if your system has NOTHING to say about those 4 other than "we just don't know".

Arm strength, mobility, TD:INT ratio etc. etc. do matter because those are all taken into consideration to get you in the first two rounds in the first place.

232
by David (not verified) :: Tue, 05/22/2007 - 2:30am

By limiting the sample size to only 1st and 2nd rounds then you are basically saying if a QB is good enough to be drafted in those rounds (meaning ALL aspects of the QB are being considered) then completion % and games are the deciding factor.

Okay, now you're just arguing semantics. Yes, gross differences in background, strength, etc. matter. But those are easy to spot, and you don't need a formula to tell you that Jordan Palmer will not be a great NFL starter. But a system that can tell the difference between Philip Rivers and Eli Manning from just their college careers would be very, very, very useful, and according to Lewin, when you make that comparison nothing but accuracy and games started matters. Unless you think that comparing high-round draft picks before they start their pro careers is already an exact science, I can't see how that's not a big deal.

233
by Sid (not verified) :: Tue, 05/22/2007 - 2:32am

RE: 116

I highly doubt he actually read PFP 2006.

234
by sn0mm1s (not verified) :: Tue, 05/22/2007 - 2:54am

#229

I agree that NFL scouts think they know more about football than they really do. Anyone could do Mel Kiper's job. I have always judged on completion percentage (hence why I have always thought Vick was a mistake). I am not saying the analysis isn't informative but it isn't ground breaking. Does FO think Drew Stanton is the best QB in the first 2 rounds?

What would make the system revolutionary is if it could bump up a player from the 4th or 5th round into the 1st or 2nd.

#231

It isn't semantics. I just think that to be predictive a system should be able to make predictions on the entire pool of QBs not just those drafted in the first two rounds. FO's system would say absolutely nothing about Brady, Warner or Brad Johnson (winner of 5 of last 10 superbowls).

235
by Alex (not verified) :: Tue, 05/22/2007 - 4:27am

#233: "What would make the system revolutionary is if it could bump up a player from the 4th or 5th round into the 1st or 2nd."

Well, it wasn't designed to do that. But, it can serve an equally valuable role by dropping a player from the 1st or 2nd round into the 4th or 5th. Avoiding huge busts in the first two rounds is arguably just as valuable (if not more valuable), as finding gems in the last few rounds, since most teams that need a QB try to find one in the first two rounds anyway.

I think you've just got an overly strict definition of revolutionary. Here's a definition from Merriam-Webster:

1 c: constituting or bringing about a major or fundamental change.

I'd say that a system that shows you how to choose correctly between Ryan Leaf and Peyton Manning, or between Tim Couch, Akili Smith, and Donovan McNabb, represents a major change in the NFL drafting process, because it would eliminate most high-profile QB busts before they happen. Considering how much money is wasted every year on QBs that become busts, I don't see how you could deny that it would be a major change if those busts stopped happening.

"I agree that NFL scouts think they know more about football than they really do. Anyone could do Mel Kiper’s job."

I don't know about that. I don't presume to know whether or not it's easy to identify whether a QB has the "right stuff", so to speak, to be considered a 1st or 2nd round talent. I imagine it's not a trivial task. Scouts are pretty good at weeding out untalented QBs, they just get a little too infatuated with some of the ones with lots of physical talent.

236
by Seth (not verified) :: Tue, 05/22/2007 - 4:54am

Does FO think Drew Stanton is the best QB in the first 2 rounds?
No, but they very well might think that Kevin Kolb is. Which I would call pretty revolutionary.

237
by jmparker (not verified) :: Tue, 05/22/2007 - 5:50am

Also, if I'm not mistaken the system says to take Quinn over Russell. Thats a 20 spot difference in the way NFL teams judged it.

238
by Seth (not verified) :: Tue, 05/22/2007 - 6:04am

You are not mistaken. I think someone pointed this out about 100 posts ago, but Lewin identified Russell last year as a potential first-round bust. Time will tell.

239
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Tue, 05/22/2007 - 8:22am

sn0mm1s,

Ideally, a perfect system would be able to predict results for every QB in the draft, plus undrafted ones. I agree that such a system would be revolutionary.

However, that is completely impossible. As Lewin pointed out in the PFP2006 article, many late pick QBs never get an opportunity to play, so it's possible that some of them would be successful and we'd never know. Additionally, in the late rounds you get QBs who played in fluky offenses like Hawaii or Texas Tech, and you can't draw the same conclusions from such players as from QBs who play in more traditional systems.

Nevertheless, there are plenty of QBs drafted in the first two rounds, and those are the most important ones from an investment standpoint. Who cares if some 5th round QB (like Troy Smith) can be predicted to succeed or not? It doesn't hurt the team very much if he is a failure, so he's only a small gamble. First and second rounders represent much bigger investments and much more risk. They are also the most likely to become successful.

It's too bad the system can't tell us about Matt Schaub or Troy Smith, but it can tell us about most of the QBs people have actually heard of before the draft, and it tells us this with (as I've mentioned) an unprecedented degree of accuracy. No other NFL analysis comes close. (This is a purely mathematical claim. The system is good for 74% of the total variability. That's almost three quarters! There is no other draft analyst or system in the world with that kind of hit rate.)

240
by James G (not verified) :: Tue, 05/22/2007 - 8:38am

The system may not be "revolutionary," but I'd say it's definitely ground-breaking. Although everyone would like to find hidden gems, and it doesn't do that, what it really does is weed out busts. Had teams figured this out before people like Leaf, Akili Smith, and even JaMarcus Russell would not have been nearly as high draft picks.

Sure, it doesn't do everything we want it to do, but in reality, it should change the way teams draft 1st and 2nd round QBs. And that's pretty ground breaking.

241
by James G (not verified) :: Tue, 05/22/2007 - 8:57am

I also want to point out to all these people claiming that stuff isn't ground breaking unless it produces a whole bunch of stuff that research is often incremental. Lewin laid some pretty important groundwork that other people could conceivably build off.

The same thing happened with DVOA, which I assume most people on this thread consider ground-breaking. The groundwork was laid in The Hidden Game of Football, but Schatz and company figured out how to use that to make predictions and calculate team strengths. It doesn't mean that Carroll et al's work wasn't ground-breaking.

I think Lewin's work is probably in a similar position, although, yes, it really does tell us something we didn't know before. Personally, I would have guessed that yards/attempt was the best indicator stat, and it's not.

242
by John (not verified) :: Tue, 05/22/2007 - 10:14am

Actually, what I think is revolutionary about it is that it's self-defeating. If people actually start to use it, and all of the sudden those low start / completion percentage guys that before were being taken in the 1st/2nd round now drop to the 3rd or 4th, the system breaks down. It's freaky-deeky circular I tells ya.

243
by hrudey (not verified) :: Tue, 05/22/2007 - 11:30am

In the context of the discussion, whether it's a truly revolutionary foolproof system or something as asinine as saying that Quarterbacks who butter their bread with their nonthrowing hands have better careers simply doesn't matter. What does matter is that a published, original (so far as we know) work was lifted and passed off as someone else's work.

244
by James G (not verified) :: Tue, 05/22/2007 - 11:48am

241 - I don't see how that makes it circular. If people used it, it would send draft busts further down the draft. Then, yes, it would have to be modified, just like any model that gets more input over time.

245
by Rich Conley (not verified) :: Tue, 05/22/2007 - 12:39pm

"241 - I don’t see how that makes it circular. If people used it, it would send draft busts further down the draft. Then, yes, it would have to be modified, just like any model that gets more input over time."

I dont think its defeatist. If you knock Russell out because of his low prediction, and make it so only guys with a good prediction are in the first two rounds, than you just pick the guy who has the best physical stuff, as they've all got the comp%, and Games started.

246
by Will Allen (not verified) :: Tue, 05/22/2007 - 1:00pm

Yes, it would seem that if the system works well (and the early indication is that it does), and is widely adopted, the result would merely be fewer 1st and 2nd round busts, along with fewer predicted busts. Perhaps, eventually, the system would apply to third rounders as well.

247
by Will Allen (not verified) :: Tue, 05/22/2007 - 1:17pm

I think Tavaris Jackson provides an interesting example to look at. He just barely eked into the 2nd round, by way of a trade which made him the last player selected. I wonder if there might be a difference between a player such as this and a guy drafted at the beginning of the 2nd round, to say nothing of a guy drafted at the top of the first. I certainly am not confident that he will end up being a decent qb, and Lewin's system doesn't predict that he will.

248
by Rich Conley (not verified) :: Tue, 05/22/2007 - 1:48pm

Will, it may be the case that as we get further out of the first round, other factors come into effect.

You dont get drafted in the first unless you have those physical things. Frankly, the physical differences between Quinn and Russel dont really make a difference, its going to depend on who has the mental stuff. When you start looking at a 3rd, 4th, etc, then maybe the differences in physical ability may come in.

It may be that other "factors" dont matter once you get to the first/second simply because everyone has good "factors" if they get drafted earlier. If a player gets drafted later, they've got a flaw somewhere, and there may be other numbers that make up for the loss of accuracy in the big 2.

249
by Pat (not verified) :: Tue, 05/22/2007 - 3:08pm

#245: The problem is numbers. There are far more crappy NFL QB prospects than good ones - obviously. If Russell's "true" QB performance level is, say, second-day, then if people started using this to correct for the scouting bias, players wouldn't move up - there would just be fewer QBs taken in the first round, and more later.

Yes, you'd then have JaMarcus Russell along with Jeff Rowe, Jordan Palmer, Jeff Smith, Lester Ricard - late round guys with excellent measurables, but either not a lot of information about them (Russell, Ricard) or just 'meh' results (Rowe, Palmer, Smith).

The question then would be "well, wouldn't you take a chance on Russell? wouldn't he be more likely to succeed?" Yeah, probably - but only minimally. There are so few first-round prospects anyway - digging through the trash is going to require a lot of digging.

250
by Will Allen (not verified) :: Tue, 05/22/2007 - 3:33pm

Yeah, Pat, I certainly don't think it could extend beyond the third round, if it even ever went that far. Guys are going to be drafted eventually; the position is just too important not to risk some draft picks on. Maybe the end effect is that fewer have first round money risked on them.

251
by Rich Conley (not verified) :: Tue, 05/22/2007 - 3:41pm

Pat, I agree with you on a lot of things here, but not all of them. I'm of the oppinion that there are a LOT more good QB prospects out there than people realize. I think theres a lot of the Flutie/Johnson effect where the better player never gets the chance because the other guy is "the guy" and has better measurables.

Theres no way to tell though, how many potential starters never got a chance, or how many potential studs got destroyed by bad teams.

The system MAY pull up the guys who have lots of games started, and good comp%, but are missing something, IE, guys who dont have cannons, or "aren't tall enough", etc.

I think the whole issue is that coaches are arrogant enough that they think they can coach up a raw physical talent, while they can't make a good player bigger/stronger.

252
by David (not verified) :: Tue, 05/22/2007 - 4:13pm

Maybe the end effect is that fewer have first round money risked on them.

You'd also get fewer season-killing busts. It's the teams with the most pressing need for a starting quarterback who draft one early; how many teams draft a QB 200th overall and then throw him into the starting job? If the Joey Harringtons and Akili Smiths of the world went in the fifth round instead of the first, they'd probably never rise above second on the depth chart.

253
by Rich Conley (not verified) :: Tue, 05/22/2007 - 4:22pm

"f the Joey Harringtons and Akili Smiths of the world went in the fifth round instead of the first, they’d probably never rise above second on the depth chart."

Theres also the possibility that 3 or 4 years of sitting on the bench learning how to play the game would turn them into actual players, as opposed to people who've skated by on physical talent alone.

254
by Pat (not verified) :: Tue, 05/22/2007 - 4:26pm

I’m of the oppinion that there are a LOT more good QB prospects out there than people realize.

I didn't make myself clear enough there. It's not the number that matters. That's just a definition of what "good" is. Raise the bar, the number drops.

What matters is this: suppose a first round QB is "100%". Then there are more "90%" QBs than "100%". There are more "80%" than "90%". There are far, far more "10%" than "80%". It's the slope of the talent histogram that matters more than the number.

255
by Alex (not verified) :: Tue, 05/22/2007 - 5:13pm

#251: "If the Joey Harringtons and Akili Smiths of the world went in the fifth round instead of the first, they’d probably never rise above second on the depth chart."

To be fair to Joey Harrington, he did rise to the top of the depth chart in Miami all on his own, beating out Daunte Culpepper fair and square. Of course, that doesn't mean a whole lot now, but still, if he was on a really bad team, he'd probably be a borderline starter.

256
by Paul (not verified) :: Wed, 05/23/2007 - 1:30pm

plager plagar
pelagai
plagairiz

Yeah, that thing!

257
by fiddycentbeer (not verified) :: Thu, 05/24/2007 - 9:18am

Leibniz and Newton invented the calculus, (virtually) simultaneously, and independently. Others, like Saki Kowa, way 'round the globe, worked a similar vein within the same period.

With all respect, Outsiders might consider that in the course of his 40+ year career, Brandt came on this independently. After all, the relationship between collegiate starts/completion % and subsequent performance on the pro level is simply a matter of observation and correlation. It's not rocket science.

Re #20: TMQ's addendum is nonsense. If, as Mr. Brandt claimed, he didn't see Outsiders work, then he is not obliged, as TMQ says, to credit FO's research and analysis. If he's lying, well, that's a different matter. However, the presumption of guilt found here is more than slightly disturbing.

258
by DB (not verified) :: Thu, 05/24/2007 - 9:51am

Wow...this thread is still going. (And, wow, I'm still checking on it.)

RE : 256 - The main question is not did Brandt figure it out on his own. It's that he lifted FO's actual text, even including the same anomaly. That's the plagiarism.

And as far as crediting FO, he should, just to state that his work isn't novel. If I had never seen, say, the laws of motion, working them out on my own wouldn't mean I should publish it and proclaim "look what I discovered". Instead I should research to see if anyone else has already figured this out. (Which isn't hard - a google search of "college qb projection system" takes you straight to FO.)

259
by Tarrant (not verified) :: Thu, 05/24/2007 - 10:57am

Re: #256

It is completely possible that Brandt came up with a similar statistic on his own.

It is, however, implausible that, in writing up that "independent research", he chose the same examples, in the same order, and happened to include the same singular error, about the same player, regarding the same statistic.

The odds of that having happened, if Brandt did all the work independently, are microscopic.

260
by jimmo (not verified) :: Thu, 05/24/2007 - 11:12am

"...in the course of his 40+ year career, Brandt came on this independently."

I agree this is a reasonable assumption. It is also reasonable to assume that within that same 40+year career, Brandt would have actually published some form of this very original analysis.
Maybe he was keeping the info for the Cowboys all those years? (If so, maybe he would've picked Marino in '83 instead of sticking w/Danny White)

261
by MCS (not verified) :: Thu, 05/24/2007 - 12:49pm

To reiterate what Tarrant said. Brandt made the same mistake of including the bowl games in the start number for McNabb, and only McNabb.

The original ESPN rough draft listed McNabb with 45 starts, because Lewin’s research only used regular-season games. A fact-checker at ESPN added McNabb’s four bowl games, but did not add bowl games to the total for any other quarterback. In the Brandt article, Donovan McNabb is the only quarterback listed with bowl games included as collegiate starts.

262
by Lou (not verified) :: Thu, 05/24/2007 - 12:53pm

256
Checkout the links in posts 107 and 112. If he had come up with this on his own you'd think he would've mentioned it when analyzing qb prospects in the past.

263
by fiddycentbeer (not verified) :: Thu, 05/24/2007 - 3:03pm

FWIW, I agree with #258 and #260 in that Mr. Brandt's replication of that "singular" error wrt game count is damning. That goes towards his, perhaps, having lifted text, but does not apply to his, perhaps, having ripped off FO's QB evaluation concept, the true subject here, or so I thought. Maybe not.

In the same fwiw-vein, I don't find the links in 107 and 112 convincing. Seems to me those were intended as prospect profiles, lighter fare than what we're discussing here.

While I have no idea what Brandt has written over all the years he's been in the game, so long as he was with Dallas, I would not expect him to share trade secrets like, say, QB evaluation.

#259:

Danny White came to the Boyz in 1976. They took DE Jeff Jeffcoat R1 in the 1983 draft, a few spots ahead of Marino. Jeffcoat did have 102 sacks over a 15 year career,12 with Dallas. Not a bad pick; not Marino great, but not awful.

264
by jimmo (not verified) :: Thu, 05/24/2007 - 3:59pm

fiddy, did you read the linked Brandt article that brought this on? It, too, is a prospect profile piece, analyzing the QBs in the 2007 draft. Not exactly the same format as the 107, etc linked pieces, but the same idea, it appears to me.
And no, I wouldn't expect Brandt to share this information while employed by the Cowboys, but that relationship ended in 1989, at which point he was presumably free to make his methods public. My admittedly limited google-centric research shows he did not do so until this year, after at least three different Lewin pieces had been published.

I'm aware that White wasn't drafted instead of Marino, just that Marino was available to the Cowboys at a time when their current QB situation could have been considered a weakness; White hadn't won a Super Bowl as a starting QB for Dallas, which at that time was a pretty big deal.

Also, though I've been blasting Brandt for stealing the system itself, I get the feeling that most here, and FO, believe the lifting of the actual text to be the bigger issue. It could be, though, simply that that particular argument is easier to prove, as outright theft of the projection system is, as you have pointed out, difficult to verify.

265
by jimmo (not verified) :: Thu, 05/24/2007 - 4:05pm

Hell, better yet check out the link on my name. Its the 2006 QB analysis, a near-copy of the 2007 story linked in the article above. A total TMQ-style cut and paste job, down to talking about the "big three" this year, when there was only the big two, except of course, with no mention of college starts and completion %.

266
by David (not verified) :: Thu, 05/24/2007 - 6:45pm

FWIW, I agree with #258 and #260 in that Mr. Brandt’s replication of that “singular� error wrt game count is damning. That goes towards his, perhaps, having lifted text, but does not apply to his, perhaps, having ripped off FO’s QB evaluation concept, the true subject here, or so I thought. Maybe not.

Well, if he lifted text and figures fromt he article, he must know that it was written a year before his own piece. In that case, even if he hadn't copied anything, it would still be courteous to give a nod to the people who thought of this first - plus it would remove any appearance of plagiarism. As it is, all the evidence fits his having lifted the whole concept, and there's nothing to specifically support his developing it on his own. We can't prove he didn't, but other than the benefit of the doubt I can't see anything that would make me think this isn't plagiarism.

Plus, remember that he pointed out both the games-started and completion-percentage figures as important in the lede, and then only discussed one. If he'd figured all this out on his own, would he really be likely to forget to write about fully half of it?

267
by fiddycentbeer (not verified) :: Fri, 05/25/2007 - 10:43am

jimmo:

I suppose I did skim Brandt's 2007 QB piece, don't remember it as standing out. My draft eval tends to focus on different info, as found at Rob Rang's site, or Brandt's own (presumably)Pro Day compilations. IMO, profiles are puffery; Kiper, a major purveyor, being unreadable.

David:

Agree: if he lifted, then he was obligated to acknowledge the original. Per TMQ, #20, Brandt claims that he didn't lift, that he was unaware of FO's work.

I don't know the truth of this matter tho, again, the "McNabb error" is damning. The rest, not so much; listing recent top 10 (or so) QB prospects per declining # of collegiate starts being an excellent means of illustrating one main point. While I know nothing of the Web site guru biz, I hazard that, maybe, the completion % portion of Brandt's article took an editor's knife. Like most of this, mere speculation.

As you say, benefit of the doubt is a factor but since it wasn't my work that was (maybe) ripped that's easy for me to say. I do suggest this:

TMQ profile of Brandt, in #20 above, dovetails with much I've seen elsewhere. Suffice to say none of that describes one who would steal another man's work. More the opposite. Then, except as he's fooled many for long these many years, I consider that, maybe, he's not a plagiarist.

268
by DGL (not verified) :: Fri, 05/25/2007 - 12:53pm

#266: The rest, not so much; listing recent top 10 (or so) QB prospects per declining # of collegiate starts being an excellent means of illustrating one main point.

There were 35 college QBs drafted in the first two rounds in the last 10 drafts (97-06), 27 of which were drafted in the first round - and he just happened to pick the same 8?

If I'm doing the math right, there are roughly 2.2 million combinations of 8 quarterbacks that could be chosen out of 27. So the odds of Brandt randomly picking the same 8 QBs to cite as examples are roughly one in 2.2 million.

Seems pretty compelling to me.

269
by Lou (not verified) :: Fri, 05/25/2007 - 1:03pm

What i don't get is, if he did this study on his own, how did Brandt define a successful QB? Lewin used DPAR. Did Brandt just use his subjective judgement?

270
by fiddycentbeer (not verified) :: Fri, 05/25/2007 - 4:00pm

#267:

Mr. Lewin's work considered R1 & 2. If Mr. Brandt didn't see that, as he says, then Lewin's field is irrelevant to Brandt's inquiry.

Brandt's text,(pilfered or no) lists 8 QB. Of those, 7 were drafted in the top 4 overall. If we imagine, for a minute, this (and within past decade) to be Brandt's field of fire, then he excluded only: Couch, Vick, Carr and Manning, the Younger. Vince Young too, but he doesn't have much of a pro track track record. Others, close but no cigar, may include: Leftwich (7th), Roethlisberger (11th), Leinart (10th)and Cutler(11th).

Taken at (almost) his own terms, within round 1A, as the scouts say, that reduces our basis from 35 to as few as 11. However...

In Brandt's, Druckenmiller is the outsider, drafted in the deep 20s or so, somewhere 'round the JP Losman range. Or Rex Grossman range...and so on.

Seems to me this Druckenmiller Anomaly ranks near the McNabb (thanx to whomsoever christened the McNabb Anomaly) on the evidence list. Again, text only; concept, well, I just don't know.

271
by Pat (not verified) :: Fri, 05/25/2007 - 5:24pm

I can't believe anyone would think that this isn't plagiarism. The ESPN article contained 10 statistics: Manning/Leaf's starts/completion percentage, and Rivers, McNabb, Palmer, Harrington, Druckenmiller, Smith's starts.

Brandt's article contains the exact same statistics. No more, no less. There was no reason for Brandt to include Manning or Leaf's completion percentage - other than the fact that he was lifting the article wholesale.

272
by Felton (not verified) :: Sun, 05/27/2007 - 10:51am

We can settle this pretty quickly - Brandt's penalty will be season tickets for the Raiders next year - he will have to sit with Aaron Brooks and Art Shell. If he still appears recidivistic, he will have to buy all the tickets for the 2011 Super Bowl in Dallas without a view of the field.

273
by Justin Zeth (not verified) :: Mon, 05/28/2007 - 3:36pm

You don't need a projection system for 3rd-5th round picks. Quarterbacks picked in those rounds almost never develop into high-quality NFL starters. Here's the list of all QBs drafted in rounds 3-5 in, let's say, 1996-2003:

Bobby Hoying. Jeff Lewis. Danny Kanell. Danny Weurffel. Pat Barnes. Shaun King. Brock Huard. Joe Germaine. Aaron Brooks. Kevin Daft. Giovanni Carmazzi. Chris Redman. Tee Martin. Chris Weinke. Sage Rosenfels. Jesse Palmer. Josh McCown. David Garrard. Rohan Davey. Randy Fasani. Kurt Kittner. Brandon Doman. Craig Nall. Dave Ragone. Chris Simms. Seneca Wallace. Brian St. Pierre.

Read over that list again. The very best quarterback on that list is Aaron Brooks. The only other guy on the list you could even make an argument for as an NFL starter is Chris Simms. And those aren't exactly the kind of guys you stake your Super Bowl hopes on.

You can get a decent backup after round 2, if you're lucky; even then, it's more likely you're getting someone who can't play. Any team thinking it's getting a future starter after the second round is either stupid or delusional.

No one needs a statistical system to figure this out; one need only spend two minutes at drafthistory.com. This is one nice thing we can say about NFL scouts: They're very good at identifying who does and doesn't have the skills to be an NFL quarterback. Nobody who lasts past the second round ever makes it, with the sole 1-in-200 exceptions of your odd Tom Bradys and Kurt Warners and Marc Bulgers.

We need a projection system for rounds 1-2, because that's where QBs are generally evenly split between smashing successes and hideous failures, and until recently, no one really knew how to tell the difference in advance. So yes, David Lewin's research IS tremendously important.

By the way, next time you point out that post-round-2 quarterbacks never make it in the NFL, be sure to cite my original research. Those were two very intense minutes of research I did there.

274
by coyotl666 (not verified) :: Sat, 06/02/2007 - 8:10am

>>The word "genius" isn't applicable in football. A genius is a guy like Norman Einstein." --Joe Theisman>>

hope i've got the link thing working. attempt is to link to a very thorough and fascinating piece on plagiarism and the need to have free and fair access to what has come before from the artist's perspective, in this case from eddie campbell a scottish artist now long transplanted to australia. perhaps not known by most here, but that don't devalue his ideas on plagiarism. he's probably best known for his collaboration with alan moore on from hell, later butchered by the hughes brothers with the superpowered assistance of johnny depp, who i generally quite like.

cheers, mark

275
by coyotl666 (not verified) :: Sat, 06/02/2007 - 8:15am

ok, the link in the above post gets you to the blog but not the article on plagiarism. scroll down on the right hand side near the bottom and hit the plagiarism link under the some labels heading

276
by coyotl666 (not verified) :: Sat, 06/02/2007 - 8:33am

272 >>Nobody who lasts past the second round ever makes it, with the sole 1-in-200 exceptions of your odd Tom Bradys and Kurt Warners and Marc Bulgers.>>

well just how disengenuous is that? if 3 of 32 starting quarterbacks are 1 in 200 exceptions i now understand why i had so much trouble in maths class. i'd have guessed that the results were close to 10 percent, maybe even higher given that your list contains 26 names (though oddly none of the three exceptions. and of those 26, even more than 3 were full time starters at one time or another. no, there are no real "household names" or clear superbowl winning talent there, but as we know neither dilfer or hostetler were on anyone's list to come away with rings.

277
by Scott de B. (not verified) :: Sat, 06/02/2007 - 10:43am

coyotl66:

The person you linked too seems not to know the difference between plagiarism and copyright violation. Two different things.

278
by Pat (not verified) :: Sat, 06/02/2007 - 11:33am

well just how disengenuous is that? if 3 of 32 starting quarterbacks are 1 in 200 exceptions i now understand why i had so much trouble in maths class.

One-in-200 is too low, but 10% is way, way, way too high as well.

A quarterback's career lasts typically ~12 years. Around 10 players are drafted each year between round 3 and the end of the draft. That's 120 players over 12 years - adding in undrafted free agents, about 150-175 sounds about right. Depending on how you count the number of players who succeed, it's about a 5% chance.

279
by Sid (not verified) :: Sun, 06/03/2007 - 9:59pm

RE: 263

I consider the system the bigger issue. The plagiarism is what proves that Brandt straight out ripped Lewin's system. Proving Brandt didn't figure this out on his own would be difficult, if not for the fact that he quite obviously directly ripped off written words.

280
by Steve (not verified) :: Thu, 06/07/2007 - 11:53am

I respect what FO has done, of course, but the problem I have here in claiming plagiarism is that the information on which FO based its College QB Projection Theory is readily available to anyone. This is not the situation of the Boston writer who took verbatim passages from an article in an another paper. FO's theory is based on research readily duplicated. And it is not a complex formula we're talking about here either. You simply look at starts and completion percentage for college QBs in the past and look at how they performed on the NFL level. From that simple research, you'd see that more often than not, the college QBs with more starts and higher completion percentage performed well on the next level. This is not like the DVOA or DPAR formula FO invented. If you want to call the DVOA and DPAR a "high school senior" level formula then the College QB Projection Theory is the stuff of kindergarten.

281
by Zac (not verified) :: Thu, 06/07/2007 - 4:44pm

Re: 279. I thought this thread was over. Anyway, see 190 for a preemptive rebuttal. Just because all the information used is readily available, that doesn't mean someone has the right to steal your analysis of the information, especially when it's so obvious that they've done so.

282
by Steve (not verified) :: Thu, 06/07/2007 - 5:41pm

Re: 280. Thanks for the response, and sorry for resurrecting this. I've read post 190, which is on point, but does not address the issue that college starts + completion % as an indicator of NFL success is not a novel or profound idea. It is merely a statistical compilation from readily available data. Brandt plagiarized not the research or theory but the actual article that the theory was described in by Lewin. That is how there is same "oddities and same order of names and same strange mistakes" as listed in post 219. FO should be complaining that an article of theirs was plagiarized, rather than this data compilation.

283
by Medea (not verified) :: Thu, 06/07/2007 - 6:34pm

At least Brandt plagiarized something I could care less about. Prior to the 2005 season, if he had brazenly announced that Kevin Jones was going to be some kind of Fantasy Jesus, I'd hate his ass too and consider him to be a blowhard hack who didn't know what he was talking about.

284
by Riceloft (not verified) :: Tue, 06/12/2007 - 7:21pm

Lets try this again. Click my name :).

285
by andy3887 (not verified) :: Wed, 04/22/2009 - 5:55pm

footballoutsiders.com can't "prove" anything, they can only give evidence, and then we weigh the evidence. the fact that brandt used the same stats isn't that noteworthy, but the fact that he used all the same examples and that he also produced similar errors (donovan mcnabb's number of starts) is a sure-fire sign of plagiarism.

whenever there are academic reviews of a researcher's work and they find plagiarism, usually the evidence that points to plagiarism conclusively are errors that are reproduced verbatim along with arbitrary examples reproduced verbatim. why couldn't gil brandt use other qb's than the ones he used? there are lots of examples that prove Lewin's Forecasting System to be quite good, brandt could have used ANY of those examples yet he didn't. that's as close to "proving" plagiarism as it gets. the evidence is very strong.