Niners Nation hits the big time
by Stoned Slacker on Nov 17, 2009 3:56 PM PST
Email
PrintWell it played a small role in a petty spat, but the people involved were fairly famous. Author Malcolm Gladwell was ripping author Steven Pinker, who had unfavorably reviewed his new book. One of his criticisms was the fact that:
"Pinker’s third source was an article in the Columbia Journalism Review, prompted by my essay, that made an argument partly based on a link to a blog called “Niners Nation."
He uses us as a belittling shot, obviously unaware of the quality analysis happening here daily.
The article he references was this one, which was actually a review of some Football Outsiders stats. Ironically I think Gladwell cited his own stats from F.O. in the original piece on QBs that caused the argument. So its all pretty convoluted and unresearched on his part.....but funny place to come across the site.
This is a FanPost and does not necessarily reflect the views of Niners Nation's writers or editors. It does reflect the views of this particular fan though, which is as important as the views of Niners Nation's writers or editors.
35 comments
|
0 recs |
Do you like this story?
Comments
Oh....
The article that cited Niners Nation is here:
Ha
Fooch actually e-mailed me about this the other day. I hadn’t had a chance to look into it with much detail, but it looks like you did all the hard work for me!
I think it’s a difficult point that’s being made, particularly because “positions 11 through 90” in the draft is, like… 80 people every year you’re looking at, while the top ten is, well, you get the picture.
The pools being sampled there don’t seem very equitable to me, because the difference in range is so vast.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
That 11-90 number
does it take into consideration QBs who never get a chance to play, or only starting QBs? If so I wouldn’t trust the conclusion that’s being drawn because it’s self-selecting for the successful quarterbacks.
It’s based on a “per play” metric, so I’m guessing that it wouldn’t punish the sample range for players who don’t hit the field.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 17, 2009 4:31 PM PST up reply actions
Gladwell is citing stats created by Dave Berri
And while I’m more familiar with Berri’s basketball stats, his extrapolations of per play stats have always seemed a bit of a stretch. He often takes guys that do limited things well in limited time and concludes that they’re among the more effective players overall.
And of all the statheads I’m familiar with, his basketball stuff has always seemed less accurate the more you know about the players in question. Its supposed to be this clever, counterintuitive system, but I just don’t think its as great or predictive as he says it is. Lots of massaging of numbers and subjective factors in his Win Score metric to make it appear accurate.
by Stoned Slacker on Nov 17, 2009 5:25 PM PST up reply actions
post
Actually after I emailed you, I found Steven Pinker’s email address and emailed him directly. Pinker pointed me to his sources and I came across the CJR article.
I think generally Gladwell was saying, the more you google the more you can find…but I’ll take any comments I can get about the site. I suppose any publicity is good publicity, right?
by David Fucillo on Nov 17, 2009 4:57 PM PST up reply actions
Just read the article and Danny's article
Is 37 games started in college really magic number? How many QB’s really start for 3 seasons in college these days? That’s a very high threshold, you almost have to be QB prodigy, but maybe that is the requirement for NFL success.
Bill Parcells' requirements
3 seasons as starter.
23 wins
Senior and/or have degree.
You’ve got to have experience before you try it in the NFL. The NFL really isn’t a place for on-the-job training (though there is some of that going on). If you don’t develop the necessary skills before coming to the NFL it’s going to be tough to develop them there.
I’m actually working on a fanpost where I’m comparing all the starting QBs in the NFL today and seeing which ones fit Parcells’ requirements. I’ll be interested to see if it’s actually a doable system.
For a little bit of a teaser.
Peyton Manning—3 1/2 year starter (stepped in his freshman year), senior, college degree and 34 wins.
Ryan Leaf—2 year starter, left as a junior, didn’t have 23 wins.
Parcells wouldn’t have even put Leaf on his board (and to think that once upon a time the debate as to which QB would be a bigger NFL star was fierce).
I’m not sure where you can find the numbers without actually looking up every individual day-one quarterback pick for the last five years… but I think the number of guys who do play three years as a starter might be surprisingly high. Since most “freshmen” are second year players to begin with, and a lot of programs emphasize graduation (like Notre Dame, though not many programs as energetically as Notre Dame).
Just off the top of my head recently:
Brady Quinn – 4 year starter (Notre Dame, though)
Josh Freeman – 2.5 year starter (34 starts)
Chad Henne – 4 year starter
Matt Leinart – 3 year starter (39 starts)
Vince Young – 3 year starter (37 starts)
JaMarcus Russell (LOL) – 3 year starter (35 starts)
Jay Cutler – 4 year starter
Sam Bradford – would have been a 3 year starter
Colt McCoy – 4 year starter
Jimmy Clausen – 3 year starter
Seems pretty common based only on that. It’s somewhat cherry-picked, sure, but mostly in the sense that it was all off the top of my head.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 17, 2009 6:21 PM PST up reply actions
Got to factor in the other two parts of the equation
23 wins and a senior/finished degree.
Quinn fits
Freeman doesn’t
Henne does
Leinart does
Young doesn’t (2 years as full starter. He alternated his freshman year)
Russell doesn’t
Cutler doesn’t (11-24 as a starter)
Bradford doesn’t (doesn’t have 3 full seasons as a starter)
McCoy does
Clausen doesn’t (isn’t a senior)
I wasn’t talking about Parcells’ criteria, though.
According to Football Outsiders (and they did some very fancy math to “figure” it out), there are something like three strong indicators of success for a draft pick. One of them is starting 37 college games. All I wanted to do in the above comment was address bignerd’s question about how often a high-profile quarterback even stays in college that long.
There are the other parts of the FO indicators. I believe that one of them has to do with completion percentage and the other has to do with scouting, of all things.
But the only one I was interested in as far as my comment was concerned was number of starts, because that’s what bignerd brought up in his comment.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 17, 2009 8:32 PM PST up reply actions
How does Aaron Rogers really qualify in this assessment? He was a starter for more than 2 years, it’s just he spent some time playing QB at the junior college before getting plucked by Cal.
I’m not sure, to tell the truth. It’s an assessment that I’ve never really been comfortable with. I don’t doubt the numbers, but I trust them only insofar as they work for what has already happened. In the past, first and second round draft picks who have x starts and y completion percentage have done well. And as far as I know, that’s all that the system was designed to tell us, because it doesn’t give us any correlative reasons for why these things should continue to hold true in the future.
Why number of starts? Is it because more college experience actually does create more preparedness? I want to say that age would be important, but you can be a senior and only start 25 games, so that’s not it. But why is that extra 5, 6, 11 starts so important? If you’ve reached a certain level of skill, that should carry over regardless of start numbers. The real reason 37 starts is important is because that’s the number that has worked out the most often in the past.
Why completion percentage? Why wouldn’t there be a distinction between a guy who throws 66% in a pro-style offense and a guy who throws 66% in a spread style? We don’t know, but the numbers bear out for the last 15 years…
Scouting is the only part that makes logical predictive sense to me. The rest of it is kind of buried in a pile of “it worked before but we don’t know why.” it seems fairly non-predictive to my mind.
But I’m still glad it’s there because it at least gives me something to look at that isn’t arbitrary. These factors have been significant in the past without a lot of question. It’s fair to assume that they might continue to be, as long as you keep that assumption in perspective.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 17, 2009 9:56 PM PST up reply actions
Gotcha
As far as the extra starts—it’s the experience factor. A QB with those extra starts under his belt is going to be able to make his throws quicker, do his checkdowns quicker, not panic if he’s behind, etc., etc., etc. You only get that by playing and you don’t want those extra games to come in the NFL where it might cost you your season.
I’d think that completion percentage is fairly obvious—if a QB can’t complete 60% of his passes in college he’s not going to be able to do that in the pros with better talent on the opposing defenses. If he’s not completing 60% of his passes he’s not going to be effective at moving your team down the field to get into scoring position.
You only get that by playing and you don’t want those extra games to come in the NFL where it might cost you your season.
And there’s my problem. Saying that, you concede that the experience can come in the NFL. Maybe you don’t want to have to sit through the growing pains, but playing is playing – right? What’s so important to development about playing those games in college as opposed to anywhere else? It’s not the “experience” factor. It is very, very specifically the “college experience” factor. Why do those extra college games equal having a better NFL career? I absolutely understand why they might make you more NFL-ready coming straight out of the draft, but being more NFL-ready at the beginning of your career has never been the same as being more likely to succeed over the course of your career.
And for percentage, I agree that accuracy is obviously important, but why contextless accuracy? I mean, I can get a guy who completes 70% of his passes in some ridiculous college offense. That doesn’t to my mind make his percentage more impressive than somebody who did somewhat less with a different, less efficient offensive scheme.
I guess if I was going to say something, it would be that the scouting factor weeds out the percentage outliers… but then why is completion percentage important at all? If it’s already being taken care of by the scouting step, then the guys who don’t fit the bill shouldn’t be part of the equation to begin with.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 17, 2009 11:22 PM PST up reply actions
We know that scouting is an inexact science to say the least. To me, it seems to make sense that QB’s with a higher completion percentage would be more likely to succed in the NFL. As you said the scout is the intelligent part of the equation, he filters out the players who excelled in a gimicky offense, or didn’t face an impressive level of competition etc. Sure you could tell your scout to only look at players that have a 60%+ completion percentage but you are still using that as a variable by doing so.
As far as “College Games Started” is concerned I think it can have an impact on a players overall success at the NFL level. This has to do with one of those pesky intangibles that is hard to measure through statistical analysis. Arrogance vs. Confidence. You need confidence to succeed in the NFL because it is based on a belief that you have what it takes to succeed. Confidence has a strong foundation and it is much more difficult to destroy by a lack of results. College game experience helps to build this foundation. Arrogance on the other hand is a perception. It is a shell that has very little foundation and collapses at the first signs that the perception is incorrect. Players who have the potential to succeed but do not have the confidence to succeed typically struggle in the NFL. If the player doesn’t have the opportunity to build that confidence in the NFL, they usually fail.
Hey, remember that one time when we were arguing and I was wrong? Yeah, me either.
by SanFranSoldier on Nov 19, 2009 11:11 AM PST up reply actions
Read the article Florida Danny did that I linked below.
I’ll steal this quote from it though:
Thankfully, several people posted lists of college QBs in response to mikev’s question. What I’m going to do is look at how each one stacks up according to the LCF. But before I do, I want to apply it to the 49ers current QB situation, or more accurately, whether the choice of Alex Smith over Aaron Rodgers was the correct one given the LCF. Here’s how they stacked up:
Alex Smith — 66.3% COMP PCT, 23 GS
Aaron Rodgers — 63.8% COMP PCT, 22 GS
Clearly, you can see that the correct answer to the question, “Alex Smith or Aaron Rodgers?” was, “Neither.”
They call their best player "Kung Fu Panda" and they complain that people aren’t taking them or the game seriously enough? -Nick
Florida Danny did an article on it.
Its called the Lewin Career Forecast: read this: http://www.ninersnation.com/2008/10/28/648563/a-statistical-look-at-draf
Basically, 37 starts with a 60% completion rate are the main criteria for Division 1 college QBs, and I think it only applies to guys taken on Day 1.
They call their best player "Kung Fu Panda" and they complain that people aren’t taking them or the game seriously enough? -Nick
Actually, I rather like Danny’s explanation for why 37 starts is so important. It’s not specifically that the experience makes the player better. It’s that the larger sample makes the scouting better. That’s almost completely satisfying to my mind.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 18, 2009 2:39 PM PST up reply actions
I was guessing sample size too
Model is still a bit sketchy. A QB still has to play 37 games, disqualifying several candidates. Relies heavily on scouts being able to determine if a QB is a 1st or 2nd round prospect.
Gladwell’s piece covered Chase Daniels, his flaw wasn’t revealed until his team was against the wall in a big game and he was pushed out of his comfort zone. Scouts might not even get a look at McCoy or Bradford outside their comfort zone when a play has broken down.
It’s a test with some accuracy but flawed with both false positives and incorrect negatives. Better than nothing.
The 37 game number might have to be larger given the changes in college football. A decade ago a QB prospect playing for a big time college program might have played in 4-5 big, challenging games a season. These days college programs fill there schedule with tomato cans and hope to play one big game at home before getting a big bowl birth, which still might not be a competitive. Even less chances to scout and find faults with QB’s. Sadly the two conferences Big 12 and SEC guilty of this trend are the ones with the potential QB prospects.
at least we get conference championship games
A hearty thank you to Rich Aurilia for all the good memories, and to the Niners for finally getting the uni's (mostly) right.
I love that Gladwell’s argument revolves around the other guy’s sources. Gladwell’s at his best when he’s dry-humping cherry-picked sources to back up his thesis while forgetting to footnote. Thesis first; sources second. Remember that, kids.
Never read Blink. I did read Tipping Point, and it frustrated the hell out of me.
by Grant Brisbee on Nov 17, 2009 5:28 PM PST up reply actions
three words...
holy fucking shit! sorry fooch for the bad language, but when one’s article gets cited by pinker in a spat with gladwell, one really can’t contain themselves. :-)
by (Florida) Danny Tuccitto on Nov 17, 2009 8:06 PM PST reply actions
How much would you pay.....
to see a steel cage match between these two?
vs. 
Not much.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 18, 2009 9:36 AM PST up reply actions
Well, I mean, sure. I’d pay much if I could afford it.
Context, people. More context is good. Less context is bad. If you're willing to be reductive, then you're willing to be wrong.
by howtheyscored on Nov 18, 2009 12:55 PM PST up reply actions
I wouldn't pay any
Who in there right mind would want to see two grown men slapping each other while making high pitched noises?
This dude called..
and he wants his hair back.
Well, we're waiting....
by drummer on Nov 18, 2009 8:15 PM PST up reply actions 1 recs
$0.001
What we've got here is a failure to communicate.
by SportsChicken on Nov 20, 2009 3:50 PM PST up reply actions
wait, are those two different people?
I thought they were just different photos of the same guy.
Maybe Gladwell is actually angry because Pinker stole his hair tonic. (or vice versa)
Jason Hill is turning the corner!
Dude
Pinker is a Douche rag
He thinks we are Darwinian survival Machines. Takes all of his Biological theory from Dawkins and Wilson. He’s basically extends the oversimplification that is Sociobiology into an over-reductive paradigm that favors his linguistic backgrounds. Many of his claims about the origin of language are increasingly being refuted.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Shutdown Corner - NFL - Yahoo! Sports
- Giants to be honored with commemorative NY license plate; some would rather see a 9/11 plate
- The hidden play of Super Bowl XLVI: Justin Tuck’s third-quarter sack
- Report: Chad Ochocinco will become Chad Johnson once again
- Video: Bill Belichick implores his defense to make the Giants throw to Mario Manningham
- Training Day: Learning the 40 with an Olympian
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
