« Pinker on "What the Dog Saw." | Main | Letting Igons be Igons »

Comments

Christopher Horn

Speaking anecdotally of Peyton Manning, I have in mind that the great QB draft debate of 1998 went something like this:

Pro-Ryan Leaf: better physical skills/stronger arm, more upside (whatever that means), no baggage from highly visible painful losses such as Manning's 0-4 record against Florida.

Pro-Peyton Manning: better work ethic/preparation/discipline.

Consensus first choice: Leaf.

Maybe its too late, and this is just stream of consciousness commenting anyway, but could there be a Moneyball-type revolution for QB evaluation based on the primary importance of 10,000-hour type preparation?

David

The problem here, I think, stems from one of cluelessness. Pinker criticizes Gladwell as clueless, and Gladwell responds in kind. Yet both of them seem to misunderstood certain aspects of the game of football, which is not surprising, as such a base, primitive sport could be little more than a rare diversion in the lives of intelligent men.

Both Gladwell and Pinker fail to understand the notion that a quarterback's "performance" is utterly dependent on a number of factors that (one presumes) are directly tied a team's draft ranking in the previous year.

In other words, the relationship between QB performance and draft rank is dependent, and these variables cannot be treated as independent. Unfortunately, that is precisely what the Journal of Productivity Analysis paper does (though their approach to indexing performance is quite reasonable).

But then, this is something more seasoned football fans know -- even the most promising quarterbacks can't do much for you when (1) their receivers drop balls, (2) their line is prone to collapse, and (3) their coaches underprepare them and/or call poor plays.

This circus of mishaps is precisely what you'd expect among low ranked teams, yes? Which is why gauging the quarterbacks' performances naked, as it were, will lead you to the wrong conclusions.

GC in DC

Interesting post, I'd echo a lot of John of Sparta's response. I'd make three other observations.

First, GMs operate in different political environments and their decisionmaking is circumscribed by different factors. Bill Belichick has enormous lattitude in New England to focus solely on football factors. But some teams are under pressure (either from the fan base or the owner, or both) to land a high-profile college player, some are under pressure to draft a player with local roots, some are under pressure to draft for specific traits favored by the owner (Raider owner Al Davis's speed obsession, for example).
Second, it's important to assess the environment within which the high-first-round QB is thrown. Peyton Manning was a top pick who is able to thrive, in large part, because his team has maintained a consistent offensive system and strategy thoughout his career, allowing him to improve from year to year. Others haven't been so lucky, and some have injured their own careers by holding out and missing critical training camp time. It would be a fairer sample group if limited to exclude differences in system and coach turnover and, roughly, preparation time.
Third, NFL coaches and GMs beyond Belichick have caught onto the notion that top draft picks aren't as valuable as they may seem. Last year's draft saw lots of teams trade first round picks for multiple lower-round picks, with the real sweet spot being late in the first round and early in the second.

Paul Q

Malcolm,
It is infinitely more complex than I think the average NFL fan percieves, or at least the ones who have not played the game. For instance, so much of a QB's success is reliant on how good the running game is and how well his line protects him. If the running game is terrible, the linebackers usually give themselves a bigger cushion from the line (because they are less worried about a run play breaking) making it more likely that the QB could be intercepted. Also, if the line can't give at least moderate protection, the QB is dead in the water. My point is that college QBs with a good team around them already have an advantage. Football is the ultimate team sport and the quarterback position is the most reliant on the other players making assessment incredibly difficult.

Steve Sailer

By the way, have you noticed that NFL quarterbacks are doing pretty good these days? Five quarterbacks have passer ratings over 100 and seven more over 90.

Vince Kellen

If someone could reliably predict future human performance, everyone would soon be able to do the same(if the method was shared). Top picks' values would inflate and competition would then move to the second tier picks: those that rank low but could do well despite what the data says. If one could reliably predict most of future human performance, none of us would be here as most or all of sport (and the stock market) would be uninteresting.

Unpredictability of some things may be a deep design of those things. Some things are more easily dealt with as random than as causal.

Also, despite its problems, overoptimism may be beneficial and a critical ingredient for becoming highly skilled in anything. One needs a bit of delusion to work through the long trough of "average" on the way to excellence.

Michael Ott

Malcolm,

I think your material and insight is great. I appreciate all the writing you do, whether i agree or understand it completely is of no importance. It helps me think outside the box and view things in a new way. I am disappointed to see the insulting and hostile comments left on some of your postings. I wish some people would focus more on the big picture, rather than sweat the details that they either dont agree with or understand. Tone is a significant indicator of educational and professional maturity. I look forward to your books, articles, and blogs. Keep up the great work and keep questioning the conventional approach to thinking. Happy Thanksgiving.

Michael Ott

Also, i think a more important indiciator of quarterback and team success depends on the ability of a team's offensive line. Tom Brady and Bret Farve would not be the quarterbacks they are without their stellar offensive lines. Or maybe instead of analyzing the makeup of the offensive line, take into account the correlation between pressured drop backs (hurries) and a quarterbacks success...or maybe elapsed time between snap and pass...quarterbacks are keeps of the ball, but they cant do much without an experienced and solid offensive line.

Ethan Hunt

I think when it comes to statistics, the way someone like Lewis did with Moneyball is the best way to approach it. Maybe the folks on Wall Street need to call up Charlie Epps?

Doug Okamoto

Overall Draft Picks and Their Passing Efficiency Ratings

As of Week 13 in the current NFL season. New Orleans Saints QB, Drew Brees, who was the first pick in the second round of the 2001 NFL draft, 33rd overall, is top-rated passer among starting quarterbacks with an efficiency rating of 111.3. Oakland QB, JaMarcus Russell, whose efficiency rating is a league-low 47.7 was chosen first overall in 2007. Other No.1 overall draft picks, Peyton Manning, Eli Manning, Alex Smith, Carson Palmer and Matthew Stafford rank 5th, 12th, 16th, 18th and 31st, respectively. Three quarterbacks who were never drafted, Kurt Warner, Tony Romo and Jake Delhomme, rank 8th, 10th and 32nd.
.
Correlation Analysis

Plot passer efficiency rating on the y-axis versus overall draft pick on the x-axis. Since the correlation coefficient, r=-0.02, is not significantly different from zero, efficiency is uncorrelated with overall draft pick. Assuming passing efficiency is a surrogate for winning, the earlier or later a QB was drafted is predictive of neither winning nor losing.

Passer Rating (Wikipedia)

http://en.wikipedia.org/wiki/Passer_rating

Douglas Macer

Overall Draft Picks and Their Passing Efficiency Ratings

As of Week 13 in the current NFL season. New Orleans Saints QB, Drew Brees, who was the first pick in the second round of the 2001 NFL draft, 33rd overall, is the top-rated passer among starting quarterbacks with an efficiency rating of 111.3. Oakland QB, JaMarcus Russell, whose efficiency rating is a league-low 47.7 was chosen first overall in 2007. Other No.1 overall draft picks, Peyton Manning, Eli Manning, Alex Smith, Carson Palmer and Matthew Stafford rank 5th, 12th, 16th, 18th and 31st, respectively. Three quarterbacks who were never drafted, Kurt Warner, Tony Romo and Jake Delhomme, rank 8th, 10th and 32nd.

http://en.wikipedia.org/wiki/Passer_rating

Correlation Analysis

Plot passer efficiency rating on the y-axis versus overall draft pick on the x-axis. Since the correlation coefficient, r=-0.02, is not significantly different from zero, efficiency is uncorrelated with overall draft pick. Assuming passing efficiency is a surrogate for winning, the earlier or later a QB was drafted is predictive of neither winning nor losing.

Douglas Macer

Overall Draft Picks and Their Passing Efficiency Ratings

As of Week 13 in the current NFL season. New Orleans Saints QB, Drew Brees, who was the first pick in the second round of the 2001 NFL draft, 33rd overall, is the top-rated passer among starting quarterbacks with an efficiency rating of 111.3. Oakland QB, JaMarcus Russell, whose efficiency rating is a league-low 47.7 was chosen first overall in 2007. Other No.1 overall draft picks, Peyton Manning, Eli Manning, Alex Smith, Carson Palmer and Matthew Stafford rank 5th, 12th, 16th, 18th and 31st, respectively. Three quarterbacks who were never drafted, Kurt Warner, Tony Romo and Jake Delhomme, rank 8th, 10th and 32nd.

http://en.wikipedia.org/wiki/Passer_rating

Correlation Analysis

Plot passer efficiency rating on the y-axis versus overall draft pick on the x-axis. Since the correlation coefficient, r=-0.02, is not significantly different from zero, efficiency is uncorrelated with overall draft pick. Assuming passing efficiency is a surrogate for winning, the earlier or later a QB was drafted is predictive of neither winning nor losing.

Chandra

Unlike a batter, a quarterback relies on an entire offensive team for success. This makes isolating "the realization of potential" problematic. If the line can't protect him and his WR's can't catch, he's gonna seem like a bust.

David Spero

Quarterback success is more complicated than how "good" a player is. Teams do well when they draft QBs who fit well with the offense of the team drafting them. There are many cases of QBs flopping until they got with teams that better matched their game, or who were just better than their old teams.

Pete__rock

i think the only thing they fail to consider is the relative value of the pick in light of the team they are being drafted by.. specifically, what player is going to benefit a bad team the most... to take it further, we need to look at it over time - how long will it take them to reach peak value and how long it's sustained. impossible to take it that far i would think.

Eric

GM's and owners may tend to overvalue high first round picks on the performance standpoint, but I also think there is a "financial incentive" as well. I think a good example is Reggie Bush. Going into the draft the media couldn't get enough of Reggie. He was an electrifying college player and the media was able to hype him up enough that every ESPN-watching sports fan simply presumed he was a lock NFL star. I can see why a front office would want him on their team based on performance potential, but I bet the potential (or not potential, but certain) revenue Bush brings in played a part in making him a second overall pick. If he did turn out to be an unstoppable force in the NFl like his was in college, he would have had the rare combination of world class talent AND mass market appeal (what owners would kill for pretty much.) Probably worth a shot. And if he doesn't? Well that actually played out... He still gets a good amount of coverage on ESPN, he's still a decent NFL player, he does commercials, his jerseys gets sold, and his face is still recognizable on TV. And if you ask the Saint's front office if they would go back and replace Reggie with a small market running back that is better on the field but not on commercials... I really don't know what they would say... And as a die hard sports fan, it hurts when your organization puts money over winning (see Golden State Warriors under Chris Cohan.)

D. Harrold

I think that Indinapolis Colt's president Bill Polian has come closer to perfecting NFL "market value" assessments than anyone else. Not only has he made outstanding draft picks he has also made tough "market value" decisions regarding veterans (i.e., Marshalll Faulk, Edgerrin James, Marvin Harrison, and more). Malcolm, Mr. Polian might be a really interesting interview, especially now that he's announced his retirement and his son has been named his replacement.

Clybern

http://tickalerts.blogspot.com/

Just made $600 off John Mayer's concert today.


All my info and trades open to the public. Check out my other positions.

Nick Matyas

Very informative post. I liked it. Thanks for your posting. :)

http://www.webroyalty.com

Nick Matyas

So nice and good posting. i liked it. keep going. :)

http://www.webroyalty.com

Matthew Knauff

I think there is one major aspect not being looked at: Lower draft picks tend to go to better teams. Isn't it realistic to think that a qb drafted #1 to a bad team would have a higher success rate if that same player was drafted say 28th simply because that player would be on a better football team? Chris johnson of the titans maybe the best player in football, he went in the 3rd round. but he also went to, if not the best certainly top 5, offensive lines in football. does that not have to be considered? Maybe the real value of a pick lies in the team, not the round or the pick number. And maybe, the best GMs do not pick the best player, maybe they pick the best players that fit thier system. Do the raiders have a system? or do they simply pick what al Davis thinks is the best player? could this explain why they draft so badly? could this explain why some first round picks become all-pros and some become truck drivers? The value of a draft pick is not in the round or number, it is in the Gms ability to draft the player that best fits the system, that best allows the player abilites to be showcased.

Sebastian

Interesting piece, but there is one factor that doesn't seem to be addressed. That of absolute value vs relative value. Above you pretty much discuss relative value (i pay $15m, but get $10m, therefore value is minus $5m). However the absolute value is still $10m that is given back. Surely that is better for a team looking to win, tha buying someone at $5m who gives you $7m back (relative value plus $2m).

Because with $10m of output vs $7m of output, your relative skill is plus $3m even if it is costing you more proportionally.

Teams surely want their players to be the best they can, not the most cost effective. they'd rather win with a neutral balance sheet than come 3rd with an extra $20m in the bank, no?

Jay

These arguments are kind of foolish. The fact of the matter is that the NFL is scripted with predetermined outcomes designed to capitalize on the betting dollar layed on each game, as well as enhance league profitability. Do we all really think that 23 people on the field really determine the true outcome of all the gambling dollar tied into the pending outcome. The NFL is a living monopoly with holds on multiple TV networks protecting the huge income potential of the league. TV advertising dollar is a huge source of revenue, therefore storylines and games are scripted to ensure the viewer is sitting in front of the TV for the full 4 quarters. If the superbowl can command 3 million dollars for a 30 second ad spot, then the league would obviously want viewership to be high on all nationally televised games (monday, sunday night) to increase the dollar they can demand from the advertisers. New Orleans wins a super bowl after Katrina, the "patriots" win after 9-11. The detroit lions need a natural disaster so that they can get a championship too

Bryan Gibson

The NFL owners are currently hoping to gain a 'rookie wage scale' in the upcoming negotiations with the players' union. Their goal is to limit the pay of rookies, particularly those taken early in the first round. This implies to me that the owners have an intuitive appreciation of the analysis of Massey and Thaler. They realize they are overpaying players a the top of the draft, but are currently unable to help themselves by reigning in their spending on these players. The negative publicity for failing to sign a top 5 draft pick allows agents for these players to apply a lot of pressure on the owners to capitulate and overpay.

The comments to this entry are closed.

My Photo

Bio

  • I'm a writer for the New Yorker magazine, and the author of four books, "The Tipping Point: How Little Things Make a Big Difference", "Blink: The Power of Thinking Without Thinking" and "Outliers: The Story of Success." My latest book, "What the Dog Saw" is a compilation of stories published in The New Yorker. I was born in England, and raised in southwestern Ontario in Canada. Now I live in New York City.

    My great claim to fame is that I'm from the town where they invented the BlackBerry. My family also believes (with some justification) that we are distantly related to Colin Powell. I invite you to look closely at the photograph above and draw your own conclusions.

My Website

Books

  • What the Dog Saw

    buy from amazon

    Outliers

    buy from amazon

    buy from amazon UK

    Blink

    buy from amazon

    buy from amazon UK

    Tipping Point

    buy from amazon

Recent Articles

Blog powered by Typepad