Moving on to Chapter 1 in my ongoing pedantic plodding through Malcolm Gladwell’s Outliers: The Story of Success
List of posts here: introduction, ch 1.
SPOILER ALERT: Dude, I can’t talk about the book without giving away what the book is about, so if you don’t want the book’s main ideas to be spoiled, don’t continue reading.
IDIOT ALERT: I’m in no way qualified in most of the fields Gladwell will touch on, so please, a grain of salt, before you start complaining about my ignorance. Yes I’m an idiot, please tell me why!
Chapter 1: The Mathew Effect
Chapter one begins with Gladwells laying out his main thesis. To summarize it, roughly, we overly stress “personal” qualities in explaining the highly successful and don’t put enough emphasis on “hidden advantages and extraordinary opportunities and cultural legacies.” Let’s just leave that where it is, because at this point its hard to tell what it really means.
| Those with the Ruby birth stone are sometimes blessed with awesome hockey birthdays, and sometimes doomed to hockey mediocrity. |
Gladwell then focuses on a case example, which is supposed to illustrate some of these “hidden advantages.” The example he picks is junior ice hockey in Canada. He shows us a lineup for a Western Hockey League team and asks us if we can spot the strange pattern. Now I love finding patterns (that’s in a way, part of my job) so it didn’t take me long to spot the pattern that Gladwell wants to talk about. (Of course I also found more patterns, but more on that latter!) In particular Gladwell points out how the majority of players were all born in January, February, March or April (seventeen out of twenty-five.) How can this be so? Well if you know a bit about the junior hockey system in Canada you’ll realize that this has to do with the cut-off date for hockey which is January 1, and this is dictated across the whole system. Thus someone born on January 2 will be playing along side kids who are, for the most part, younger, and thus not as well developed physically. And, because the kids who start out relatively older get streamed into the elite teams and they get an advantage that translates into there being more likely to get extra practice. Thus because of the initial cut-off establishes an relative age advantage, and the system then selects for those who are at this advantage, the chances of making it through to the upper levels if you are not born early in the year are highly skewed. Gladwell then discusses, briefly, similar effects having to do with performance in school: the relatively older kids to better.
| An angel tell’s Matthew to make the first four months the hockey months, but Matthew refuses to add this to his Gospel. |
Finally Gladwell gets to explaining how he views these results in light of this book. In particular he points to the idea of the Matthew Effect (which he attributes to sociologist Robert Merton). The naming comes from a line from the gospel of Matthew (Matthew 25:29)
For unto everyone that hath shall be given, and he shall have abundance. But from him that hath not shall be taken away even that which he hath.
(it also appears, unsurprisingly given the history of the gospels, in the gospel of Luke. Good thing the internet cheat scanners weren’t around when Luke turned in his gospel!) The idea here is that if you’re successful, then because you are successful you get advantages that lead you to be even more successful. If you aren’t successful, then you get tracked into the “not successful” track and aren’t likely to overcome being not given the advantages. (Correct me if I’m wrong, however, but my recollection was that most Christians interpret this slightly differently: that those who have faith/enact God’s will shall be given to and those who don’t have faith/don’t enact God’s will shall have their advantages taken away.) Thus Gladwell tells us:
The professional hockey player starts out a little bit better than his peers. And that little difference leads to an opportunity that makes that difference a bit bigger, and that edge in turn leads to another opportunity, which makes the initial small difference bigger still – and on and on until the hockey player is a genuine outlier. But he didn’t start out an outlier. He started out just a little better
Curmudgeonly Pedants take: Well first of all I have to say that the relative age effect doesn’t surprise me much at all. I played little league baseball growing up, and it didn’t take much observation to note that the older kids had a huge advantage and were the ones who ended up all-stars. Further I used to use this as a strategy growing up: on the playground pick the kids who had been held back a grade on your team! And, just to be clear, given that this effect exists, it seems that it would be the duty of good sportsmenship of the the Canadian hockey system to figure out a way to alleviate this effect (see this paper for proposal on how to fix this.)
So I don’t doubt the effect of the Canadian junior hockey cut-off date having the effect described: if you aren’t born in one of the first four months you are in deep trouble in this system. But there are a lot of problems with using this as an “explanation” for the outliers who end up playing professional hockey.
The first problem is that the effect Gladwell describes, while certainly there, is “small.” Wait! How can it be small: it nearly excludes something like two thirds the participants! Well look at it this way. There are around 480,000 players registered in the Hockey Canada system. Of these, around 71 per year make it to the NHL. There is something like fifteen years of hockey players in those 480,000, so if you assume there are equal numbers across these age groups (certainly this is wrong: it is more likely the distribution shrinks with age, but this would make the odds I’m about to state worse) gives you a one in 450 chance of making it to the NHL from this system. So great, you say, the effect described, explains one third of this. But this is the wrong way to think about it. Of the one in 450 chance of making the NHL, to explain just getting into the NHL you have to “left” to explain the one in 150 chance of making the NHL after having been screened for this effect. This is because explaining “outliers” is a multiplicative effect: you can’t explain an outlier by giving me an effect which narrows down the chances by such a small amount of the long odds. Sure the early effect of this relative age discrimination grows with time, but this is very different from saying that it becomes the overwhelming explanation for getting into the NHL! If you read the quote I’ve pasted above, you’ll note that Gladwell is making a huge category error: just because an effect arises from a selective process does not mean that it can be used to explain the entire one in 450 odds of becoming an outlier!
| Roy Worters is the real outlier |
And this dovetails into the second problem. It is trivial, and I mean trivial, to find and effect which is nearly as large as this effect, and has nothing to do with a “hidden advantage” or is “cultural” in nature. Indeed it should have jumped out at you when you read the list of hockey players that Gladwell presented in the book. All but two (or four, see below) of the twenty five players listed in Gladwell’s list are above average height for their age! The average height for Canadian boys of age 19 is 174.6 centimeters which is about 5’8.7″. Two of the players are listed at 5’8″ and two are listed at 5’9″, so lets just be generous and say that all but four are above average height. Take a look at a random NHL team and you’ll see that there is some pretty strong selection going on in heights: almost all are above average height (but not too tall.) This, then, is roughly an effect which can “explain” about one half of your kids chance of making the NHL. If the kid is going to be short, then, well you’d better give up! In other words, it’s pretty damn easy to find an effect which is pretty close to the opposite of the effect Gladwell wants to emphasize, one based on “nature”, which is nearly as large as the effect he describes.
Again, just to reemphasize, I am not doubting the relative age effect in the hockey system. It’s there. But it is a small effect and there are easily other effects, like the eventual height of the child, which has a large effect and is, of the essence very different from the one Gladwell emphasizes. So I’m left with just one interesting point of the chapter: that the relative age effect exists. But should we be surprised by this? I’ll leave that up to you, but for me I’d say no (maybe because I’ve seen similar effects before: for example alphabetical ordering of names on scientific papers appears to influence your employment and tenure chances!)
| Full of woe: to be born on the wrong day! |
Improv’s take: Suppose we lived in a world where your future occupation was directly influenced by the date of your birth. Oh boy would that lead to a lot of scheduled sex! In such a world we’d probably have holidays around the major desired occupations. Today it’s “Doctor Day”, everyone go home and make whoopie! And boy the stress this would put on hospitals around those dates. Not to mention the parents who want to induce the pregnancy on a given day! Or the father mad at the wife for not being able to make it to the desired date. (For fun such a system should schedule the jobs no one wants around the ones everyone wants.) I think in such a world I’d make my living selling tables which calculate your expected salary if you procreate on a given day.
thenks-
If your summary above is correct, then I think the example indeed supports his argument. No, we shouldn’t be surprised by the cutoff date effect, nor does it have to explain the whole thing.
The final measurement is very precise (70 out of 480k), and (most) people think that it’s equally accurate. But, it is not. I think that’s all that he’s saying.
disclaimer: I haven’t read the book.
I had this weird feeling while reading the book that he was sometimes very close to contradicting his thesis. And its absurd to suggest that there could have been (or are today) millions of people with the potential of Bill Gates, who, if only they had been given the opportunity to spend 10,000 hours in front of a computer (with no WoW) could have transformed the world.
Only a few people can revolutionize the world at any given time. It is not totally random who those people are. Outliers is an attempt to make broad policy arguments that support providing more opportunities for everyone, because as Gladwell describes, opportunities are the difference. (He makes this point in a more sophisticated way, I think, in The Tipping Point when he discusses poverty as being cut off from contacts and opportunities).
I’m not against providing opportunities to everyone. But I think that if one wants America to compete in the global economy then the policy should be to put more supply side controls on opportunities such as advanced training or small business loans, level the playing field by eliminating regressive economic policies, providing safety and security as well as three meals a day, hike up the inheritance tax so that legacies are not inherited, really excellently compensate the people who work their tails off. Build up the idea that through hard work is born transformational change.
Blah!
Tipping Point > Blink >> Outliers
A practical remark: Professional surfers are considerably shorter and more slender than average folks, so there’s a good answer for you slender Canadian boys and girls … move to Vancouver Island (or Newfoundland) and take up surfing! (I could tell you where to surf, but then I’d have to kill you)
An ethical remark: focusing on “the best of the best” is a tendency that is deeply embedded in human cognition … that makes zero ethical or economic sense!
Wouldn’t it be a whole lot better to study extraordinary achievements by average people? Not too many people study this, but we do have Dirac’s “A Golden Era is one in which ordinary people can make extraordinary contributions.”
If we look for Golden Communities, then there are a surprising number of them … family farms … non-championship high-school team sports … non-professional surfing … and also (my own great interest) system engineering.
Getting to the moon, via the Apollo Project, was pretty cool. And it was achieved by the extraordinary efforts of (mainly) ordinary folks. System engineering provided both the technical context and (equally important) the social context, which catalyzed the Golden Achievement of going to the moon.
In its broadest sense, system engineering is “The design of the whole, as contrasted with the design of the parts” (Ramo). This is not achieved by cherry-picking four-sigma outliers, but by making best use of the centroid of the distribution.
Marc:
Yes I’m being very pedantic, and I have only read the first chapter, so I can’t even comment on whether he has achieved his main thesis, or whether he should be forgiven for his current sin. But at this point, he’s done what I deem a classic snow job: only pointing out the effect he thinks is important and not quantifying it compared to other effects. I think we both agree the effect is important, but without context of what an effect of this size means, he’s being a bit dishonest at this point.
Or, more probably, I’m just reading too much into his title. Explaining a one in three effect for something that requires much longer odds doesn’t really help me understand who outliers are and why they get where they are. In other words, does the effect describe have much to do with actual outliers?
While I’m not a huge fan of Outliers, your critique here does seem to live up to your description as “pedantic.” You don’t really make an attempt to vew the chapter in the context of Gladwell’s overall objective, which is to explore what factors create “outliers” – not just successes, but _extreme_ successes. Are they born, or are they “cultivated”?
First: You say its trivial “to find and [sic] effect which is nearly as large as this effect, and has nothing to do with a ‘hidden advantage’ or is ‘cultural'” in nature.” I assume you mean “trivial” in the sense of “very easy”, but I think Gladwell would agree with you that finding other factors, such as size, that influence hockey success is indeed trivial in the sense of “not being particularly interesting or relevant” to his purpose. I don’t have the book in front of me, but I do recall him explaining (explicitly or implicitly) that having the basic toolset suited to a given endeavor is of course a primary determinant for success – being born in Canada, athletic, larger than average, etc. will explain why hockey players make success; you can pretty much assume that by late teens most competitive hockey players are above average size and strength.
But Gladwell is interested in exogenous factors that, across disciplines, make the best-of-the-best stand out from the rest, and here the Relative Age Effect does have significant explanatory power – one that holds true for hockey, tennis, acamdemic performance, social leadership, etc., etc. It’s real, it’s interesting, and it’s exogenous, and therefore potentially controllable or exploitable.
Second: Gladwell doesn’t claim that the Relative Age Effect it the only, or even primary, explanation for what creates “outliers”. It’s one factor that combines and interacts with the phenomena described in subsequent chapters. In fact, when you read the next chapter on the “10,000 hour rule”, I think you’ll be amazed that the Relative Age Effect has as pronounced an impact on hockey success as it does.
Dude I’m not going to stop reading just because I disagree with the book! Sorry, I read books because they tell me things I don’t know, not because they reinforce my thinking. Sheesh, what’s up with that comment, that’s like telling me to remain stupid.
Two points:
1.
“He’s trying to find out what exogenous factors – in general, across fields – separate _extreme_ successes from the rest of the merely “very compentent” actors within any given field.”
But his hockey example does nothing of this sort, right? Tons of very competent actors have exactly the relative age advantage, but they aren’t extreme successes. Leaving out all the 149 out of 150 seems to me to be a “snow job” if that is his goal.
“Gladwell says – lets take all those innate intellectual/physical attributes that influence aptitude for a field as a given (i.e., let’s stipulate that every field has its own general set of “suggested minimum hardware requirements”). Now, assuming everyone meets the common bar, what determines who becomes an extreme success?”
But he doesn’t, at least in the two chapters I’ve read (dude, that’s one hand behind my back) make any such analysis. He simply says: look here is a component that most people don’t notice because their dense. Wah, lah, it must explain the extra amount needed to become an _extreme_ success! And further, why should this be “the effect” which is needed. What if, for example, an innate physical ability has longer odds? Isn’t this them “more important?” So far (and again, I’m only on Chapter 1, so I’m willing to be swayed) he hasn’t made any such rigorous comparison. That’s called spin, in my book.
2. Again, I don’t understand why you claim the hockey example is such an extreme effect. As I explained above it leaves a considerable amount to be left explained. Is it a component: yes. I didn’t argue it wasn’t. Is it a big deal: yes if your a hockey dad or hockey mom and want to flatten the playing field. But does it explain the one in 450 odds? Not by a long shot. He has explained why, when drawing two cards from a deck, the suits are the same color, but not why they were the ace of spaces and the ace of clubs (okay that’s a bit off on the odds, but it’s in the ballpark.)
“Snow job” is a bit of a cheap shot! The problem I have with your critique is that you assume Gladwell is trying to determine what makes someone a success at something specific (like hockey) or, well, anything. He’s not. He’s trying to find out what exogenous factors – in general, across fields – separate _extreme_ successes from the rest of the merely “very compentent” actors within any given field. Arguing that Gladwell hasn’t found the factors that most influence whether any random individual in Canada becomes succesful at hockey is off target, because that’s not his goal. (And it would have made for an extremely uninteresting analysis relevant only to hockey coaches, or totalitarian regimes interested in increasing their Winter Olympics medal counts.)
Again, it’s obvious (and not very generalizable or interesting) that success in different fields depends to large degree on possessing certain natural characteristics. Height for basketball, IQ for computer science, being shorter than average for gymanstics, etc. For Gladwell to meet your criteria, he would have had to write a chapter quantifying the relative impact of variables such as upper body strength, lower body strenght, 40yd dash speed, lung capacity. Relative age becomes a minor variable in whether any given person becomes a hockey player, but that’s a no duh.
Gladwell says – lets take all those innate intellectual/physical attributes that influence aptitude for a field as a given (i.e., let’s stipulate that every field has its own general set of “suggested minimum hardware requirements”). Now, assuming everyone meets the common bar, what determines who becomes an extreme success? The conventional wisdom is that prodigies are just at the extreme ends of these variables – Michael jordan was born with freakish hand/eye coordination, Einstein’s genes had a mutation that coded for execessive glial cell production, etc. (= most ram, fastest cpu in the hardware analogy).
But Gladwell’s thesis is that this conventional wisdom is wrong. When you look at a chart like the one here (http://www.socialproblemindex.ualberta.ca/relage.htm#Elite) which shows that 40% of elite hockey players were born in the first three months of the year, and 10% born in the last 3 months of the year – it becomes clear that it’s NOT all about innate abilities; a huge component of extreme success (and, probably, relative success in a field as well) is determined by controllable environmental factors. That’s a fascinating discovery that has real world applicability (either at a policy level or at an individual level – e.g., whether parents should “red shirt” their child in kindergarten). I think he starts the book off with the relative age effect not because it is the most important such factor (that’s actually covered in Chapter 2 – the 10,000 hour rule), but because showing the huge impact of a seemingly arbitrary variable such as birth month on who becomes an “outlier” clearly undermines the conventional wisdom, priming us for the rest of the book.
The remaining chapters explore (with varying degrees of success/relevance., IMO) other controllable and non-obvious factors that separate the “best of the best”. If you don’t think that question is interesting, and require a soup-to-nuts explanation of why any given outlier is so much better than _the average person_ (vs. so much better than the average person _in a given field_), you should probably stop reading (well, after trying chapter 2).
What will we accomplish by identifying these unanticipated factors? Gladwell doesn’t address this issue at all. I think he wants to remove them, but then what? The elite circles are elite because they have seemingly arbitrary barriers to entry and are by design limited in size. By removing unexpected factors, I don’t think that the size of the elite pool changes at all. All that happens is that it becomes more arbitrary to enter.
Maybe its more fair at the early stages, but life isn’t fair. Is it reasonable to lead an increased number of people down the path toward the elite circle when you know that only 1 in whatever large number (say 10,000) will be successful?
Of course maybe my opinion is colored by the rejection letters I’ve gotten saying there were “almost 700 applicants” for the positions I am apply to. That means that even if you are in the top 1% of applicants you slightly over a 50% chance of getting invited to interview.
Ok, now I’m going to switch gears.
I also agree with Dave’s point that you will always be able to identify factors that correlate with extreme success. Correlation does not imply causality. For example, I couldn’t find whether anyone had looked at whether the C150T mutation in mitochondrial DNA that is significantly over-represented in Italian centenarians (Attardi et al Nature 2003) is also present in the Rosetian population? Perhaps genetic predisposition to extended lifespan leads to close knit community structures.
Someone brought up Bill Gates and 10,000 hours. Why would Bill Gates need 10,000 hours to resell someone else’s operating system. Bill Gates’s talent was being in the community that learned about IBM’s search for an operating system and recognizing how important and valuable IBM’s operating system monopoly could be. Yes, he also had to keep a straight face when he bought that first OS, but he didn’t need any special programming skill.
John Sidles said,
” Professional surfers are considerably shorter and more slender than average folks, so there’s a good answer for you slender Canadian boys and girls … move to Vancouver Island (or Newfoundland) and take up surfing! (I could tell you where to surf, but then I’d have to kill you)”
I’m 6’1″, 52 years old, and rode my first wave in 1963. I’m a decent surfer, despite my old age and height. I still surf every swell because I never stopped, never got complacent, and surfing is still the best thing in my life.
When professional surfers show up at our very localized break, they get beaten up and shoved to the parking lot. Unless it were Gerry Lopez, who’s an acquaintance, I still have no use for pros. Hell, Gerry would get a serious case of stink eye at my break.
Jeff
Jeff Watson (aka Masteroftheuniverse) says: “I’m 6’1″, 52 years old, and I still surf.”
Hmmmm … localized break … intolerant of pros … Gerry Lopez … Cro-Magnon affect … my guess is, you’re a S**s*d* local! Heck … I’m 6’3″, 57 years old, and I still surf. And at 212 lbs (on a good day), it’s a pretty safe bet that I’m fatter than you too! 🙂
I could mention the break where Seattle’s quantum-surfers paddle out … but gosh, the younger PNW locals play mighty rough … they’ve been known to wax a rent-a-car windshield … with a cheeseburger … and let the local bears do the rest. 🙂
If there’s an older, heavier surfer than me out there … who also is into quantum information theory … let’s hear from you! 🙂
And Dave Bacon … please … let’s hear some of your promised crazy ideas! … `cuz the QIT blogosphere has been just too gloomy recently!
http://www.merriam-webster.com/dictionary/voila