Jen on Chess Outliers and Unlucky Birthdays
By Jennifer Shahade   
December 10, 2008
Outliers210.jpgI was born on New Year's Eve, the biggest party night of the year aka "amateur's night." December 31 is also the worst birthday for a young chessplayer, and to a larger extent, an aspiring Canadian ice hockey professional. When I was 14, I started playing in World Youth events and was miffed that my eligibility was determined by my age as of January 1, making me a year older in chess age than I would have been if I was born one day later. So I was immediately drawn in by personal experience to the "Matthew Effect", the first chapter of best-selling author Malcolm Gladwell's third book, Outliers: The Story of Success. Outliers is great brain candy, and in this review I'll pick out treats relevant to chess.

The "Matthew Effect", refers to how an age cut-off in sports creates a glut of athletes who are born just after the cut-off. It's named after the gospel of Matthew in the New Testament: "For unto everyone that hath shall be given, and he shall have abundance. But from him that hath not shall be taken away." In Canadian ice hockey, which Gladwell focuses on, the effect is particularly extreme, because all young Canadian boys are funneled into a training system in kindergarten or even earlier, ages at which 9-11 months will likely make a big difference in weight and height. The most skilled undergo a series of grueling training sessions. In this sport, January, February and March birthdays dominate even professional league rosters, with November and December kids under-represented. Gladwell explains that the bigger January boys will be more likely to be chosen for an intensive training program, the benefits of which will extend even when boys born later in the year will have caught up in size. Gladwell doesn't examine gymnastics, which prizes smallness and flexibility, but I'd imagine that sport would show the opposite effect.

Is there a Matthew Effect in Chess?

Upon examining the birthdays of the majority of the players at the 2008 World Youth Championship in Vietnam,  I also found evidence of the Matthew effect. The data (715 birthdays including 28 Americans) was tapered in this admittedly moderate sized sample with the most frequent birthmonths occurring in the beginning of the year and the less frequent at the end. The median birth month was in May (you'd expect it to be around July 1.) January was the most common month to be born and December was the least likely. Here is the breakdown:

Birth Month    Number of World Youth 2008 participants

January-        99
February-      61
March-           82
April-             65
May-              58
June-             58
July-              60
August  
         49
September-   42
October-        60
November-    43
December-    38

Unsurprisingly, the effect was most pronounced at the Under Age 8 category (where late April was the median birthday) but in four out of the six age categories, January was the most likely month to be born. The other two were February (Under 16) and March (Under 12.)

Taller and bigger chessplayers don't have any special edge, but maturity helps in chess, especially in the younger age categories. In my experience teaching chess, it's much easier to teach an average fifth grader a complicated chess concept than a third grader. 

The effect is probably more likely to show up in countries with intensive training programs. The average and median birthdays for the 28-player American delegation to the World Youth Championships in Vietnam was actually in the second half of the year. There was even one U.S. player with my birthday, December 31. Yay for the calendar-challenged!

The statistically inclined may wonder whether babies are born more in certain months, skewing the stats. I found a lot of data online about birth month frequency (mostly related to the likelihood of contracting various diseases), including this chart for America and one for Pakistan (pdf) , but nothing I read seems to invalidate the Matthew Effect findings.   The Freakonomics New York Times blog, the first mainstream venue where I read about this, considered a conspiracy theory in a follow-up article:

Some other readers have offered a clever, very Freakonomics-y alternative explanation for these age patterns: the parents are lying about their child’s birthday. If the parents want the kid to be a star, they take an older kid and change his date of birth to make him eligible to play with younger children.


Even if birthday cheating happened occasionally, it would doubly skew the stats. Kids would be added to January months and taken away from November and December. 

10,000 hours of chess


Chapter two in Gladwell's book, "the 10,000 hour rule" refers to chess explicitly a few times.  10,000-hours is the minimum amount of deliberate practice that Gladwell and other researchers think is required to become world-class in anything. Gladwell mostly discusses music and computer programming in this chapter, but on page 41, he stumbles trying to stretch his argument to chess, "To become a chess grandmaster also seems to take about ten years. (Only the legendary Bobby Fischer got to that elite level in less that amount of time: it took him nine years.)" Obviously the kids who made GM at 13 or 14 were not all studying chess systematically from the age of 3 or 4, so this is just wrong. So it could take less than 10 years to become a GM, but Jonathan Hilton has a beef with another number; he thinks 10,000 hours is a deflated starting point for chess:

I’m pretty sure it takes far more than 10,000 hours to become a GM. If Bobby Fischer spent just three hours a day on chess, that’s roughly 1,000 hours a year, which would make one a GM in ten years at an average of three hours a day. Perhaps it’s possible (it seems reasonable that if those were 3 quality hours, one could certainly do it over ten years) so long as the 10,000 hours doesn’t include time actually playing in tournaments. But from talking (to many major talents) including Caruana and Ludwig… they’ll often spend entire days, 10-14 hours, working on the game. Professor Ken Kierwa estimated it took 8,000 hours for a child prodigy to become a master-level player. 


 Despite such qualms, "10,000 hour rule" is fascinating reading.   Gladwell doesn't think that 10,000 hours or hard work alone is enough to become very successful, but that after a certain threshold of intelligence or talent in a specific area, it's by far the most important factor. This may sound obvious, but it's actually counter-intuitive to me. I'd think that talent would matter more as you go up the ladder, and that hard work would matter more in the middle. Gladwell is saying that talent matters less the higher up you go. To put it in chess terms, a talented chessplayer would do a better than a hard-worker with average talent but a talented chess player who works really hard will surpass a less diligent super-talent.

Emil Sutovsky, a 2650+ Grandmaster and a trained opera singer agreed with this theory. ( Emil may be most known to CLO readers for training and managing Gata Kamsky, and securing the details for the upcoming Sofia match.) 
     
I absolutely agree, that talent becomes LESS important the higher you go up the ladder, as HARD WORK becomes more important. I am clearly less professional in music (than in chess), but having some knowledge, I would say in singing , talent plays a key role. You might be barely capable of learning the words in your aria, but your ingenious voice would win hearts. But playing piano or violin is absolutely different. As in chess - you MUST work really hard in order to get to the very top. 


Most of Gladwell's examples in this chapter are young devotees. Judging from Gladwell's recent New Yorker article on late bloomers it's far from clear that he'd disqualify adults from the potential to reach mastery after logging 10,000+ hours of their own. Wishful thinking, many may say, but  writing does not just represent reality but also influences it. If Malcolm writes a whole book on Late Bloomers, and the depression deepens, maybe in three years, we'll see a litany of unemployed adult chess prodigies at the World Open.

The 10,000-hour rule is informed by another pop psychology/economics book, Talent is Overrated , which expands on the idea of deliberate practice. Author Geoff Colvin shows how businesses can use the lessons of extraordinary achievement in sports, chess and music to improve employee performance.  The "chess mode" of improvement involves challenging workers and oneself to case studies, in the way that chessplayers often to pose themselves questions like "White to Move and Win" or "What did Kasparov do in this position?" All disciplines are not equally suited for this type of problem. Can you imagine a writing book that was filled with puzzles such as "What word did Hemmingway write next?"

Talented is Overrated also classifies key aspects of deliberate practice:  it is designed specifically to improve, it can be repeated, it's highly demanding mentally and it isn't much fun. Colvin writes: "The reality that deliberate practice is hard can even be seen as good news. It means that most people won't do it. So your willingness...will distinguish you even more."
 
Gladwell, on the other hand, espouses the importance of fun in deliberate practice. Bill Joy, co-founder of Sun Microsystems, says that sharing time on computers instead of waiting for others to finish was like, "the difference between playing chess by mail and speed chess." Gladwell goes on: "Programming wasn't an exercise in frustration anymore. It was fun." It's fitting that chess was mentioned in this section, as I'm sure many top GMs have honed their skills through blitz chess. So deliberate practice should be fun (according to Gladwell) but also not fun (according to Colvin). I think there is word for this: passionate.  
 
Renaming the Bishop


In Chapter 6, "Rice Paddies and Math Tests", Gladwell tries to justify the stereotype that "Asians are good at math." Gladwell thinks it's all rooted in language; in Chinese languages, the construction of numbers is more intuitive and length of syllables is shorter than in English. Gladwell cites research by Stanislas Dehaene, author of The Number Sense

In languages as diverse as Welsh, Arabic, Chinese, English and Hebrew, there is a reproducible correlation between the time required to pronounce numbers and the memory span of its speaker... in this domain the prize for efficacy goes to...Cantonese...a dialect of Chinese.


Cantonese speakers can memorize 10 numbers in a 20 second span while Americans can memorize an average of 7. Indeed, "seven" is a definite killjoy in my attempts to memorize phone numbers.

So maybe we'd all be faster at chess calculation if the pieces were all one syllable. Any ideas on how to rename the bishop? If we can get through this hurdle, we'll be on even more on our way to passing Russians, who waste a whooping three syllables on the pawn (peeyeshka). If you're a dork like me, you'll probably want to scan this table of chess terms in 64 languages for syllabic efficiency. 

Gladwell argues that because of the more intuitive numeral system, Chinese students might be more confident in chess as they grow up, and therefore more likely to enjoy math and by extension, math-like pursuits such as chess calculation (I'm personally not sure how much chess and math are alike, but for the sake of this argument, let's assume they're related). He also talks about the difference between rice cultivation, which can be done year around and other types of more seasonal farming. Gladwell harps on an old proverb to explain the long lasting nature of the Chinese work ethic: "No one who can rise before dawn 360 days a year fails to make his family rich." This goes back to the 10,000 hour rule, and the obvious equation around which many fine points can be made, that more time=more success.

Slow Learning


Gladwell also writes about the determination to understand concepts inside-out in this chapter. A professor at Berkeley University, Alan Schoenfeld makes videos of his students puzzling out the details of math problems like Renee, who spent 22 minutes on an algebra problem. She may have seemed slow to some, but Schoenfeld was impressed, "If I put the average eighth grader in the same position as Renee... they would have said 'I need you to explain it.' There's a will to make sense that drives what she does." At the risk of being a naysayer on this sweet story, absorbing material without being overly inquisitive is something that seems to help kids learn quickly. But there's a time for deep understanding too. When I tried to learn the Open Sicilian, it helped me to write out the logic behind all of Black's major alternatives. I was reminded of the slow learning section in Outliers when going over Sam Shankland's games from Vietnam for Interview with an American Medallist. Sam analyzed these games for Chess Life Online, but because he is such a strong player, I went through the analysis and picked out points that required more explanation for the average CLO reader to understand. In the final round battle that clinched his bronze medal, Sam gave Rb1 two exclamation points.

50...Kf5G11.jpg
Position after 50...Kf5 in Shankland-Liem


Even with a few example variations, it took me time and words to grasp why it deserved two exclams. Here is what my thought process sounded like:
   
White threatens b8=Q because after Rook takes queen, Black can queen but White will also queen and White can easily run his king up the board to escape a perpetual. After Rb1, Black can blockade the pawn with Rb8, but now when White plays Rd1, Black is forced to play Rd8. White has gotten the same position he had a few moves ago, but now his Rook is on d1 for "free" So he plays Kc3 and after now ke6 Kc4 Rd7 b8+Q Rxd1 Qe8+ wins for White. White could have started with Kc3, but then Ke6 Kc4 Kd6. So protecting the d7 pawn is not really relevant to this variation. That's just an illusion since the d7 pawn is already protected tactically because of b8=Q. The point of getting the R to d1 for free is actually to cut the Black king off from coming to d6-c7.


So anyway, congratulations if you got through all that, I guess there is a reason chess books aren't written in stream of consciousness style. But my point is similar to the math professor's: We shouldn't be ashamed of asking obvious questions and putting variations into words when trying to understand endgames and openings. Sam's endgame advice to CLO readers is telling, " When you play a game that reaches an endgame, take it very seriously. After the game, COMPLETELY TAKE THAT ENDGAME APART (caps are Sam's) until you know it like the back of your hand."  

Inspiring or Obvious?


After finishing Gladwell's book, I thought about the most successful people I know and looked for patterns between them, even though they range from entrepreneurs to champions to the very rich. One thing they all have in common is a personal and intense relationship with their free time. I'm not saying they work all the time or that they insist on setting their own schedules, but they give off the sense that their time is theirs alone and you really shouldn't screw with it.  This is a chicken/egg situation- the very successful gain a sense of importance, often don't have bosses, or are the boss and can set their  schedules.  So you can boil this down to a cliché, that success=hardwork=time in the same way a negative Outliers review, "It's True: Success Succeeds and Advantages Can Help " criticized Gladwell for being too obvious: "Much of what Mr. Gladwell has to say about superstars is little more than common sense: that talent alone is not enough to ensure success, that opportunity, hard work, timing and luck play important roles as well." I agree with some of this, but I don't see it in such a negative way. Gladwell inspires us to unpack our own convictions, and if I am also guilty of glazing the obvious in this review, then I  consider reading Outliers time well spent.  

It's that time of year again: Not only for my unlucky birthday but also for Best of CLO 2008. Please send your nominations for Best Article of the Year to [email protected].