Are Elite Athletes Marrying Elite Athletes?
A conversation with David Epstein, author of The Sports Gene
Posted Sep 03, 2013
His latest book, The Sports Gene, is being widely discussed, and for good reason. We are all fascinated by what makes some athletes great, and David explains through just the right balance of stories and science how these elite people are simply extraordinary in so many ways.
After reading David’s eye opening book, I had many questions for him because I also study individual differences, have played sports much of my life, and understand that talent in so many different domains is relative. Even among the very best, some are just that much better. But why?
I sent him an email asking for an interview, and to my delight, he said he would be happy to talk. We ended up chatting for well over an hour about wide ranging topics, including the ones in this interview. He holds graduate degrees in journalism and environmental science from Columbia University, so he is obviously extremely smart and educated. What I did not realize until I got him on the phone is how kind and down to earth he is, considering he is such an accomplished writer and author.
I had read, in bits, and pieces some individual studies from the sports literature. However, academics tend to study very narrow topics so it is hard to see the big picture regarding a field without a clear synthesis. Review articles in academic journals are written for that purpose, but unfortunately, academic writing is often not very engaging or easy to understand, and rarely do academics read work outside their own discipline. This is why The Sports Gene is so important. You are being entertained as David teaches you all about the science of extraordinary performance. You could be an academic in a different (yet related) field like me, or a smart person who knows nothing about the topic, and David would teach you through his storytelling and science writing. I sincerely believe this synthetic work will be drawn upon by both writers and academics for many years to come.
There are eight main sections to this interview:
1. The Only Real Rule Is Tremendous Natural Range
2. Are Elite Athletes Marrying Elite Athletes?
3. “You Can’t Train Speed”—Slow Kids Never Make Fast Adults
4. What Is The Likelihood Of Becoming An Elite Athlete?
5. The Worldwide Search for Elite Athletic Talent
6. Could Motivation and Grit Actually Be, In Part, Genetic?
7. Do We Rely Too Much On Our Personal Narratives?
8. Do The Extraordinary Think They Are Ordinary?
1. The Only Real Rule Is Tremendous Natural Range
JON: You write that among athletes “the only real rule is that there is tremendous natural range.” Could you provide some examples from the people you interviewed?
That fallacy shows up in a lot of sports. You’ll see this at the NFL combine for example, where physical traits are measured that are separate from the ability to play football which are things like strength and speed. And there will be some guy who gets drafted in the 4th round instead of the 1st and eventually turns out to be better than guys in the 1st round. And so inevitably every year the story will be, well it’s all these intangibles, it’s all these things that you just couldn’t measure, but really first of all, the combine measures are stupid, so it’s not that you can’t measure those things, it’s that you’re measuring the wrong things, and second, if you took every player in high school and college they would be outstanding predictors, but because you’ve restricted it to the top 20 guys in the country, then all of a sudden it’s a lot more difficult to have real substantive predictive value. To me it’s just this total fallacy. They’re saying “well you just can’t measure these things.” You sure could if you put all of humanity against those measures—they would be amazing predictors. I know Malcolm Gladwell wrote an article in response to me the other day and it seemed to me that the brunt of his argument was something along the lines of “once we prescreen the population, then practice matters a lot and more practice is better than less practice” and I wholeheartedly agree with that. He seems to be saying, in part, that “The rule that I called a magic number is neither a rule nor a magic number.”
That’s a really interesting question. I think it probably depends on the sport but happens quite often. I come from an endurance sports background where there is quite a bit to measure like oxygen carrying capacity and I’m still of the opinion—I’ve seen how these tests are conducted—and I think in the labs where they test both normal people and elite athletes, they aren’t actually getting the elite athletes to the point where they’ve maxed out. Some of the physiologists who are used to working with elite athletes can do a better job of that but that’s not most labs and I’ve looked at some of the data and you can tell that they haven’t leveled off yet. So I think that’s probably quite common.
I think there are some measures where elite athletes are compared to normal people and there are variables that aren’t considered. So, at the NFL combine again, bench press is maybe the second most important measurement that occurs there, and in bench press guys with short arms have a huge advantage. But that’s a trait that you don’t want for the actual football field. The guys who actually play better probably tend to have longer arms. So I think their actual functional strength advantage is even more disparate from normal people than they appear in the measure.
You note that Yao Ming’s parents were gargantuan themselves and brought together by the Chinese basketball federation. Do you think this practice—of having people with extraordinary traits team up—is common based on your interviews and research?
No I don’t think it’s common. But I do think it is becoming more common by accident because now there are many more female elite athletes than there ever were in the past and there are more professional athlete couples. So I think Yao Ming’s example is pretty extreme—although it certainly worked—but I still expect regression to the mean. You wouldn’t expect Andre Agassi and Steffi Graf’s progeny to just miraculously walk out onto the tennis court and be professional tennis players, but I do think we will see an increasing number of kids of professional athlete parents become professionals themselves. There just weren’t many professional athlete couples in the past because there just weren’t that many female professional athletes. The women have a lot better opportunity now to find out if they have athletic talent now than they ever did in the past. In some of the studies that look at aerobic capacity, there was some evidence of assortative mating, even just among normal people. So people with higher aerobic capacity would tend to be with each other and maybe that’s because they like outdoor activities or they exercise or something like that. So I think there’s some evidence of that even among normal people. I can tell you from my college track team that there were multiple couples of people who ended up getting married. And Sheldon Williams and Candice Parker just got married. [David later sent me the article “Ranking The Most Athletic Couples In Sports” which shows there are many out there now.]
A mantra of football coaches is that “You can’t train speed.” In fact, Justin Durandt, manager of the Discovery High Performance Centre at the Sports Science Institute of South Africa, is in the business of testing for speed. He pointed out that “We’ve tested over ten thousand boys, and I’ve never seen a boy who was slow become fast.” You note that “slow kids never make fast adults.” What types of things, like speed, did you find simply can’t be trained as much as we would like?
That’s interesting. To add to that I just saw a study today that followed Oklahoma State football players over four years of strength training in college football, and they improved their strength immensely in the weight room, so they got much better at what they were doing, but they didn’t improve in running speed at all. So either that weight training is just disconnected from the performance benefits they want, or it’s really important to recruit guys who are fast in the first place. There were really important similar traits that didn’t change. So I talk about the Achilles tendon which is really important for everything explosive. Ideally you want an Achilles tendon that is both long and stiff, and stiffness you actually can change to a degree through training. But the length is purely a function of the distance between your calf and your heel bone, and that’s something you’re born with and you can’t change. By all accounts, visual acuity, which I write about with baseball players, can change but your maximum visual acuity seems to be set because it’s based on the density of cones in your macula, and that’s like the megapixel rating of a camera, and doesn’t seem to be changeable. There are actually all these vision systems that are marketed for baseball players to improve their visual acuity, depth perception, and things like that, and I actually cut out a little part where I assessed those in the book, but the vast majority of them have been utter failures and have shown no difference. There are some visual skills like contrast sensitivity that some of these systems showed they can improve, but for most of the visual traits it’s just total nonsense, it’s just marketing and they fail every single test. But they’re still quite popular. There were things from height to things as micro body type as brachial index, which is the length of your forearm relative to your total arm. And we know that bone length can change a little bit, that was a surprise to me. That for kids who grew up playing tennis from the time when they were really young, bone does respond to stresses the way that muscle does, sort of. And yet, it’s amazing, but the limits are pretty distinct and quite narrow. So more bone can be built to support more muscle, your arm can lengthen a little, but it’s not a whole lot.
And the same is true for the weight of your skeleton. I loved the analogy that Francis Holway gave about how important your skeleton is to the amount of muscle that you can ultimately hold. There, he likens it to a bookcase, where he says, well, a professional shot putter’s skeleton might only be 6 pounds heavier than a normal person’s, but because each pound of bone can support a maximum of 5 pounds of muscle, that’s a difference of 30 pounds of muscle. So that’s like a bookcase in that when a bookcase is a little bit wider it’s not much heavier when it’s empty, but when you fill it with books it’s way heavier. So there are all these sorts of traits that ultimately set constraints on where you’ll get with your training. Even if they’re not sort of constraints to begin with because you have a lot of room to improve, eventually they become constraints. But athletes who use steroids are able to surmount that, at least temporarily.
The winner-take-all effect along with the global marketplace has essentially altered the gene pool within elite sports. You write about the bivariate overlap zone (BOZ) which gives the “probability that a person randomly selected from the general public has a physique that could possibly fit into a given sport at the elite level” and that “the BOZ for most sports has decreased profoundly” over the years. For example:
Professional soccer players: 28% of men have the required height and weight combination
Elite sprinters: 23%
Professional hockey players: 15%
Rugby Union forwards: 9.5%
Regional catwalk models: < 8%
International models: 5%
This basically means that most people don’t have the appropriate height/weight/body type combination to even have a chance of becoming an elite athlete, is that correct?
That’s true. And I go on to write that we can change our weight to a degree but it’s within constraints. But for some of those things, like height, it is what it is, and if you don’t have it you don’t have it. And the more competitive sports have become—the wider the global talent search—the more specific the body types have gotten, so there have been more people ruled out from the get go.
Why do you think body types have been getting more specific?
In the first half of the 20th century there was this idea that the average body type was the best for multiple endeavors. But this wasn’t really based on any legitimate science, and partly had these racial agendas, and that idea eventually went by the wayside. As sports scientists realized that was bunk and the rewards for sports became much larger, more people wanted to participate, and a self-filtering system brought these body types into sports. So with water polo players, where their forearm length has grown proportional their arm, there was no indication as far as I saw that anyone was proactively selecting for that. But as the sport has grown more competitive, those are the people who have risen to the top naturally. As sports have gotten more competitive, both biology and training have become more restrictive to who can make it. You have to have better access to training and better training than ever before, and you need to be more talented to begin with.
You also note that there is such a thing as a “naturally fit fraternity” where approximately 6 out of 1,900 men are part of this. Could you explain what this means?
So that came from a study at York University where men who had no training whatsoever were tested for their aerobic capacity which is the amount of oxygen they can use when they are running as hard as they can. And that’s a really powerful predictor of your endurance, although not the only one. Even men who had a light training history were screened out of this population so they were left with people with no training history whatsoever. And 6 of those 1,900 men who were screened had about 50% higher aerobic capacity or VO2 max as runners call it. Those guys had aerobic capacities in line basically with college runners, and they hadn’t trained at all. And despite being a little higher body weight than you would want, that’s pretty amazing. I was floored by that because those guys were where I was in aerobic capacity after I had been training at the Division I college level, without having done anything and not knowing at all about themselves. Even with only 6 out of 1,900, that’s really not that rare for someone to have a trait that amazing. And maybe intuitively we know that there seem to be some people who are in better shape without much training, but if you had asked me before if I thought there were people with VO2 max of 66 with no training whatsoever, I would have guessed no, that nobody had that. But some people very clearly do, and not only were they found in this study at York University, but now when I’ve been looking through data on elite athletes when they are untrained, they appear as if they start around the level of those naturally fit people, basically, and then just improve upon that. So considering the number of people at the level of elite athletes, 6 out of 1,900 is not that rare, if you think about it.
Part of it could be that we go through life only seeing a certain sliver of people, and if we could see the entire population only then could we fully understand the variance that’s out there.
Yes. So there were five books that took the all nuture approach, or at least minimized the role of genes, to the development of expertise over the last couple of years. In my opinion, every single study they cite in defense of the more strict nurture hypothesis suffers from a pretty severe restriction of range problem.
Yeah you mention in your book that when you are studying experts, you have such small samples and a restriction of range that it is nearly impossible to tell whether there were any differences between the people in talent to begin with.
Some of these studies are set up in a way to be very much biased against finding any evidence of any innate talent, and I would think that the scientists would have recognized that. And say, “okay we can make conclusions that practice is important,” but to go on to make the conclusion that because practice is important talent doesn’t exist, that’s a surprise to me to see scientists making that conclusion when they should understand from their experimental setup that they can’t even make that conclusion.
I wonder if that’s because this is the palatable conclusion, that we can all be come great.
Janet Starkes who is one of the most famous scientists, now retired, studying motor skill development and perceptual expertise in sports, told me, “Yeah, I always knew that people had different talents, but I felt like I had to stake out the other side of the argument in order to get people to listen, and then once people started listening more I could swing back to the center.” And I understand that from a practical standpoint, but the social message doesn’t bear on the scientific truth. And my opinion is that if we pretend like differences don’t exist, is that really the best way to get the optimal outcomes for all people? To get the best results for all people, I think we need to figure out which differences are real and important, and then figure out how to work with them, not to pretend as if they don’t exist.
In 1994, Australia launched its National Talent Search program, where children ages fourteen to sixteen were examined in school for body size and tested for general athleticism. Some of the athletes were channeled away from sports in which they had experience to unfamiliar sports that suited them. You wrote: “The successes with talent transfer attest to the fact that a nation succeeds in a sport not only by having many athletes who practice prodigiously at sport-specific skills, but also by getting the best all-around athletes into the right sports in the first place.” Are talent searches (within countries and worldwide) for sports commonplace today? Could a cultural emphasis on one sport likely channel most people in that culture to pursue that sport, when, in fact, their talent lies in another sport?
Largely talent searches are not common unless they are already at a high level. So I don’t think you need a talent search for the most popular sport in most countries. So, for example, in Brazil for soccer, Canada for hockey, or the United States for football. In the U.S. the best athletes in high school want to play football so you end up with a self-filtering. And this is what happens in Kalenjin runners in Kenya, they try a gold medalist training plan and only a couple of them can make it. So I think you’re having a self-filtering when you’re talking about the most popular sport in a given country. And then only at the top level do they start doing a more concerted talent search. But for any sport that is not the country’s most popular sport, in the Olympics the countries who do well in what are their minor sports, they most definitely are conducting talent searches for those minor sports. This is exactly why when a country is awarded the Olympics, seven years out from the Olympics they start a talent search and you see this huge increase in the number of medals the country wins when they host the Olympics. Because when they’re hosting they want to win on their home territory, so they start a talent search, and the search is almost always focused on sports that don’t have a big input of people in the first place, not sports like track and field. The Australians sort of exemplified this; they literally went and asked what sports have fewer competitors, and said we’re going to put people in these sports, even if they’ve never participated in them at all. Great Britain hugely increased their medal count in their home Olympics last year and the first gold medalist for the home team was a woman named Helen Glover, who was identified in a program called Sporting Giants where basically U.K. sports officials went to schools and clubs and took body measurements and told people “hey, you should try this sport.” And they picked Helen Glover out and they said, “Hey, you’ve got a body type that will fit in rowing, why don’t you give it a try?” Four years later she’s a national hero because she wins the first gold medal for Great Britain. It was a really interesting program, because they’ve started funneling people to rowing who had low brachial index, so short forearms compared to their total arms, long legs, and usually tall. And I remember there was this guy, and they picked him and put him in rowing, and he had this high aerobic capacity, and he was in the world championships in his first or second year of ever competing in the sport and he was actually winning at world championships and actually fell out of the boat. Athletically he had superior gifts compared to the people he was competing against, but had no technical skill. So you never really see someone fall out of a boat at a rowing championship but that’s what happened to this guy. So I think 100% the reason you see you this huge increase in medal count when a country hosts the Olympics is because they started a talent search that was focused on sports that don’t have a big input to begin with. The difference never comes from a huge increase in medal count in the most popular sport like in swimming or in track and field, it always comes from an accumulation of medals in these less widely contested sports. I saw a video of selection for Chinese divers, and you see these rows of kids who are made to put their arms above their head and if their elbow joints didn’t meet above their head they’re out because they’ll make too wide of an imprint when they hit the water.
Here’s my personal story. So I started running and it became the greatest thing in my life because I broke my arm playing football. And football is so popular, my colleague Peter King who is our most well-known writer, joked that the NFL is the most popular sport in America and the NFL draft is the second most popular sport in America. And when I go to high school football games and I see like 60 kids on the sidelines I see at least five I want to tell, “you should be on the cross country team, just by looking at you you should give it a shot because here you are sitting on the bench through a whole football game.” That’s not even good exercise. I really think it does some athletes a disservice, where they just all cluster to the most popular sport without regard to their body type.
In Jamaica, at the national high school championships, this is like 30,000 some odd people standing room only, it’s a crazy atmosphere, it’s amazing, it’s like the World Cup. But when an event comes up that is over 400 meters on the track people go to the bathroom. You know you can see a kid who has a body type that is good for an 800 meter runner and yet everyone is trying to run the 100. And you know, even I did the same thing. My racing weight in college was 130 pounds, but I was still playing football in high school until I broke my arm.
Your parents were okay with you playing football?
Nah, my mom wasn’t, but there’s only so much she could do. We got to wear the jersey into school on Friday’s during class. We would get pulled out of class to watch the movie The Program to get psyched up before our rivalry game. That was compelling, but I’m just really thankful that I broke my arm and found track.
Tiger Woods noted that “To this day, my dad has never asked me to go play golf. I ask him. It’s the child’s desire to play that matters, not the parent’s desire to have the child play.” Could motivation or grit actually be, in part, genetic? Can herculean will be separated from innate talent?
There has been plenty of research showing that exercise can alter the dopamine environment in your brain shown in animal models and humans. But I didn’t realize researchers who work in that area know full well that it can go in the reverse. That someone’s dopamine system and potentially variations in genes involved in dopamine systems can influence the amount of physical activity someone undertakes. And that’s done over and over again in animal models, and there is some proof of concept work for humans as well. I guess I knew this intuitively in that I know I’ve had friends and training partners some of whom really have to be managed in order to get them to train as much as they should and others who have to be managed to get them to stop training. This actually led to one of the coolest interviews in the book which was with Pam Reed one of the greatest ultra-marathoners of all time. When I interviewed her she had just finished the U.S. Nationals Iron Man triathlon in New York and the next day she was in La Guardia airport because her plane was delayed—and she told me she doesn’t feel calm unless she’s moving—and so I was interviewing her as she was running laps around the parking garage. So that’s obviously an extreme case.
That just shows individual differences. Most people could never imagine being like her, right?
Yes, not even close. And I’ve gone on TV and radio so far and a lot of people have picked up on this. They’ve said “so, you really can have couch potato genes” because I quoted someone using that phrase. And yes that’s true and they always make the joke, “oh, now I have an excuse for not getting off the couch.” But that’s not the way I would take it, I would take it as you would have to work harder to manipulate your environment to be conducive to you exercising than the next person does. And, for example, I know some very basic things that make my environment more conducive to my training. To have a training group, for example. And that gets at how genes have been really incorrectly portrayed and I partly blame the media for the one gene deterministic portrayal. So for example this week it’s the alcoholic gene, next week it’s the obesity gene, the third week it’s the promiscuity gene, as if these single genes mean you’re destined to do something.
You wrote that “In all likelihood, we overascribe our skills and traits to either innate talent or training, depending upon what fits our personal narratives.” This is an insightful point. Do you think this tendency to rely on our personal story has prevented us from understanding that success comes from both “innate hardware and learned software” as you put it?
I do. And let me tell a story that I cut from the book that I thought was funny. When I was looking at twin studies, there were these identical twins that were separated at birth and they were maybe in their 30’s or something when they were interviewed, and they both were fastidiously clean, what we would call neat freaks. One of them said, yeah well my adopted mother, she was really clean, so I took after her. And the other said, well, my adopted mother was such a slob so I never wanted to be like her. Maybe they’re right that these very different inputs sort of pushed them to the same outcome, or maybe they just can’t see their own genes and something in their genome predisposes them to being that way. And I think people construct narratives based on what they can see, and they can’t see their genes. Primarily, we take only what we know and construct a narrative through that. At least in the sports sphere I think the journalists tend to sort of fan that flame because of the fear that implicating innate talents might, aside from height, devalue somebody’s work.
You write: “If there exists a scientist or sports fan who would denigrate [Michael] Jordan’s hard work and skill because of his obvious gift of height, I didn’t meet him in the reporting of this book. In fact, the opposite extreme—ignoring gifts as if they didn’t exist—is much more common in the sports sphere.”
For example, one of the athletes you interviewed, Eero Mantyranta, could never be convinced that he had a beneficial genetic mutation (elevated hemoglobin levels) that gave him an advantage in the Olympic sport of cross country skiing. This is despite that fact that the scientist who studied him, Albert de la Chapelle, has stated “It’s an advantage, there’s no question.”
Could you give more examples of how the tendency to ignore gifts is commonplace? And how often do you think athletes want to believe they had no advantage and that their success came entirely from their personal efforts?
I actually think that elite athletes, when you talk to them, acknowledge that talent exists because their life is dedicated to doing what they’re doing. And they still all get beaten sometimes. They know. They’ve been around somebody who picks up something more quickly or beats them. Elite athletes of course talk about how much hard work they’ve put in because they have put in a lot of hard work, no question about it. They have to put in hard work to separate themselves from other people like them. But elite athletes are where I’ve gotten the least pushback to the idea that innate talent exists. So I think that’s often journalists and fans speaking on behalf of athletes, not the athletes speaking themselves. For Eero Mantyranta, it was a little different because his example was so tied up in people accusing him of doping and all this other stuff, that he got so used to saying “Well if I was doping this also could be a disadvantage because my blood would get too thick.” But for the most part I think if you really grill elite athletes, they will say that their hard work separated them but they’re talking about their peers. They’re not talking about if you compared them to all of humanity.
As another example, anyone who has been part of an elite running training group has been with others who have done the exact same thing and despite this fact you still don’t cross the finish line at the same time. In fact, sometimes you even get more different not more similar.
© 2013 by Jonathan Wai