The Sports Gene: Inside the Science of Extraordinary Athletic Performance

by David Epstein

  In the shoulder muscle of one Danish national team kayaker (as well as in that kayaker’s brother) Andersen found more than 90 percent slow-twitch muscle fibers. The kayaker was trying to qualify for the Olympics in either the 500- or 1,000-meter race, but his competitors were so much more explosive off the starting line that even though he always caught up late in races, he constantly fell short and had no chance to make the Olympic team. Andersen told the kayaker about his muscle fiber type distribution and suggested he switch races. The kayaker moved to long-distance competitions and quickly became one of the top racers in the world.

  Despite his successful applications of muscle fiber research in track and field and kayaking, soccer vexes Andersen. Soccer coaches all want the fastest athletes, so Andersen wondered how it could be that many of the Danish pros have fewer fast-twitch fibers than an average person on the street. He turned to F.C. Copenhagen’s development academy, where he found that the swiftest players are lost to chronic injuries before they ever reach the top level. “The guys that have the very fast muscles can’t really tolerate as much training as the others,” he says. “The guys with a lot of [fast-twitch fibers] that can contract their muscles very fast have much more risk of a hamstring injury, for instance, than the guys who cannot do the same type of explosive contraction but who never get injured.”

  The less injury-prone players survived the development years, which is why the Danish elite level ended up skewed toward the slow-twitch. “In American football,” Andersen says, “the big fat guy becomes one position and the fast guy becomes a wide receiver and they train differently. But the soccer players are trained all alike. I hear coaches say all the time, ‘We can’t use him because he’s always injured.’ If he gets injured all the time, it’s probably because we do something wrong to him and we need to change that. We shouldn’t lose the fastest players.”

  Even with all the money and glory available in international soccer, coaches—at least in Denmark—may be losing some of the most fleet players before they ever reach the professional pitch. The same medicine should not be prescribed for every athlete. For some, less training is the right medicine.

  •

  If innate body type differences that are hidden from our naked eyes, like fiber type proportions, are not accounted for, some athletes are sacrificed to the idea that the same hard training works for everyone. The slow-twitch kayaker who turned from sprints to long-distance might have squandered his career losing shorter races if Andersen had not steered him toward the long-distance races he could win.

  In other cases, it is much more obvious how fixed physical traits fit into particular sports amid the rapidly shifting gene pool of competitive athletics.

  7

  The Big Bang of Body Types

  Decades ago, particularly in Europe, local club sports teams supported a large number of regionally competitive, or even semiprofessional, athletes who often made up humanity’s elite performers. Until technology tilted the landscape.

  Today, literally billions of customers have a ticket to the Olympics, the World Cup, or the Super Bowl with the flick of a remote control. As a result, most sports enthusiasts are now spectators to the elite as opposed to participants in the comparatively quotidian, a huge population of recliner-bound quarterbacks paying to watch a tiny number of real QBs. That scenario creates what economist Robert H. Frank termed a “winner-take-all” market. As the customer base for viewing extraordinary athletic performances expanded, fame and financial rewards slanted toward the slim upper echelon of the performance pyramid. As those rewards have increased and become concentrated at the top level, the performers who win them have gotten faster, stronger, and more skilled.

  A group of sports psychologists, particularly acolytes of the strict 10,000-hours school, have argued that improvements in individual sport world records and team sport skill levels have increased so vastly in the last century—faster than evolution could have significantly altered the gene pool—that the improvement must come down solely to increasing amounts of practice. As the rewards for top performers have grown, more athletes have undertaken greater quantities of practice in an attempt to earn them.

  A portion of the improvements, though, even in straightforward athletic endeavors, are very clearly the result of technological enhancements. Biomechanical video analysis of legendary sprinter Jesse Owens, for example, has shown that his joints moved as fast in the 1930s as those of Carl Lewis in the 1980s, except Owens ran on cinder tracks that stole far more energy than the synthetic surfaces where Lewis set his records.

  But technology is not the only source of improvement that is often overlooked. Undoubtedly, the increasing amount and precision of practice has helped push the frontiers of performance. But the winner-take-all effect, combined with a global marketplace that has allowed many more people to audition for the minuscule number of increasingly lucrative roster spots, has indeed altered the gene pool. Not the gene pool in all of humanity, but certainly the gene pool within elite sports.

  •

  In the mid-1990s, Australian sports scientists Kevin Norton and Tim Olds began compiling data on the body types of athletes to see whether there had been significant changes over the twentieth century. The sports science, after all, had changed drastically.

  In the late nineteenth century, researchers of the science of body types—known as anthropometry—arrived at conclusions influenced by classical philosophy, like Plato’s concept of ideal forms; by art, such as Leonardo da Vinci’s Vitruvian Man, the famous depiction of a man’s body inscribed in a circle and square indicating the ideal human proportions; as well as by racially charged agendas. “There is a perfect form or type of man,” reads a late-nineteenth-century article enumerating the characteristics of an athlete, “and the tendency of the race [i.e. the white race] is to attain this type.”

  At the time, anthropometrists felt that human physique was distributed along a bell curve, and the peak of the curve—the average—was the perfect form, with everything to the sides deviating by accident or fault. So they asserted that the best athletes would have the most well-rounded, or average, physical builds. Not too tall or too small, neither too skinny nor too bulky, but rather a just-right Goldilocks-porridge version of a man. (And it was only men.) That was the belief for any sport: the average human form would be ideal for all athletic pursuits. This confluence of subjective theory and philosophy dominated the agenda for coaches and physical education instructors in the early twentieth century, and it showed in athletes’ bodies. In 1925, an average elite volleyball player and discus thrower were the same size, as were a world-class high jumper and shot putter.

  But, as Norton and Olds saw, as winner-take-all markets emerged, the early-twentieth-century paradigm of the singular, perfect athletic body faded in favor of more rare and highly specialized bodies that fit like finches’ beaks into their athletic niches. When Norton and Olds plotted the heights and weights of modern world-class high jumpers and shot putters, they saw that the athletes had become stunningly dissimilar. The average elite shot putter is now 2.5 inches taller and 130 pounds heavier than the average international high jumper.

  On a height-versus-weight graph, the duo plotted the average physiques of elite athletes in two dozen sports; one data point for the average build of an athlete in each sport in 1925, and another for the average build of an athlete in that same sport seventy years later.

  When they connected the dots from 1925 to the present for each sport, a distinct pattern appeared. Early in the twentieth century, the top athletes from every sport clustered around that “average” physique that coaches once favored and were grouped in a relatively tight nucleus on the graph, but they had since blasted apart in all directions. The graph looked like the charts that astronomers constructed to show the movement of galaxies away from one another in our expanding universe. Hence, Norton and Olds called it the Big Bang of body types.

  Just as
the galaxies are hurtling apart, so are the body types required for success in a given sport speeding away from one another toward their respective highly specialized and lonely corners of the athletic physique universe. Compared with all of humanity, elite distance runners are getting shorter. So are athletes who have to rotate in the air—divers, figure skaters, and gymnasts. In the last thirty years, elite female gymnasts have shrunk from 5'3" on average to 4'9". Simultaneously, volleyball players, rowers, and football players are getting larger. (In most sports, height is prized. At the 1972 and ’76 Olympics, women at least 5'11" were 191 times more likely to make an Olympic final than women under five feet.) The world of pro sports has become a laboratory experiment for extreme self-sorting, or artificial selection, as Norton and Olds call it, as opposed to natural selection.

  Big Bang data in hand, Norton and Olds devised a measure they called the bivariate overlap zone (BOZ). It 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. Not surprisingly, as winner-take-all markets have driven the Big Bang of body types, the genes required for any given athletic niche have become more rare, and the BOZ for most sports has decreased profoundly. About 28 percent of men now have the height and weight combination that could fit in with professional soccer players; 23 percent with elite sprinters; 15 percent with professional hockey players; and 9.5 percent with Rugby Union forwards.

  In the NFL, one extra centimeter of height or 6.5 extra pounds on average translates into about $45,000 of extra income. (Particular professions that require unique physiques have an even more concentrated winner-take-all structure and outdo even professional sports. The BOZ for regional catwalk models is less than 8 percent, dropping to 5 percent for international models, and to just 0.5 percent for supermodels.)

  And the Big Bang of body types goes down to the body-part level as well. While tall athletes have grown taller at a much faster rate than humanity as a whole, and small athletes have shrunk relatively smaller, athletes in certain sports have increasingly needed extremely specialized body traits. Measurements of elite Croatian water polo players from 1980 to 1998 show that over two decades the players’ arm lengths increased more than an inch, five times as much as those of the Croatian population during the same period. As performance requirements become stricter, only the athletes with the necessary physical structure consistently make the grade at the elite level. The shorter-armed athletes are more often weeded out.

  In addition to having longer arms overall, the bone proportions in the arms of top water polo players have changed. Elite players now have longer lower arms compared with their total arm length than do normal people, giving them a more efficient throwing whip. The same is true of athletes who need long levers for powerful, repetitive strokes, like canoeists and kayakers. Conversely, elite weight lifters have increasingly shorter arms—and particularly shorter forearms—relative to their height than normal people, giving them a substantial leverage advantage for heaving weights overhead. One of the many failings of the NFL combine that tests prospective draft picks in physical measures is that arm length is not taken into account in the measure of strength. Bench press is much easier for men with shorter arms, but longer arms are better for everything on the actual football field. So a player who is drafted high because of his bench press strength may actually be getting a boost from the undesirable physical characteristic of short arms.

  Top athletes in jumping sports—basketball, volleyball—now have short torsos and comparatively long legs, better for accelerating the lower limbs to get a more powerful liftoff. Professional boxers come in an array of shapes and sizes, but many have the combination of long arms and short legs, giving greater reach but a lower and more stable center of gravity.

  The height of a sprinter is often critical to his best event. The world’s top competitors in the 60-meter sprint are almost always shorter than those in the 100-, 200-, and 400-meter sprints, because shorter legs and lower mass are advantageous for acceleration. (Short legs have a lower moment of inertia, which essentially means less resistance to starting to move.) Sprinters hit the highest top speeds in the 100- and 200-meter races, but the 60-meter race has a proportionally longer acceleration period. Perhaps the advantage of shortness for acceleration explains why NFL running backs and cornerbacks, who make their livings starting and stopping as quickly as possible, have gotten shorter on average over the last forty years, even while humanity has grown taller.

  On occasion, technique changes in sports have changed the advantaged body types almost overnight. In 1968, Dick Fosbury unveiled his “Fosbury flop” method of high jump, which gives an advantage to athletes who have a high center of gravity. In just eight years after Fosbury’s innovation, the average height of elite high jumpers increased four inches.*

  In other cases, body types have more nuanced effects. While smallness is generally a boon for endurance runners, Paula Radcliffe, the world record holder in the women’s marathon, at 5'8" is literally head and shoulders above most of her world-class competitors. It didn’t keep the iconically tough Brit from winning eight marathons in the prime of her career, 2002 to 2008. But Radcliffe’s size may have helped confine most of her victories to autumn. One reason that marathon runners tend to be diminutive is because small humans have a larger skin surface area compared with the volume of their body. The greater one’s surface area compared with volume, the better the human radiator and the more quickly the body unloads heat. (Hence, short, skinny people get cold more easily than tall, hefty people.) Heat dissipation is critical for endurance performance, because the central nervous system forces a slowdown or complete stop of effort when the body’s core temperature passes about 104 degrees.*

  While Radcliffe in her prime was unbeatable on autumn mornings when races were held in cool temperatures, she was feckless in summer heat. At the Athens Olympics in 2004, when the marathon was held in 95-degree heat, despite having by far the fastest time coming into the race she was unable to finish and crumpled in a heap by the side of the road. The woman who won the race was 4'11". At the Beijing Olympic marathon in 2008, the temperature was 80 degrees and humid and Radcliffe finished a distant twenty-third. From 2002 to 2008, Radcliffe was 8-0 in marathons contested in cool or temperate conditions, and 0-2 and never even in contention in the sweltering summer Olympic races.

  •

  Collecting data for the most famous study of athletic body types ever conducted took an international research team a full year and included 1,265 athletes who competed at the 1968 Mexico City Olympics, representing every sport (except equestrian) and 92 different countries. It took six more years for the results to be compiled and published in a 236-page book. Half the book is simply tables of body measurements. Even without text, they convey an obvious message: in most Olympic sports, athletes are generally more physically similar to one another than I am to my own brother.

  Within track and field, most of the athletes could be pinned to an event simply by their body measurements. The men and women who raced the 400- and 800-meters or the high hurdles were the tallest of the runners—no surprise, given that the goal in hurdling is to clear the barriers with as little movement of the center of gravity as possible—while the marathoners were the shortest. No surprise there, either. But the similarities extended to less obvious physical traits of the skeleton.

  Athletes in a sport or event tended to be similar in height and weight—and often different from a control population of nonathletes—and also with respect to the breadth of their pelvic bone and the skeletal structure of their shoulders.

  Nonathlete women who were measured as a control group for the study had, of course, wider pelvic bones than nonathlete men. But female swimmers had more narrow pelvic bones than the normal, control population of men. And female divers had more narrow pelvic bones than the female swimmers. And female sprinters more narrow than the female divers. (Slim hip
s make for efficient running.) And female gymnasts had slimmer hips still.

  Female sprinters had much longer legs than the control population of women, and about as long as the control men. Male sprinters were around two inches taller than the control men, and 100 percent of that was in their legs, such that when they were seated the sprinters were the same height as the control men.

  The male swimmers were, on average, more than 1.5 inches taller than the sprinters, but nonetheless had legs that were a half-inch shorter. Longer trunks and shorter legs make for greater surface area in contact with the water, the equivalent of a longer hull on a canoe, a boon for moving along the water at high speed. Michael Phelps, at 6'4", reportedly buys pants with a 32-inch inseam, shorter than those worn by Hicham El Guerrouj, the Moroccan runner who is 5'9" and holds the world record in the mile. (Like other top swimmers, Phelps also has long arms and large hands and feet. That elongated body type can be indicative of a dangerous illness called Marfan syndrome. According to Phelps’s autobiography, Beneath the Surface, his unusual proportions led him to get checked annually for Marfan.)*

  The more that elite sports markets have shifted from participatory affairs to events for bulging masses of spectators, the more rare the bodies required for success have become, and the greater the money needed to attract those rare bodies to a particular sport. In 1975, athletes in the major American sports averaged roughly five times the median salary for an American man. Today, the average salaries in those sports are between about forty and one hundred times the median full-time salary. In order to match a single year’s salary of the highest-paid athletes, an American man making the country’s median annual income for a full-time job would have to work for five hundred years.