In fact, in absolutely every single study of sports expertise, there is a tremendous range of hours of practice logged by athletes who reach the same level, and very rarely do elite performers log 10,000 hours of sport-specific practice prior to reaching the top competitive plane, often competing in a number of other sports—and acquiring a range of other athletic skills—before zeroing in on one. A study of ultraendurance triathletes found that the better athletes had practiced far more on average but that there was a tenfold difference in practice hours among athletes who performed similarly.
Studies of athletes have tended to find that the top competitors require far less than 10,000 hours of deliberate practice to reach elite status. According to the scientific literature, the average sport-specific practice hours to reach the international levels in basketball, field hockey, and wrestling are closer to 4,000, 4,000, and 6,000, respectively. In a sample of Australian women competing in netball (sort of like basketball but without dribbling or backboards), arguably the best player in the world at the time, Vicki Wilson, had compiled only 600 hours of practice when she made the national team. A study of athletes on Australia’s senior national teams found that 28 percent of them started their sport at an average age of seventeen, having previously tried on average three other sports, and debuted at the international level just four years later.
Even in this age of hyperspecialization in sports, some rare individuals become world-class athletes, and even world champions, in sports from running to rowing with less than a year or two of training. As with Gobet’s chess players, in all sports and skills, the only real rule is that there is a tremendous natural range.
•
In 1908, Edward Thorndike, who would become known as the father of modern educational psychology, came up with a way to test whether nature or nurture dominated an individual’s ability at a task. Thorndike was a leading proponent of the then-controversial idea that older adults—meaning, at the time, those over thirty-five—can continue to learn new skills. He figured that the way to distinguish nature from nurture was to give people the same amount of practice at a certain task and to see whether they became more or less alike. If their skill levels converged, Thorndike reasoned, then the impact of practice was overwhelming any innate individual differences. If they diverged, then nature was overpowering nurture.
In one experiment, Thorndike had adults practice multiplying three-digit numbers by three-digit numbers in their heads as quickly as they could. He was astounded by their improvement. “The fact that these mature and competent minds improved in the course of so short a training so much,” Thorndike wrote, “is worthy of attention.” After one hundred practice trials, many of the subjects cut their mental computation time in half. And every single subject improved. Just as in chess, language, music, and baseball, as practitioners improve at mental multiplication, they internalize patterns and systems of breaking problems into pieces that allow for increasingly rapid calculation.
But while Thorndike saw across-the-board improvement, he also noted what sociologists often call a “Matthew effect.” The term derives from a passage in the biblical Gospel of Matthew:
For to all those who have, more will be given, and they will have an abundance; but from those who have nothing, even what they have will be taken away.
Thorndike saw that the subjects who did well at the start of the training also improved faster as the training progressed compared with the subjects who began more slowly. “As a matter of fact,” Thorndike wrote, “in this experiment the larger individual differences increase with equal training, showing a positive correlation with high initial ability with ability to profit by training.” The passage from the Bible doesn’t quite capture Thorndike’s results accurately because every subject improved, but the rich got relatively richer. Everyone learned, but the learning rates were consistently different.
When World War I erupted, Thorndike became a member of the Committee on Classification of Personnel, a group of psychologists commissioned by the U.S. Army to evaluate recruits. It was there that Thorndike rubbed off on a young man named David Wechsler, who had just finished his master’s degree in psychology. Wechsler, who would become a famous psychologist, developed a lifelong fascination with tracing the boundaries of humanity, from lower to upper limits.
In 1935, Wechsler compiled essentially all of the credible data in the world he could find on human measurements. He scoured measures of everything from vertical jump to the duration of pregnancies to the weight of the human liver and the speeds at which card punchers at a factory could punch their cards. He organized it all in the first edition of a book with the aptly momentous title The Range of Human Capacities.
Wechsler found that the ratio of the smallest to biggest, or best to worst, in just about any measure of humanity, from high jumping to hosiery looping, was between two to one and three to one. To Wechsler, the ratio appeared so consistent that he suggested it as a kind of universal rule of thumb.
Phillip Ackerman, a Georgia Tech psychologist and skill acquisition expert, is a sort of modern-day Wechsler, having combed the world’s skill-acquisition studies in an effort to determine whether practice makes equal, and his conclusion is that it depends on the task. In simple tasks, practice brings people closer together, but in complex ones, it often pulls them apart. Ackerman has designed computer simulations used to test air traffic controllers, and he says that people converge on a similar skill level with practice on the easy tasks—like clicking buttons to get planes to take off in order—but for the more complex simulations that are used for real-life controllers, “the individual differences go up,” he says, not down, with practice. In other words, there’s a Matthew effect on skill acquisition.
Even among simple motor skills, where practice decreases individual differences, it never drowns them entirely. “It’s true that doing more practice helps,” Ackerman says, “but there’s not a single study where variability between subjects disappears entirely.”
“If you go to the grocery store,” he continues, “you can look at the checkout clerk, who is using mostly perceptual motor skill. On average, the people who’ve been doing it for ten years will get through ten customers in the time the new people get across one. But the fastest person with ten years’ experience will still be about three times faster than the slowest person with ten years’ experience.”
Scientists who study skill performance attempt to account for “variance” between people. Variance is a statistical measure of how much individuals deviate from the average. In a sample of two runners, if one athlete completes the mile in four minutes and the other runs it in five minutes, then the average is four and a half minutes and the variance is half a minute. The question for scientists is: What accounts for that variance, practice, genes, or something else?
It is a critical inquiry. It is not enough for scientists to say that practice matters. That point is entirely uncontroversial. As Joe Baker, a sports psychologist at York University in Toronto, says, “There isn’t a single geneticist or physiologist who says hard work isn’t important. Nobody thinks Olympians are just jumping off the couch.”
Scientists must go beyond saying that practice matters and attempt the difficult task of determining exactly how much practice matters. By the strictest 10,000-hours thinking, accumulated practice should explain most or all of the variance in skill. But that never, ever happens. From swimmers and triathletes to piano players, studies report that the amount of variance accounted for by practice is generally between low and moderate.
In a study that K. Anders Ericsson himself coauthored of darts players, for example, only 28 percent of the variance in performance between players was accounted for after fifteen years of practice. At the rate of skill convergence documented in that study, a 10,000-years rule might be more likely than a 10,000-hours rule—if, that is, the players would ever reach the same level at all.
The data quite clearly su
pport a view of skill—from chess and music to baseball and tennis—that is based on a paradigm not of “hardware not software,” but of both innate hardware and learned software.
3
Major League Vision and the Greatest Child Athlete Sample Ever
The Hardware and Software Paradigm
In 1992, his first year of research on the Los Angeles Dodgers, Louis J. Rosenbaum met with an unexpected problem. The players were literally off the charts.
Rosenbaum had been the team ophthalmologist for the NFL’s Phoenix Cardinals since 1988, and now he was in Dodgertown, the spring training facility in Vero Beach, Florida, to test eighty-seven players in the Dodgers organization, major league players as well as minor leaguers hoping to earn their spot in the show.
From eight A.M. to five P.M., Rosenbaum tested players for traditional visual acuity, dynamic visual acuity (the ability to see detail in moving objects), stereoacuity (the ability to detect fine differences in the depth of objects), and contrast sensitivity (the ability to differentiate fine gradations of light and dark). For the visual acuity test, instead of the usual eye chart with the big E on top, Rosenbaum and his colleagues used Landolt rings—circles with a gap in one section that the viewer must pick out as the rings get progressively smaller toward the bottom of the chart.
The trouble was that Rosenbaum used commercially available Landolt ring charts, which tested visual acuity down to 20/15.* Nearly every player maxed out the test.
Fortunately, the other vision tests were successful. So when gruffly skeptical Tommy Lasorda, the Dodgers’ legendary manager, challenged Rosenbaum to predict which minor leaguer would thrive in the majors, Rosenbaum had plenty of data to pore over. He did not have the players’ baseball statistics and so had to rely purely on the vision testing data. He chose a minor league first baseman with outstanding scores.
The player was Eric Karros, a mere sixth-round pick in the 1988 draft. By ’92, though, Karros was starting at first base for the Dodgers and won the National League Rookie of the Year award. It was his first of thirteen full seasons as a major leaguer.
The following spring, Rosenbaum returned to Dodgertown with a custom-made visual acuity test that went down to 20/8. Given the size and shape of particular photoreceptor cells, or cones, in the eye, 20/8 is around the theoretical limit of human visual acuity.
One’s maximum visual acuity is determined by the density of cones in the macula, an oval-shaped spot in the retina of the eye. Cone density in humans is akin to the megapixel rating in digital cameras, and it is highly variable between people. Scientists who have collected retinas from deceased adults, ages twenty to forty-five, found a range from 100,000 cones/mm2 to 324,000 cones/mm2. (If one’s cone density is below 20,000 cones/mm2, a magnifying glass will be needed to read the newspaper.) As Michael A. Peters, author of See to Play and an eye doctor who works with pro baseball and hockey players, puts it: the number of cones appears to be “genetically predetermined for each of us.”
Armed with a custom test at the 1993 spring training, Rosenbaum could finally measure how well pro ballplayers see. Again, Lasorda challenged Rosenbaum to predict which minor leaguer would make a distinguished pro. This time, the player whose vision tests stood out to Rosenbaum was Mike Piazza, a lightly regarded catcher.
Piazza had been picked by the Dodgers five years earlier in the sixty-second round of the draft, the 1,390th player taken overall, and only because Piazza’s father was a childhood friend of Lasorda’s. Nonetheless, Piazza would make good on Rosenbaum’s prediction. He won the National League Rookie of the Year in 1993 and went on to become the greatest hitting catcher in baseball history.
Over four years of testing, and 387 minor and major league players, Rosenbaum and his team found an average visual acuity around 20/13. Position players (players who have to hit) had better vision than pitchers, and major league players had better vision than minor leaguers. Major league position players had an average right eye visual acuity of 20/11 and an average left eye visual acuity of 20/12. In the test of fine depth perception, 58 percent of the baseball players scored “superior,” compared with 18 percent of a control population. In tests of contrast sensitivity, the pro players scored better than collegiate baseball players had in previous research, and collegiate players scored better than young people in the general population. In each eye test, pro baseball players were better than nonathletes, and major league players were better than minor league players. “Half the guys on the Dodgers’ major league roster were 20/10 uncorrected,” Rosenbaum says.
The two largest population studies of visual acuity, one from India and one from China, give a sense of just how rare 20/10 vision might be. In the Indian study, out of 9,411 tested eyes, one single eye had 20/10 vision. In the Beijing Eye Study, only 22 out of 4,438 eyes tested at 20/17 or better.
Smaller studies focused only on young people, though, have documented average vision that is better than the standard 20/20. Seventeen- and eighteen-year-olds in a Swedish study had average visual acuity around 20/16. So we should expect that Major League Baseball hitters—their average age is around twenty-eight—would have better than 20/20 vision just because they are young, but not an average of 20/11. (Coincidentally, or perhaps not, twenty-nine often is the age at which visual acuity starts to deteriorate and the age when hitters, as a group, begin to decline.)
Mark Kipnis shared with me his first recollection of his baseball-playing son Jason’s visual acuity. It was during a ski vacation when Jason was twelve years old. The Kipnis family was sitting in a large restaurant in a lodge and Mark wanted to see the score of a football game on a television in the far corner. He was tired, so he asked Jason to get up, walk over to the television, and tell him the score. “He just turned his head and told me the score,” Mark says, “and a little light went off in my head.” A decade later, Jason was selected by the Cleveland Indians in the second round of the 2009 draft. By 2011, he was starting at second base.
Ted Williams, the last man to hit .400 over a major league season, used to insist that he only saw ducks on the horizon before his hunting partners because he was “intent on seeing them.” Perhaps. But Williams’s 20/10 vision, discovered during his World War II pilot’s exam, probably didn’t hurt either.*
About 2 percent of the players in the Dodgers organization dipped below 20/9, flirting with the theoretical limit of the human eye. Daniel M. Laby, an ophthalmologist who worked on the Dodgers study and later with the Boston Red Sox, says that he encounters a few players at that level every year in spring training. “I can pretty comfortably say that in twenty years of caring for people’s eyes I’ve never seen someone outside pro athletics achieve that, and I’ve seen over twenty thousand people,” Laby says. David G. Kirschen, an optometrist who also works with professional athletes and is chief of binocular vision and orthoptic services at the Jules Stein Eye Institute at UCLA’s medical school, says that he has seen a few patients outside of elite sports with 20/9 vision, “but you can count them on one hand over thirty years.”
So while major league hitters might not have any faster reaction time than you or I do, they do have the superior vision that can help them pick up the anticipatory cues they need earlier, making raw reaction speed less important.*
Baseball players have to know before the final two hundred milliseconds of a pitch where to swing, so the earlier they pick up anticipatory cues the better. One such cue, as psychologist Mike Stadler writes in The Psychology of Baseball, is the “flicker” of a pitch, or the indication of the spin of the ball by the flashing pattern of rotating red seams. Two-seam fastballs and curveballs are foretold by signature red stripes on the side of the ball. A four-seam slider shows the batter a bright red dot in the center of a white circle. “That circle right out of the [pitcher’s] hand, you identify in your brain, ‘Oh, okay, slider,’” Keith Hernandez, the five-time All-Star first baseman, once said in television commentary of a Mets ga
me. “If you didn’t have those little red seams on the ball, you’d be in a world of trouble.”
The importance of picking up ball rotation has been demonstrated in virtual-reality batting studies in which baseball players were asked to identify or to swing at digital pitches. When players picked up the rotation of the ball, they identified pitches more accurately and executed more precise swings. Hitters performed better when the red seams of the ball were accentuated, and worse when the seams were covered with white paint.
•
It’s easy to understand why an athlete with outstanding visual acuity but without the mental database of what to look for is as useless as Albert Pujols facing Jennie Finch. But once the data is downloaded into the brain, it’s advantageous to see those signals as clearly and as early as possible, all the better not to have to rely on pure reaction speed.* Al Goldis, a longtime major league scout who studied motor learning in grad school, says: “If a player has better visual skills, he can pick up the pitch while it’s five feet or ten feet closer to the pitcher. If he doesn’t, his mechanics might be outstanding but he reacts so late that he breaks his bat because the ball is in on his hands. It’s not the bat speed, it’s the visual skills. That little bit is the difference between ordinary and extraordinary.”
When Laby and Kirschen studied U.S. Olympians from the 2008 Beijing Games, they found that the softball team had an average visual acuity of 20/11, outstanding depth perception, and better contrast sensitivity than athletes from any other sport. Olympic archers also had exceptional visual acuity—they scored similarly to the Dodgers—but not particularly good depth perception. That makes sense, Laby says, because the target is far away, but it’s also flat. Fencers, who must make rapid use of tiny, close-range variations in distance, scored very well on depth perception. Athletes who track flying objects at a distance—softball players and to a lesser extent soccer and volleyball players—scored well on contrast sensitivity, which is “probably set at a certain ability you’re born with,” Laby says.*
The Sports Gene: Inside the Science of Extraordinary Athletic Performance Page 5