You might have heard that it takes 10,000 hours of practice to become an expert—an intuition Malcolm Gladwell discusses in his book “Outliers”, that’s morphed into a pseudo-fact given its circulation by countless writers. Hopefully you’ve also heard that this practice needs to be “deliberate” — practicing with the specific intent to improve performance — a type of practice that requires a lot of mental and physical effort. The implication of this is that you can’t simply do something a lot and get better at it. Years of experience doesn’t make one an expert. But if you put in those 10,000 hours of hard work, you too can achieve the kind of greatness you see from the world’s top athletes, musicians and even business people. At least that’s how the intuition goes.
Well the authors behind this intuition must have spent 10,000 hours overgeneralizing, because they’ve truly become experts at it! For one thing, the original research that gave rise to the number 10,000 was a study from 1993 on musicians 1 (violinists, to be specific) where it was found that the “best” had accumulated over 10,000 hours over the course of their music career, the “good” about 8,000 hours, and the least accomplished about 5,000 hours. This study has been highly influential, having been cited over 5000 times since publication. However, if you take a closer look at expertise research you’ll see that it’s not so simple — there’s considerable variation between different people and skills.2 For instance, the hours of deliberate practice reported by chess players who have reached “master” status ranged from 3,016 to 23,608 hours!3 Clearly there is more to expert performance than amassing hours of practice (we’ll discuss what other factors might be important near the end).
This isn’t to say deliberate practice isn’t important. On the contrary, practice remains one of the most important factors to consider when explaining expert performance. Survey the professional ranks of any sport and you won’t find anyone at that level who just picked it up yesterday. Genetics alone can’t get you Steph Curry’s three-point shot, or Mike Trout’s impeccable swing. You must practice. But how important is it? Well, a recent meta-analysis (a “study of studies”, combining and analyzing many data sets) of 52 studies (including a total of 2765 athletes) investigated the relationship between deliberate practice and performance in sports.4 The authors of the study picked apart the data to get a handle on how the results differed between different types of sport skills (“task characteristics”) and people (“participant characteristics”). Here, I’ll give a brief review of their findings and some of my thoughts.
As you can see from the chess example above, there’s a lot of variance in how much deliberate practice explains expert performance. Overall, the authors of the present study found that deliberate practice explained only 18% of the variance observed in sports performance. That seems surprisingly low — but it gets even more interesting when you look at the data more closely.
When tallying up someone’s hours spent practicing deliberately, should you include the time they spent in a competition, or playful activities? Apparently, it doesn’t matter — when the measure included such activities the variance explained was still 17%. However, I suspect that because competition is a very deliberate form of practice (if the athlete is trying hard to win), and playful activities are not very deliberate at all, the effects of each might cancel each other out (although one could make the case that play can be as important as deliberate practice given its exploratory nature).
Interestingly, there wasn’t a big difference between team sports (17% variance explained) and individual sports (18%). Although team and individual sports can differ substantially — primarily in whether you need to worry about what other people are doing — it seems the skills required are equally amenable to practice. Somewhat similarly, there wasn’t a huge difference between ball sports (20% variance explained) and non-ball sports (15%). The authors also investigated whether there’s a difference between “open” and “closed” skills. Open skills involve a changing environment which often forces an athlete to make decisions about what skill to perform and when. Closed skills involve a predictable environment in which an athlete can focus on executing their skills. Most sports require both and fall on a continuum between being dominantly “open” or “closed”. The authors found that there were little differences between sports defined as primarily “open” (17% variance explained) or “closed” (19%).
However, there was a notable (although not “statistically significant”) difference between externally paced (17% variance explained) and internally paced sports (41%) — the latter of which means you can choose when to “go” (e.g. when to throw while playing darts) compared to being demanded to perform by the situation (e.g. taking a shot in Hockey, the timing of which can be very unpredictable). Personally, I wonder whether this result would be different if people approached practicing skills more intelligently. For the most part (at least with coaches I’ve spoken to, and the research I’ve read) people still aren’t approaching skill development the way science recommends… people still try to “cram”, used blocked practice schedules, and isolate skills outside of realistic sport situations. If people practiced with more variability and contextual uncertainty, and “sparred” or played more short-sided games, maybe athletes would develop more flexible skills and strategies, and practice would have a greater effect on skills requiring external pacing.
The authors found no differences between young athletes (19% variance explained) and adult athletes (18%). Interesting result, as you might think adults would have amassed more practice and their performance would be more dependent on it as a result. As a bit of an aside, the authors performed a sub analysis to determine whether starting a sport younger made one more likely to become elite. One school of thought suggests that starting early is critical as it allows you to amass more practice earlier, but another school of thought is that one should avoid specializing early to avoid burnout and injuries, and because trying a variety of sports early develops more general skills and coordination. The sub analysis in this paper seems to indicate that starting age doesn’t seem to matter. I’m sure this isn’t the last we’ve heard on this topic.
The authors found no significant differences between practicing alone (22% variance explained) and with others (18%). This seems to be in line with the indifference between individual and team sports discussed above.
Does it matter how we determine an athletes level of expertise? Maybe. When an athlete is rated by their “group membership” (are they on an elite team?) deliberate practice explained 25% of the variance in their performance. This is only slightly higher than using standardized objective scores (like a batting average; 20% variance explained), but way higher than when using an expert’s rating (a coach’s opinion of an athlete; 2%). I’m not too sure what to make of this. I suppose that while points win games — and therefore one would think it’s the objective scores that we should be more concerned with — it’s clear that coaches know the athletes more intimately and understand their strengths and weaknesses, their consistency and potential, and therefore rate athlete performance in far more nuanced ways.
One of the biggest differences found in this meta-analysis is that deliberate practice appears to be far more important for sub-elite athletes (29% variance explained) compared to elite athletes (1%). Wait… only 1% of elite athlete performance is explained by deliberate practice hours? This is very interesting. I would contend, however, that this is not terribly surprising.
Presumably, all elite athletes were once sub-elite… they must develop their skills, and therefore practice matters. But once you’re bumping up against the upper asymptote of skill and experiencing diminishing returns from practice, other factors start to matter more. Three factors I can think of off the top of my head:
- Genetics — think about Michael Phelps and how his body is practically genetically engineered for swimming. Of course, he still had to train a lot, but the athletes he competed against at the Olympics probably trained for similar amounts of hours.
- Lifestyle factors — really, anything that affects your ability to perform. To execute those skills you’ve developed, you must manage consistent sleep, nutrition, general fitness, mental and physical fatigue, stress management, etc.
- Practice Consistency — yes, you might have banked 10,000 hours, but when you stop training, skills decline (more on this later).
What else matters?
If the amount of deliberate practice time an elite athlete banked isn’t enough to explain their high level of performance, what else might matter? The authors briefly comment at the end and cite a few studies illustrating the importance of genetic factors. This could include anything from the shape of your body (anthropometrics) to your mental inclinations, neither of which may be affected by practice. Further, genetics appears to put a cap on how much you can improve on certain aspects of performance, like endurance capacity, ease of gaining muscle mass, and certain features of your nervous system. The authors also discuss a variety of psychological factors, like confidence, tendency to experience performance anxiety, aversion to negative outcomes and sensitivity to reward, general intelligence, working memory capacity, attentional control, perceptual and psychomotor speed. That’s a lot of stuff!
Personally, I think practice is a bit more important than the results of this study imply. For two reasons:
- These studies almost never include a valid measure of practice quality — one might be trying to get better but doing the wrong things. They would still include this poor-quality practice in their total “deliberate” practice hours. What is or isn’t “good quality” practice is a frequent topic of this blog.
- Consistency matters. The negative effects of de-training are well recognized in the physical training literature where it’s been shown that athletes lose aerobic capacity, strength, power and muscle mass rapidly when they stop training.5–7 It’s the same with skill — it’s well known that consistent, spaced practice is far more effective for learning than cramming with large breaks.8 Yet studies of deliberate practice simply tally practice hours with no regard to consistency.
Human performance is complex. Whether you’re talking about sports performance, musical ability, or even business acumen, a person’s ability is determined by numerous interacting factors. While it’s hard to fight your genetics (for now…), you can still improve your chances of achieving your highest potential by considering all these factors, addressing what you can, and training hard.
- Ericsson, A., Krampe, R. & Tesch-Römer, C. The role of deliberate practice in the acquisition of expert performance. Psychological Review 100, 363–406 (1993).
- Macnamara, B., Hambrick, D. & Oswald, F. Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis. Psychological science 25, 1608–18 (2014).
- Gobet, F. & Campitelli, G. The role of domain-specific practice, handedness, and starting age in chess. Dev Psychol 43, 159–72 (2007).
- Macnamara, B. N., Hambrick, D. Z. & Moreau, D. How Important Is Deliberate Practice? Reply to Ericsson (2016). Perspect Psychol Sci 11, 355–8 (2016).
- Izquierdo, M. et al. Detraining and Tapering Effects on Hormonal Responses and Strength Performance. J Strength Cond Res 21, 768 (2007).
- Liao, Y.-H., Sung, Y.-C., Chou, C.-C. & Chen, C.-Y. Eight-Week Training Cessation Suppresses Physiological Stress but Rapidly Impairs Health Metabolic Profiles and Aerobic Capacity in Elite Taekwondo Athletes. Plos One 11, e0160167 (2016).
- Koundourakis, N. et al. Discrepancy between Exercise Performance, Body Composition, and Sex Steroid Response after a Six-Week Detraining Period in Professional Soccer Players. Plos One 9, e87803 (2014).
- Smolen, P., Zhang, Y. & Byrne, J. The right time to learn: mechanisms and optimization of spaced learning. Nat Rev Neurosci 17, 77–88 (2016)