10,000 hours vs training debate: No scientific limits making it impossible for any individual to become an elite athlete with practice?
Dear Anders Ericsson...a request on behalf of sports science to stop telling people that the world is flat
The 10,000 hours vs genetic debate, and correcting Prof Ericsson's mistruths
So last night, I was (un)fortunate enough to be involved in a radio debate with Prof Anders Ericsson on the concept of talent vs training. For those who don't know, Ericsson is the father of the 10,000 hour concept, where he prescribes that ANY individual can become an elite athlete if they engage in the required hours of deliberate practice. He sets that number at 10,000 hours, which is really more marketing than it is science, and I had the chance to "debate" this on air last night.
Unfortunately, the debate ended before I was able to adequately respond to some of Ericsson's claims, and so this is a post to do just that - respond, put the sports science side of the debate across. I address the article to Ericsson somewhat tongue-in-cheek, and I don't mean to appoint myself on behalf of sports science, but the truth is that someone has to point out that the books, the popular media, and Ericsson are misrepresenting the evidence (either deliberately or ignorantly). And besides, Ericsson did ask in the radio interview (see below).
The debate was a glorious seven minutes long (I was told it would be much longer), and it involved two opportunities for Ericsson to state his case, and two for me to try to explain the physiology of elite athletes. Going in, I was under the impression we would debate the points, but that never really happened, mostly because I didn't think it was going to be cut short at 7 minutes.
You can listen to the podcast here. Just click "Listen Now" The interview portion starts at 9:00, as the section before is an interview with Chrissie Wellington (this provides some context for some of my comments in my first response).
A stunned reaction
I was, throughout the interview, stunned at what I was hearing. And it's not as though I'm new to this particular debate - I've recently written two review articles on this topic with a colleague of mine (a geneticist, because unlike Anders Ericsson, I don't like the idea of commenting about a field that I'm not an expert in - he's a psychologist, but he was throwing physiology around with abandon, as you'll hear and read later). These articles will be published in peer reviewed journals later this year, I'll let you know when. There are also the two articles (PART I and PART II) that I wrote here on The Science of Sport last year, and then I presented on this at the UK Sports and Exercise Medicine conference in London last November.
I have also read the books - Bounce, Outliers, so in theory, I've heard it before. But I was just absolutely stunned that Ericsson was saying some of the things he did - you can hear this in my reaction in the podcast as I start my response to both questions! What he says is just ludicrous, empty and baseless, and I can only think he's misinformed, or has some other agenda to push. Maybe he is writing a book...
Truth is, you don't even need research, you just need common sense and a tiny bit of experience with elite athletes in training groups. For example, if any of you have ever run with a training group, you have seen and felt the reality of "individual responses" to training - you know that 1,000 hours of identical training will not produce an identical result in ten different people. There are examples all over the place that show that practice is not sufficient for elite performance, and there are as many examples of athletes who have succeeded on far, far less than this (there are even cases in chess, where, dare I say it, performance is a little less complex because there's no risk of overtraining, injury, etc)
Also, Ericsson's theory that it is the training done during the adolescent years that matters is not only wrong (look how many talented young athletes fail at senior level despite accumulating far more hours than their peers by the age of 18, and how many endurance athletes only take up the sport in their 20s and become world class in a few years despite zero training when adolescents), it's also very irresponsible, because it compels parents, teachers and coaches to start training young athletes too soon and that's detrimental to the person (see Cote et al for review).
The statements - no scientific evidence showing that genes or physiology limit performance?
In his second response in the podcast, Ericsson makes the following statement in response to my argument that the scientific evidence suggests without doubt that elite athletes and champions are BORN AND MADE:
"I would argue here, and reading all the reviews, and we've had reviews where every scientist from the exercise physiology field and sports psychology. And I find it kind of remarkable that Ross is making these claims because I've never seen them made in print in any peer-reviewed publication" - 13:55 in the podcastHe goes on to say the following:
"I have to say that I'd be very interested to see Ross finding any scientific studies that support the kind of claims that he was making at the beginning of the programme" - 15:01 in the podcast
Outside the scope of knowledge - don't tread where you shouldn't unless you have a guide
Before I continue, just have to mention that I have provided a list of peer reviewed publications as references at the end of this post - they are both review articles (Ericsson made the claim that he's read ALL the reviews - clearly he's missed these ones), and they are original research studies that show the importance of genetic factors and physiological variability to training. Clearly, he's never read these either. I'd excuse this on the basis that Anders Ericsson is a psychologist, so one would not expect him to have a firm grasp of sports science, performance, physiology and the genetic literature, but the fact of the matter is that he's making claims in those fields, so therefore it's fair-game to challenge his knowledge and understanding of the literature and the sports science performance fields.
And I must just make this point - I don't for a second think we should create intellectual "silos" where you can ONLY comment on your field. I think that would be foolish because it's the process of thinking, the scientific approach to a question that matters more than the actual content. And I'd like to think that the biggest advances in our understanding often come from thinking outside the "constraints" of what we know, and by integrating research from different fields by different experts.
In other words, someone may be trained as a physiologist, but it's their application of the scientific approach, allied to some small physiological understanding, that may allow them to contribute to the field of biomechanics of barefoot running, for example. Key to this are two things: a) you must be dilligent about doing your research, and b) find someone who IS an expert to assist. That's why when in writing review articles on talent vs training and elite performance, I partnered with a geneticist (Malcolm Collins) who does understand the field at the depth that is required to put scientific statements out there. Similarly, I'm now doing barefoot running research, where my interest is the physiology, but we have a team that includes an engineer and biomechanist, so that I don't have to tread where I'm not capable.
Responding to Ericsson - one example of "limited" physiological response to training
In any event, during the debate, I tried to respond to Ericsson, and there are four things that I think are essential to understand here:
- It is true that genetic "proof" has yet to be provided. But elite sporting performance is too complex, and genetic factors too varied to ever "prove" the link Ericsson seems to require. Consider this: height is a pretty straightforward characteristic, and it's known to be highly heritable (tall parents = tall children). In fact, 80% of the variance in height is known to be genetic. However, studies have found that it takes an astonishing 300,000 genetic variants to account for only 45% of this variance. That's just height - how much more then would it take to explain something as complex as sports performance?
The reality is that the field of genetics is young, and with time, more evidence will emerge. But there's a massive difference between something being "proven" (where I agree with Ericsson) and saying that it is absent (which is what he implies). Genetic evidence is not absent - most physiological factors that are known to limit performance have been associated with genes (including injury risk, aerobic capacity, muscle fiber type), and others can be easily related to heritable factors (think height for basketball, limb proportion, bone mass etc). When Ericsson suggests there is no evidence, it is because he is ignorant of the evidence.
- Ericsson's own work disproves his theory - his studies have tried to explain performance level as a function of training, yet research he has been involved in shows that only a very small part of performance can be explained by practice. Only 28% of the variance darts performance is explained by the number of hours practiced! That's astonishingly low, and it means that time spent in practice is a very poor predictor for performance. The question you should be asking is what accounts for the other 72%, and could some of it be innate? It's definitely enough to throw out the deliberate practice, 10,000 hour theory, because Ericsson is clearly predicting that most (or all, in some of his articles) of performance is explained by training.
- You cannot prove that practice is necessary AND sufficient to produce champions or elite performers based on retrospective studies. They're weak, because there are so many other ways to explain the findings. For example, Ericsson's famous violin study showed that the expert performers did the most practice, and he concluded that the practice turned them into experts. However, it's equally possible, in this study design anyway, that the children with the innate violin ability were encouraged by others and their own success to practice more. Retrospective studies are poor ways to show that practice makes perfect. You have to do prospective studies.
- Prospective studies have been done. Most notably, Bouchard published a study in 2011 (reference below) in which he found that the response in VO2max (a measure of aerobic capacity and adaptation to training, and ultimately performance) of a large cross-section of the population to a standardized training programme was enormously varied. Some individuals improved by less than 5%, others improve by 30%. And here's the key point - it is possible, using genetic techniques, to identify which genetic polymorphisms (think of them as variants of genes) are responsible for this huge difference.
It turns out that Bouchard's work has provided some pretty important findings:
- About 50% of an individual's starting VO2max and 50% of the "trainability" in VO2max is heritable
- 21 Genetic polymorphisms have been associated with 50% of the training response to VO2max
- If a person carries NINE OR FEWER of these genetic variants, then they are low responders and improve VO2max by only 200 ml/min.
- If a person carries NINETEEN OR MORE of these variants, then they are high responders and improve VO2max by over 600 ml/min
Further, as one of you commented on Facebook yesterday, biomechanical factors such as the muscle's moment arm exert huge effects on performance, and so characteristics that we are born with determine the level of performance that we can attain. I would point out the most obvious example of this is basketball, but there are countless others.
A pointless polarization of the debate
Ultimately, however, the idea that elite sporting performance can be explained by one factor is foolish. That's why when Ericsson makes the claims he does, in the field of physiology, it's so absurd, and potentially damaging because people believe it at face value, and they implement sports systems and strategies that buy into this flawed concept. It's quite clear, from what we observe in athletes, what we study in laboratories, what we know from geneticists, that there is a significant contribution of all kinds of factors to performance. Physiology matters, but so does practice. Psychological factors are crucial, but so too are financial and economic considerations.
The dominance of Kenyan runners, for example, will never be found to be due to ONE factor. Those who are looking solely at genes are doomed to failure, but so are those who want to say that it's purely an altitude, diet, socio-economic, lifestyle, or incentive-driven phenomenon. All these factors contribute, and the environment interacts with the genes to produce a champion. I've said this before, but training should be defined as the realization of genetic potential.
Every single person improves as a result of training - some, as Bouchard has shown, improve by very little in a variable like aerobic capacity. Perhaps they are better suited to skill-based sports. Some improve enormously, and those who do are more suited to endurance sport. Then there is injury - this is vital and completely overlooked. We know that certain genes are associated with different performance characteristics, and there are genes that are associated with injury. Some people will never even reach 10,000 hours because they are susceptible to injury at five hours per week of training and cannot do more - they'd need 40 years to get good enough if that's all it took.
So the point I'd like to conclude with, before I list some more points for Anders Ericsson to consider, is that we should not polarize the debate. We should recognize that there are many paths to elite performance, and that a one-size or one-number fits all approach is foolish. We should learn what we can from those who succeed, including that they are dedicated and practice a lot, which is obvious. And we should learn why people fail. And we should avoid generalizations and simplifications that help us sell books to motivate people.
Training is the realization of genetic potential - practically, that means that every single one of you reading this, discussing this, can improve through training. That's the motivation. But will we all become Olympic caliber athletes in any sport we choose? Keep dreaming. The world is not flat, Prof Ericsson. Please stop telling people it is...
By way of an "Appendix", here is a little more on Ericsson's views, because I don't want to take him out of context in a 7 minute radio interview...
Ericsson's first entry into this field was his work looking at skill acquisition in activities such as music - his seminal study of violinists showed that expert performers engaged in at least 10,000 hours of training whereas those violinists judged merely as "good" or "average" did about 8,000 and 5,000 hours respectively.
But then Ericsson moved beyond education and skill acquisition and began to tackle sport. He wrote a review article in the New York Academy of Sciences Journal in 2009, in which he states the following:
"the distinctive characteristics of exceptional performers are the result of adaptations to extended and intense practice activities that selectively activate dormant genes that are contained within all healthy individuals’ DNA." - Ericsson et al, NYAS, 1172: 199-217, 2009So in other words, he is now going to tackle genetics. He is saying (and I want to be careful here about taking this out of context), that exceptional performers become exceptional because they practice, and this training activates dormant genes, and these genes are present in ALL healthy individuals' DNA - his word, my emphasis.
Right, so this is fine, if he sticks to "performance" in skill-based activities. I would disagree with him - studies on chess show clearly that some people get good very quickly, others never improve to Master level no matter what they do. The same is true of darts, tennis, golf, any activity. But nevertheless, let's assume that he is referring to his study on musicians and things like mathematical ability.
But he doesn't stop there. He then tackles physiology, and writes the following in the same paper:
"From this evidence it would appear that VO2max/kg (aerobic capacity) would not be a good candidate for a factor that was constrained by heredity"This comes from a section in the paper where Ericsson, a psychologist, tackles the PHYSIOLOGY of elite performance and comes to this incredible conclusion that there is no evidence that aerobic capacity is constrained by genetic factors. If you read the study, you will discover that Ericsson arrives at this conclusion based on THREE studies - one review, and two other studies, one of which actually finds the opposite to what he concludes.
I have explained above the recent work that shows clearly that genetic factors influence VO2max, and admittedly, this review precedes that series of studies. But there were still others that had found a) huge inter-individual differences between people in response to the same training and b) accounted for large parts of VO2 as being heritable.
Ericsson's approach to the physiology side of this argument is simply not good enough when physiologists can cite dozens of physiological systems or factors that are known to affect performance, and when geneticists can show associations between these systems and our genes.
And yes, I agree that this area is not yet developed - it's so "young" a field that it will take time to understand the genetic complexity. But even here, there's a difference between something being absent and something being proven. Neither side will "prove" their argument, but I think it's pretty clear that evidence shows conclusively that BOTH genes and training make champions.
And finally, here are some references that Ericsson may have missed:
Duffy L, Baluch B. Dart performance as a function of facets of practice amongst professional and amateur men ana women players. Int J Sport Psychol. 2004;35:232-245.
Vaeyens R, Güllich A, Warr CR et al. Talent identification and promotion programmes of Olympic athletes. J Sports Sci. 2009;27:1367-80.
Elferink-Gemser MT, Jordet G, Coelho-E-Silva MJ et al. The marvels of elite sports: how to get there? Br J Sports Med. 2011;45:683-4.
Phillips E, Davids K, Renshaw I et al. Expert performance in sport and the dynamics of talent development. Sports Med. 2010;40:271-83.
Huijgen BC, Elferink-Gemser MT, Post WJ et al. Soccer skill development in professionals. Int J Sports Med. 2009;30:585-91.
Gobet F, Campitelli G. The role of domain-specific practice, handedness, and starting age in chess. Dev Psychol. 2007;43:159-72.
Gibbons T, Hill R, McConnell, A. et al. The path to excellence: A comprehensive view of development of U.S. Olympians who competed from 1984-1998 United States Olympic Committee. 2002.
Baker J, Côté J, Deakin J. Expertise in Ultra-Endurance Triathletes Early Sport Involvement, Training Structure, and the Theory of Deliberate Practice. J Appl Sport Psychol. 2005;17:64-78.
Oldenziel K, Gagne F. Factors affecting the rate of athlete development from novice to senior elite: How applicable is the 10-year rule. Athens 2004: Pre-olympic Congress Sport Science Through the Ages: Challenges in the New Millennium. Athens. 2004.
Hodges NJ, Starkes JL. Wrestling with the nature of expertise: A sport specific test of Ericsson, Krampe and Tesch-Römer’s (1993) theory of “deliberate practice”. Int J Sport Psychol. 1996;27:400-24.
Helsen WF, Starkes JL, Hodges NJ. Team sports and the theory of deliberate practice. J Sport Exerc Psychol. 1998;20:12-34.
Bullock N, Gulbin JP, Martin DT et al. Talent identification and deliberate programming in skeleton: ice novice to Winter Olympian in 14 months. J Sports Sci. 2009;27:397-404.
Roescher CR, Elferink-Gemser MT, Huijgen BC et al. Soccer endurance development in professionals. Int J Sports Med. 2010;31:174-9.
Vaeyens R, Lenoir M, Williams AM et al. Talent identification and development programmes in sport : current models and future directions. Sports Med. 2008;38:703-14.
Tucker R, Collins M. Athletic performance and risk of injury - Can genes explain all? Dialog Cardiovasc Med. In Press.
Collins M, Raleigh SM. Genetic risk factors for musculoskeletal soft tissue injuries. Med Sport Sci. 2009;54:136-49.