Predicting sport superstars: How well are we doing?
In 2004, the DC United soccer club made United States history by signing the youngest athlete ever to a professional sports contract. At just 14-years-old, Freddy Adu--born in Ghana and moved to Rockland, Maryland when he was 8--was dubbed the next Pelé and the future of U.S. soccer.
Three months later Freddy again made history by being the youngest player to ever score a goal in the MLS. That season he would go on to play in 30 games, scoring 5 goals and 3 assists--a solid start for someone who should have been completing the 8th grade at the time.
However his development as a soccer player wasn't panning out as expected. After two more seasons with DC and a brief stint with Real Salt Lake, he boarded a plane to Europe where he spent the next five seasons being loaned between teams in Portugal, Greece, and Turkey--but never finding success he was looking for.
In 2011, Freddy returned to the MLS for a few seasons with the Philadelphia Union before again trying his luck abroad in South America. Same result.
Four years later, the former can't-miss-kid signed a deal with the Finnish club Kuopion Palloseura. After his signing, Marko Rajamaki the team's head coach was quoted as saying,
I can't say what went wrong in his career. He told me he has been ill-advised in the past. Maybe the problem is that everybody told him he was the saviour who would revolutionise football.
My intention is not to pick on Freddy. I'm sure he's a great dude and despite his lack of success is still one hell of an athlete. But what I find unique about Freddy's story--unlike other prospect busts--are two things:
First, there is no logical explanation for why his career didn't unfold as it was predicted. There was no injury. No drug abuse. No scandal. Nothing to point to where you could say "Ah. That makes sense...bad luck."
The superstardom just didn't come.
Second, Freddy wasn't just supposed to be good. He was supposed to be one of the all-time greats. He was such a rare young talent that it seemed impossible for him not to develop into a soccer icon. Yet while he has continued to play professionally for over 10 years, his career so far could best be described as average.
Was Freddy a casualty of the American sports-hype machine? Or are we just that bad at predicting talent?
How good are we and predicting athletic success?
The natural argument to the Freddy Adu story is Sidney Crosby or Andre Agassi--athletes that were tapped for greatness at a young age and succeeded in reaching the pinnacle of their sports. But looking to case studies of success or failure doesn't really examine the issue or answer the question of how good we are at foreseeing athletes' futures.
In sport, it’s easy to get focused on the athletes at the top. However these athletes are exceptional not only in their talent, but in their rarity. They represent such a small piece of the pie that looking to them for insight leads us astray. They are the outliers. The statistical freaks.
It's easy to spot the Freddy Adus and Lebron James' of the world. And you know something? With a couple of exceptions, we're fairly good and predicting them to rise to the top of their class. But that breed of athlete is so exceptional that we learn little about how well we're identifying and developing the other 99% of the athletes out there. To do that we must zoom out to see what's really going on.
A little research
A 2013 study on young soccer players tracked the turnover rate of the German national junior programs and found that only one-third of those who represented Germany at the junior level reappeared on the senior national team. They also found that the earlier an athlete made their debut in the national program, the less likely they were to advance to the most elite German soccer leagues. (1)
Most young players selected at a particularly early age were replaced within a short time by others who had developed more prosperously outside the youth academies and national U-teams. - Arne Güllich
Another German study on cyclists looked at the careers of 8004 athletes between 1980 and 2004 and found that only 34% of cyclists that participated in junior World Championship events went on to compete in elite competitions. (2)
In 2014, a longitudinal study tracked the careers of 395 Portuguese male and female athletes born between 1974 and 1981 who competed internationally in soccer, judo, swimming and volleyball. Again, they found that about one third of international pre junior athletes reappeared to compete as seniors. (3)
Another longitudinal study in 2011 tracked 242 elite swimmers through the ages of 12 to 18 looking at performance times, when they showed the greatest improvement, and if they could use their data to predict future performance. They found that the younger the swimmers were, the more variance they showed in their performance times and that age 16 was when performance began to steady and they could begin to accurately predict their performance as adults. (4)
To explain some of this Roel Vaeyens and his colleagues published a paper in 2008 entitled Talent Identification and Development Programmes in Sport, emphasizing that adolescents aren't guaranteed to retain the their athletic qualities as they transition into adulthood. And many of the distinguishing characteristics of elite athletes aren't apparent until late adolescence, making early identification totally unreliable.
They go on to discuss that physical development is highly unstable and multidimensional. The muscular system, bone growth, tendon and ligament elasticity, cardiovascular development, and motor coordination don't all develop in unison. Despite what we would like to believe, talent, athleticism, potential--whatever you want to call it--is not a singular or measurable quality. It is the output of multiple systems and abilities that are constantly evolving and interacting throughout the athlete's development. (5)
But elite performance can be influenced by social, psychological, economical, and geographical factors as much as physical ones. The Swiss owe their success in skiing to their geographical proximity to mountains as much as their leg strength. New Zealand produces the best rugby talent in the world not because Kiwis are born with thicker necks or bigger hands, but because of their cultural obsession with the sport.
To make a future prediction based on a snapshot in time is not only short-sighted, but it fails to consider what went in to produce that performance in the first place. It's like saying a 12 second 100m dash isn't very impressive without realizing that the 15-year-old kid who ran it didn't have any shoes on.
Performance is obvious, but foreseeing performance needs context.
So what are we good at?
Just because there are some flaws in our ability to spot future top performers doesn't mean we are at a total loss. Recent research into what successful athletes have in common has exposed some interesting things.
Claude Bouchard, a genetic researcher from Baton Rouge, Louisiana is at the forefront of research on trainability: the idea that certain individuals respond better and faster to training than others. The idea of trainability is not new. Anyone involved with training athletes can testify to the phenomenon that if you apply the same training plan to a large group of athletes, some will just respond better than others. But we’re starting to see real data for this phenomenon. Recent research has emerged out of Bouchard’s lab showing that an athlete’s trainability may be genetically predicted (6, 7).
Beyond physical talent
Although we usually look for physical indicators of talent and ability, another area that can not be ignored is the mind.
For example, self-regulation is the process of internally setting performance objectives and then continually monitoring and adapting practice to move closer to those goals. It’s a coach's way of saying "paying the hell attention and taking ownership."
There is a significant body of research showing elite athletes score higher on self-regulation abilities, indicating they take ownership over their own success and failure, engage in more reflection, and have higher levels of effort. As a result they are able to make progress faster and more consistently than others. (8-10)
Athletic potential spans far beyond just an athlete's biological abilities. Several recent studies have shown that the size of the town where an athlete grew up can have an impact on their success. (11-13) According to this research, athletes from small to moderate-sized communities are at an advantage because they provides more training and talent development opportunities than larger cities. In looking at where pro athletes originate from, we in fact see an overrepresentation of athletes from smaller communities playing in the NHL, NBA, PGA, and MLB.
The consequence of an assumption
Think about how we've organized our athlete development system. Starting from the beginning of a young athlete's career, we cluster the top performers together with the expectation that they will remain the top performers as they develop.
But the idea that the best young athletes are most likely to become the best adolescent or adult athletes is nothing more than an assumption. While this certainly can happen, it isn't the rule. And we definitely aren’t seeing it in the data.
But I understand the logic of the system. I get that we want the best athletes together to foster an environment that grows and nurtures talent. I’m not arguing for the banishment of all travel programs or all-star teams.
My issue is in the timing.
A performance-based system of talent development that is by definition elitist and exclusionary should, I think, not be instituted until later in the youth sports tract.
I just have a problem calling any human being younger than 10-years-old elite in anything. I think it sends the wrong message both to the kid that makes the team, and the kid that doesn't.
And just to be clear, this isn't coming from some Nanny-state notion of participation trophies where no-one's feelings get hurt. Adversity, struggle, and learning to deal with defeat should be part of the youth sport experience. I think it's critical that we force our kids to deal with difficult times. But we need to be careful when it comes with messing with kids' identities.
And that is the real problem.
By structuring youth sports in this way, we fail to capture late developers and athletes who may have extremely high potential, but for whatever reason can't realize it at 9-years-old. For every kid that gets selected to the team, two get cut--relegated to "B" status, sent home and told to simply work harder. One of those kids might have the desire to keep persisting, the other might walk away with the idea that "I'm just not good, and that's that."
I know this because I've lived it, I've seen it, and I've coached kids that are walking around with labels of their ability stamped in their consciousness.
If we've learned anything here--it's that although we seem to be good at recognizing top performance, we're not all that good at predicting it. And this ability to predict gets worse and worse the younger the athlete gets.
As much as we love to stand behind the glass and comment on how good this kid or that kid is--the reality is that our view from the stands isn't a peek into their futures.
Accurate prediction just isn't reliable at young ages, so let's just give it a rest, focus on the things that matter, be great teachers, and keep kids in the game. It's probably the single best thing that can be done for their future.
- Gullich A. Selection, de-selection and progression in German football talent promotion. European journal of sport science. 2013 Nov 19. PubMed PMID: 24245783. Epub 2013/11/20. Eng.
- Schumacher YO, Mroz R, Mueller P, Schmid A, Ruecker G. Success in elite cycling: A prospective and retrospective analysis of race results. Journal of sports sciences. 2006 Nov;24(11):1149-56. PubMed PMID: 17175613. Epub 2006/12/21. eng.
- Barreiros A, Cote J, Fonseca AM. From early to adult sport success: analysing athletes' progression in national squads. European journal of sport science. 2014;14 Suppl 1:S178-82. PubMed PMID: 24444203. Epub 2014/01/22. eng.
- Costa MJ, Marinho DA, Bragada JA, Silva AJ, Barbosa TM. Stability of elite freestyle performance from childhood to adulthood. Journal of sports sciences. 2011 Aug;29(11):1183-9. PubMed PMID: 21777055. Epub 2011/07/23. eng.
- Vaeyens R, Lenoir M, Williams AM, Philippaerts RM. Talent identification and development programmes in sport : current models and future directions. Sports medicine (Auckland, NZ). 2008;38(9):703-14. PubMed PMID: 18712939. Epub 2008/08/21. eng.
- Bouchard C. Genomic predictors of trainability. Experimental physiology. 2012 Mar;97(3):347-52. PubMed PMID: 21967902. Epub 2011/10/05. eng.
- Epstein D. The Sports Gene: Inside the science of extraordinary athletic performance. New York, New York: Penguin Group; 2013.
- Kitantas A, Zimmerman, B.J. Comparing self-regulatory processes among novice, non-expert, and expert volleyball players: a microanalytic study. Journal of Applied Sport Psychology. 2002;14:91-105.
- Jonker L, Elferink-Gemser MT, Visscher C. Differences in self-regulatory skills among talented athletes: the significance of competitive level and type of sport. Journal of sports sciences. 2010 Jun;28(8):901-8. PubMed PMID: 20544490. Epub 2010/06/15. eng.
- Cleary T.J. ZBJ. Self-regulation differences during athletic practice by experts, non-experts, and novices. . Journal of Applied Sport Psychology. 2001;13(2):185-206.
- Baker J, Shuiskiy K, Schorer J. Does size of one's community affect likelihood of being drafted into the NHL? Analysis of 25 years of data. Journal of sports sciences. 2014 Apr 16:1-6. PubMed PMID: 24738661. Epub 2014/04/18. Eng.
- Baker J, Logan AJ. Developmental contexts and sporting success: birth date and birthplace effects in national hockey league draftees 2000-2005. British journal of sports medicine. 2007 Aug;41(8):515-7. PubMed PMID: 17331975. Pubmed Central PMCID: PMC2465449. Epub 2007/03/03. eng.
- Cote J, Macdonald, D.J., Baker, J., & Abernethy, B. When "where" is more important than "when": Birthplace and birthdate effects on the achievement of sporting expertise. Journal of sports sciences. 2006;24(10):1065-73.