Past Athletic Performance Doesn’t Guarantee Future Results

My favorite rags-to-athletic-riches story is Reid Coolsaet, a Canadian ultrarunner from near where I grew up who’s a few years younger than me. As a high-schooler, he was mediocre. In university, he started to improve—and didn’t stop. Starting in 1998, when he turned 19, his annual 5K bests were: 15:56, 15:16, 14:39, 14:28, 14:12, 13:53, 13:31, 13:23. By that time, in 2005, he was representing Canada at the world track and field championships, and he went on to run (and place in the top 30) in two Olympic marathons.

I’ve always loved Reid’s success because of how unlikely it seemed, given his humble start. But a new study suggests that I may have been misjudging him, and falling into a common trap in my assumptions about what’s “typical” for athletes who ascend to the highest levels of the sport. Published in the journal Sports Medicine by a team led by Arne Güllich of the University of Technology Kaiserslautern in Germany, the new review argues that athletes who succeed in junior age categories are, for the most part, completely different from those who succeed in adult competition. For anyone coming to the end of their high school or university sports career and assuming that they’ve maxed out their talents, that’s a pretty important message.

There has been plenty of research on talent identification and development over the years, and plenty of debates about ideas like the 10,000-hour rule. Notably, Güllich led a meta-analysis back in 2021 that assessed the training histories of top juniors (usually defined as under 20, though it varies from sport to sport) and top seniors. The overall pattern was that top juniors tended to pick a sport early, practice it to the exclusion of other sports, and progress rapidly. But those who made it to the top as seniors had precisely the opposite pattern: they had spent less time training in their main sport and more time playing other sports as kids, and they made slower initial progress in their main sport.

In the new paper, Güllich and his colleagues ask a more fundamental question: when we compare top juniors to top seniors, are we even comparing the same people, or are they two different populations? To find out, they pooled the results of 110 prospective studies with 38,000 elite junior athletes to figure out how many of them went on to achieve a similar level of success (national, international, or world medalist) as seniors. And conversely, they pooled 79 retrospective studies with 23,000 elite senior athletes to find out how many of them had previously achieved similar levels of success as juniors. A wide variety of sports were represented, with the biggest category being Olympic sports like track and field, cycling, and swimming.

The results (which are free to read here) are clear: most successful juniors don’t become successful seniors, and most successful seniors weren’t successful juniors. One example: 89 percent of international-class under-17 and under-18 athletes never reach that level as seniors, and 83 percent of international-class seniors didn’t make it to international class at the under-17 and under-18 level. To put it another way, these junior and senior populations are 93 percent different and just 7 percent the same.

These results undermine both of the main theories of how outliers get so good—i.e. that it’s all about natural talent, or that it’s all about how much and how effectively you practice. Both theories imply that how good you are as a junior will predict how good you are as a senior, and that success at both levels is predicted by the same factors. Instead, Güllich argues that what predicts junior success—a focus on training to maximize immediate performance—might actually work against the prospects for sustained long-term improvement. If Reid Coolsaet had spent his high-school and university years tapering for every dual meet, he might have won a few more ribbons but probably wouldn’t have built the base that enabled him to keep improving for so long.

Of course, the boundary between the present and the future gets hazier as you get older. Another recent paper, this one published in the International Journal of Sports Physiology and Performance by Lidia Alejo of the Universidad Europa de Madrid and her colleagues, put 65 national and international-class under-23 cyclists through a battery of tests, then followed up a few years later to see which tests were best at predicting the 16 cyclists in the group who successfully turned professional. There were a bunch of lab tests, including VO2 max and ventilatory thresholds, various strength tests, body composition assessments, and an eight-minute time trial.

The single best predictor of future pro career was the time trial. Many of the factors they tested had a little bit of predictive power when combined in an overall statistical model, but knowing a rider’s VO2 max or maximal squat power in isolation wasn’t very useful. And really, the time trial itself is an overall model that combines all these factors. It reminds me—yet again—of what sports scientist Carl Foster once said about DNA testing for athletic talent: “If you want to know if your kid is going to be fast, the best genetic test right now is a stopwatch. Take him to the playground and have him face the other kids.”

There’s a similar point that applies to all of us, even those on the wrong side of the aging curve: proxy measures of fitness, whether it’s heart-rate variability or lactate levels or smartwatch-calculated VO2 max values, are never the same as the thing itself. If you really want to know how your training is going, you’ll eventually need to race.

One final point: Alejo’s results would almost certainly have been different if the cyclists were, say, 16 years old. At that point, differences in physical maturity would have a significant influence, as would their training history. But by the time you’re 20 (the approximate average age of cyclists in her study), those factors have mostly equalized. By that point, if you’re going to be a world-beater, the signs are probably showing, assuming you’ve been training for at least a few years. In that sense, Reid Coolsaet is still an outlier, because even at 20 there’s no algorithm in the world that would have predicted his future success. That’s probably what I like so much about his career progression—the reminder that, no matter what the odds, there’s always a chance.

For more Sweat Science, join me on Twitter and Facebook, sign up for the email newsletter, and check out my book Endure: Mind, Body, and the Curiously Elastic Limits of Human Performance.