90% of Elite Youth Performers Never Become Elite Adults • Sport for Life

For years, the Long-Term Development framework has made a case that runs against the grain of how most youth sport actually operates: don’t specialize early, embrace multisport participation, let development happen gradually, and stop assuming your best 12-year-old will be your best adult.

The response from skeptics has been predictable. That’s fine in theory, but elite sport requires early focus. The kids who make it started early and trained hard. You can’t get to the top without putting in the hours from a young age.

A December 2025 paper in Science—one of the world’s most prestigious research journals—just delivered a comprehensive answer: the skeptics are wrong.

The Study

Researchers synthesized data on more than 34,000 adult international top performers across multiple domains: Nobel laureates, the most renowned classical music composers, Olympic champions, and the world’s best chess players.

They asked two questions that hadn’t been systematically investigated at this level before:

Are exceptional performers at young ages and at peak performance age largely the same individuals?
Do the predictors of exceptional youth performance also predict exceptional adult performance?

The answers were clear—and consistent across every domain they studied.

Finding 1: They’re not the same people.

World top-10 youth chess players and world top-10 adult chess players are nearly 90% different individuals.

Top secondary students and top university students are nearly 90% different people.

International-level youth athletes and international-level adult athletes are nearly 90% different individuals.

The kids who peak early are rarely the ones who peak highest. Early exceptional performance and later world-class performance belong to largely discrete populations.

Finding 2: The predictors flip.

Here’s where it gets interesting. The researchers found that early exceptional performance and adult world-class performance don’t just involve different people—they’re predicted by opposite developmental patterns.

Early exceptional performance is associated with:

Extensive discipline-specific practice
Little or no multidisciplinary practice
Fast early progress

Adult world-class performance is associated with:

Limited early discipline-specific practice
More multidisciplinary practice
Gradual early progress

Read that again. Across Nobel laureates, Olympic champions, elite musicians, and top chess players, the pattern holds: the adults who reach the highest levels practiced less in their specific discipline during their early years and did more across multiple disciplines. They progressed more slowly early on.

The early specializers peak early. The samplers peak highest.

Finding 3: Most world-class performers were behind their peers early on.

The paper notes that “most top achievers demonstrated lower performance than many peers during their early years.” And across the highest adult performance levels, peak performance is negatively correlated with early performance.

The late developer isn’t the exception. At the world-class level, the late developer is the norm.

What This Means for Sport

None of this will surprise anyone who’s worked with the Long-Term Development framework. The principles have been clear for years: early sampling builds a broader foundation, early specialization creates early peaks but limits long-term ceiling, and the developmental pathway to elite adult performance looks different from the pathway to elite youth performance. Two of Sport for Life’s Seven Guiding Principles of Long-Term Development in Sport and Physical Activity—Physical Literacy to Multisport to Specialization and Excellence Takes Time—have made exactly this case. Now the evidence backs them up at the highest levels of human achievement.

But there’s a difference between “the framework says this” and “a synthesis of 34,000 world-class performers across multiple domains confirms this in Science.”

The implications are direct:

For talent identification: Stop assuming your best young performers are your future stars. The research says 90% of them aren’t. The kids who will eventually reach the highest levels are more likely to be in the middle of the pack right now, accumulating diverse experiences.
For program design: Multisport participation isn’t a detour from development—it’s the pathway the world’s best actually took. Programs that push early specialization are optimizing for early results at the cost of long-term potential.
For coaches and parents: Gradual progress isn’t falling behind. It’s what world-class development actually looks like. The pressure to accelerate—to get ahead, to specialize, to lock in an advantage—produces early peaks, not high peaks.
For selection systems: Elite academies and development programs that select top-performing youth and accelerate their discipline-specific training are, according to this research, selecting the wrong kids and training them the wrong way. They’re building programs around the 10%, not the 90%.

The Harder Conversation

The research explains something uncomfortable: why early-specialization programs keep producing evidence that early specialization works.

If you select the best 12-year-olds and give them intensive discipline-specific training, they’ll get better. They’ll outperform peers. They’ll win at the youth level. The program will look successful.

But the kids who would have eventually reached the highest adult levels? They’re the ones who didn’t make the cut, who were told they weren’t elite material, who moved on to other things or other sports or no sport at all. The system is optimized to identify and develop early peakers—and to filter out the eventual world-class performers before they get there.

The authors put it plainly: “Many elite schools, universities, conservatories, and youth sport academies around the world typically aim to select the top-performing young people and then seek to further accelerate their performance through intensified discipline-specific practice.”

That approach, the evidence now shows, is backwards.

What Comes Next

The paper proposes three hypotheses for why this pattern holds—why breadth beats depth and gradual beats fast. The “search-and-match hypothesis” suggests that multidisciplinary experience helps people find the domain they’re best suited for. The “enhanced-learning-capital hypothesis” suggests that diverse practice builds transferable skills and learning capacity. The “limited-risks hypothesis” suggests that slower early progress reduces burnout, injury, and dropout.

These will be tested and debated. But the core findings are robust: across domains, across 34,000 performers, at the highest levels of human achievement, the developmental pattern is consistent.

Long-Term Development works. The evidence confirms it.

Learn More

If you haven’t explored the Long-Term Development framework, start with Long-Term Development in Sport and Physical Activity 3.0—the resource for building programs focused on long-term development, not short-term wins. Sport for Life’s Seven Guiding Principles of Long-Term Development in Sport and Physical Activity will be available to the public in 2026.

Questions about applying Long-Term Development framework in your organization? Reach out to Sport for Life.

Reference

Arne Güllich, Michael Barth, David Z. Hambrick, and Brooke N. Macnamara, “Recent Discoveries on the Acquisition of the Highest Levels of Human Performance,” Science 390, no. 6779 (2025): eadt7790, https://doi.org/10.1126/science.adt7790.