Research · Talent Identification

Your Kid's 'Talent' at Age 11 Might Just Be a Birthday — What the Research Actually Says

By Muhammad Dzulhisham · MSc Sports Coaching, NSCA-CSCS

Every year, a version of the same sorting happens on fields and courts across Singapore: a squad of ten- and eleven-year-olds gets split into an "A" team and everyone else. The kids who make the cut usually look the part — taller, faster, stronger through a tackle or a jump than the kids left on the sideline. Coaches call it spotting talent. The research on how that sorting actually works says something less flattering: a meaningful share of it is really just spotting who was born earlier in the year.

This is the relative age effect (RAE) — the tendency for children born in the early months of a sport's selection year to be systematically overrepresented in elite youth squads, and for children born in the closing months of the same year to be systematically underrepresented, purely as an artefact of the calendar cutoff used to sort them into age groups.¹ It's one of the best-documented biases in sport science, and it sits underneath a second, harder question this article is really about: how good are early "talent" labels at age 10–12 at predicting who becomes a strong adult athlete? The evidence says: not very.

The size of the effect, in real numbers

The relative age effect isn't a fringe finding from one small study. A 2009 meta-analytical review in Sports Medicine, pooling 38 studies and 253 independent samples across 14 sports and 16 countries, found the pattern reliably present across annual age-grouped sport systems worldwide² — it's the organisational habit of sorting children into single-year age bands, not any one sport's culture, that produces it.

Soccer academies provide some of the starkest numbers. A study of youth national-team selections across ten European countries, covering 2,175 birthdate records, found that 38.7% of selected players were born in the first quarter of the football selection year, against just 10.5% born in the last quarter³ — nearly four times as many players drawn from January–March as from October–December, despite births being roughly evenly distributed across the calendar. A more recent look inside England's own elite pathway found the same skew still holding at the top of the system: a 2024 study of 487 Under-18 players, 350 Under-21 players, and 396 senior professionals across the country's highest-level academies and clubs found a significant selection bias toward relatively older, earlier-born players in both the Under-18 and Under-21 squads, across every playing position.⁴

Why the biggest kid in the group gets mistaken for the best one

The mechanism isn't mysterious once it's named. Within any single-year age band, the oldest children can be close to a full year more physically and neurologically mature than the youngest — a genuinely large gap at age 10 or 11, even though it will mean nothing by adulthood. That maturity gap shows up as more strength, more coordination, more speed, and better decision-making under pressure, all of which look exactly like "talent" to a coach running a two-hour trial.

What happens next compounds the mistake rather than correcting for it. The relatively older, more physically mature child gets selected for the representative squad, which means more coaching hours, better competition, and more match minutes than the child left out. Given a year or two of that gap in investment, the selected child usually does become measurably better — not because the original selection correctly identified superior long-term potential, but because the selection itself created the advantage it was supposedly detecting. Researchers reviewing the relative age literature describe both a physical maturational-selection mechanism and this kind of self-reinforcing, opportunity-driven mechanism operating together, and note that awareness of relative age among the adults doing the selecting is one of the few things shown to reduce the bias.¹

It reverses, almost exactly, once the physical gap closes

If the relative age effect were really tracking durable talent, it should persist once every athlete reaches full physical maturity. It doesn't. The same 2024 EPPP study that found a strong selection bias toward early-birth players at Under-18 and Under-21 level found no significant birthdate skew at all among the senior professional squads — and, in a detail that undercuts the original selection logic directly, players born in the last quarter of the selection year were found to carry the highest average market value of any birth quartile at senior level, even though that difference didn't reach statistical significance.⁴

Ice hockey shows the same pattern even more sharply, in the opposite direction. A study of National Hockey League rosters from the 2008–09 through 2015–16 seasons found the familiar relative age effect among the broad player pool — but among the league's actual elite performers, the pattern reversed: players born in the last quarter of the year outscored those born in the first quarter, with the gap reaching roughly nine points per season among the highest-scoring players in the sample.⁵ The leading explanation is what researchers call the underdog effect — a relatively younger player who survives a youth system built against them has spent years competing up against bigger, more mature peers, and the competitive and psychological toughening that requires doesn't show up until the physical gap it was forged against has disappeared.

The kid who looks like the best athlete on the field at age eleven is very often just the kid born nine months earlier in the selection year — and by the time both of them reach adulthood, that gap has not just narrowed. In some sports, the data says it has already flipped.

Junior success and adult success are, statistically, two different questions

The deepest problem with labelling a child "talented" at 10–12 isn't only the relative age bias baked into who gets seen in the first place. It's that even a clean, unbiased read of who is best right now is a weak predictor of who will still be best as an adult. A 2023 systematic review in Sports Medicine, synthesising data across 110 samples and more than 38,000 junior athletes and 79 samples covering nearly 23,000 senior athletes, quantified exactly how weak that link is: 89.2% of athletes who reached international level as Under-17/18 juniors failed to reach an equivalent international level as seniors, and 82.0% of athletes who reached international level as seniors had not reached that level as juniors at all.⁶ Overall, the review found successful juniors and successful seniors were roughly 7.2% the same population and 92.8% disparate — two overlapping but largely distinct groups of people, not one continuous pipeline running from precocious child to elite adult.

This isn't a new or isolated finding, and it has already been presented directly to a Singapore audience. At the National Youth Sports Institute's Youth Athlete Development Conference, held at the Singapore Sports Hub in November 2016 under the theme "Talent < Timing < Tenacity," sport scientist Professor Arne Güllich of the Technical University of Kaiserslautern — whose research underpins much of this body of work — presented data showing that success at age 10 carries essentially zero correlation with success as a senior athlete, and put the point plainly: junior success is a poor indicator of long-term senior success.⁷ Part of why is that the two are driven by different, sometimes opposite, developmental patterns. Athletes who look most impressive as juniors often got there by specialising early and progressing fast; athletes who go on to senior elite success more often sampled multiple sports and progressed more slowly as children — the profile that looks, at age 11, the least like a "prospect."

Static talent tests miss the athletes who develop late

Part of why this keeps happening is structural. A widely cited 2008 review in Sports Medicine on talent identification and development programmes pointed out that most models used to identify talented young athletes are cross-sectional — a single snapshot of size, speed, and skill measured at one point in time — when the qualities that actually matter for adult performance are dynamic and multidimensional, changing shape as an athlete grows.⁸ A cross-sectional model built around "who is best right now" will systematically miss children who are still years away from their growth spurt, however much raw coordination or trainability they're carrying underneath the current gap in size. The review's authors argued this isn't a minor calibration problem: this kind of static, single-point talent identification is likely to exclude a real share of promising children — particularly late maturers — from development programmes altogether, simply because the model was never built to account for a trajectory that hasn't unfolded yet.⁸

What actually helps: separating maturity from merit

None of this means talent identification is worthless, or that trying to develop promising young athletes is a mistake. It means the tools most programmes still default to — a single trial, sorted mostly by who looks biggest and fastest that day — are measuring the wrong variable a meaningful amount of the time. A few approaches from the research point toward something better:

• Bio-banding. Rather than grouping children purely by birth year, bio-banding groups them by biological maturation status — using measures like predicted adult height attainment to sort players for competition, assessment, or strength and conditioning work. Sport scientists writing in the NSCA's own Strength & Conditioning Journal describe it as a way to reduce the competitive inequality created by maturity mismatches within a single chronological age group, giving both early and late developers a fairer read on their actual skill rather than their current size.⁹
• Tracking trajectory, not a single snapshot. Rate of improvement over repeated testing sessions is a fundamentally different measurement than a one-off ranking, and it's far harder for a temporary maturity advantage to fake.
• Awareness of relative age among selectors. Simply knowing which players in a squad are relatively older or younger — and correcting for it consciously — is one of the few interventions the research base actually shows reduces the bias in selection decisions.¹
• Treating an early cut as information, not a verdict. Given that over 80% of senior international athletes in the Güllich review were not international-level juniors, a child left off a representative squad at 11 is not receiving reliable information about their adult ceiling — only about where they stood, that day, against a group that may have simply been born earlier.⁶

What this means for a young athlete in Singapore

None of this is an argument against school teams, national development centres, or structured pathways — competitive opportunity has to be organised somehow, and age-grouping is a reasonable default for running a league. It's an argument against treating the result of that sorting as a verdict on a child's ceiling. A December-born child who doesn't make an academy squad at 10 is very often standing next to a January-born teammate carrying almost a year of extra physical maturity, in a system that has no mechanism built in to separate the two. The research says that gap is real, it's large enough to look like a talent difference at that age, and it is not something a parent or a young athlete did anything wrong to be on the losing side of.

The more useful question at 10–12 isn't "did this child get selected." It's whether they're developing the underlying qualities — movement competence, trainable strength, decision-making, a genuine enjoyment of the sport — that the evidence actually ties to long-term progression, regardless of which squad they're on this season. A testing-based read of where an athlete actually stands, repeated over time rather than taken once, is a far closer match to what the research says talent development really looks like than a single early cut ever was.