Annual Age-Grouping and Athlete Development: A Meta-Analytical Review of Relative Age Effects in Sport

38 studies, 253 independent samples

Youth to elite athletes across 14 sports, 16 countries; moderators = age category, skill level, sport context

First RAE meta-analysis: 38 studies (1984-2007), 253 independent samples, 14 sports, 16 countries; random-effects pooling of birth-quartile odds against expected population birth distributions. Quasi-natural-experimental (birth date is arbitrary relative to the cut-off) but purely descriptive of roster composition — it measures SELECTION, not skill acquisition. Sensitive to population birth-rate assumptions.

38 studies, 253 samples, 14 sports. The relative age effect is robust but small, peaking in 15-18y males at representative level. IMPORTANT: the RAE is a CALENDAR artifact of the selection cut-off date — it is essentially uncorrelated with genotype and must NOT be read as heritable maturational timing (a separate, partly heritable variable). The RAE persists within early- and on-time-maturing subgroups. It is an argument against chronological age-grouped talent identification.

None — this is a calendar/selection artifact, not a genetic one. Do not conflate relative age (cut-off date) with biological maturity timing (partly heritable).

Replicated and extended: Smith et al. 2018 (Sports Med) female-sport RAE meta pooled OR 1.25 (95% CI 1.21-1.30), smaller in females than males. Senior-elite reversal reported repeatedly (see Brustio 2023).