Calculating maturity overcomes the bias towards early-developing youth athletes

Maturation is a natural biological process through which children and adolescents progress towards adulthood. This process encompasses a range of physiological and cognitive developments.^[1]

One of the most notable landmarks within this process is peak height velocity (PHV), the maximum rate of growth in stature during the adolescent growth spurt.^[1]

In girls, PHV typically occurs between 9 and 15 years of age, whereas in boys it generally occurs between 11 and 17 years.^[1] These wide age ranges highlight the substantial inter-individual variability in maturation timing and tempo. Early maturing athletes have a greater tempo, while later maturing athletes have a reduced tempo.

Individuals of the same chronological age may be at very different stages of biological development. This variability has important implications within sport and exercise. Differences in maturation can influence performance characteristics, injury risk, training adaptation, and athlete selection.

Consequently, assessing maturation status has become an important component of talent development and identification provisions.

Implications of maturation in sport development

Maturation can be described using three key concepts. First, maturity stage refers to the level of biological development an individual has reached at a specific time. Examples include pre-PHV, circa-PHV, and post-PHV.

Maturity tempo refers to the rate at which an individual progresses through the stages of maturation.

Maturity timing refers to when maturational events occur relative to peers. Individuals may be classified as early, on-time, or late maturers depending on when developmental milestones occur.^[3]

Within sport, maturation influences several important processes, including talent identification (athlete selection and perceptions of performance / potential), talent development (provisions applied by coaches and key stakeholders), and training management (organisation of training and injury mitigation strategies).

In many sports, particularly those where physical performance contributes significantly to success, early maturing athletes may initially outperform their later maturing peers. Early maturation biases have been documented in football, rugby, ice hockey, basketball, and American football.

Early maturers often demonstrate advantages in physical tests such as strength, speed, and power. These advantages emerge during adolescence, alongside growth-related increases in stature, muscle mass, and strength.

However, it is important to recognise that the physical advantages associated with early maturation are often temporary.

As development progresses, lesser maturing athletes frequently “catch up” in physical capabilities if they remain within the development pathway. In contrast, later maturing athletes are desirable in sports such as cross-country skiing, figure skating, and gymnastics.

Both scenarios demonstrate that when athlete evaluation neglects maturation, there is a risk that potentially talented athletes may be incorrectly deselected.

Providing all athletes with the appropriate challenge and development opportunities is crucial for talent development.

Environments with a bias towards early maturing athletes may repeatedly expose late maturing athletes to physically dominant opponents. This delivers a particularly high level of challenge relative to their more mature peers. While there are positive effects such as greater resilience, a critical quality of highly successful sports people, athletes are first required to “survive” excessive overstimulation amid these atypically challenging demands.

Conversely, early maturing athletes may be under-stimulated. If they can rely on their physical prowess to excel against their less-mature teammates and competitors, they may experience insufficient development of essential sporting qualities.

Subsequently, once their less-developed peers progress through adolescence, their physical advantage is neutralised and gaps in wider sporting ability are magnified.

One strategy to address these issues is bio-banding. Bio-banding groups athletes according to maturity status rather than chronological age. This approach aims to reduce maturity-related variation within competition and training environments. Bio-banding has been widely applied in youth football and appears to provide both early and late maturing athletes with different developmental challenges and opportunities.^[2]

Differences in maturation influence how athletes respond to training. Practitioners should consider individual differences in growth and maturation when designing and implementing training programmes for young athletes.

During pre-pubertal stages of development, training adaptations are largely driven by neural adaptations, including improvements in coordination and motor control.

The landmark stage of maturation is PHV.

An athlete’s maturation tempo aligning to their maturation timing, is often accompanied by rapid growth of limbs and reduced proprioceptive awareness. An athlete’s gross motor skills can decline, portraying a short-term dip in athletic ability and performance, until the body has “recalibrated” to its newly developed size.

As athletes progress beyond PHV, increases in hormonal activity contribute to structural adaptations, including increases in muscle fibre size and cross-sectional area.

These developmental changes mean that the same training stimulus may produce different responses depending on an athlete’s maturity status.

Adolescence also represents a period during which athletes may experience an increased risk of injury. Understanding growth-related injuries, such as Osgood’s and Sever’s disease, is important, as they can significantly affect an athlete’s ability to train and compete, with negative impacts that may persist for several years.

Monitoring and calculating maturation can aid practitioners’ awareness of such growth-related diseases, allowing them to intervene early and reduce the athlete’s injury burden.

Comparing ways to measure maturation

The increasing use of maturation data within sport highlights the need for reliable methods of assessing maturity status.

Several biological systems can be used to assess maturation status. Sexual maturation refers to the development of a fully functional reproductive system. Skeletal maturation refers to the development of a fully ossified adult skeleton. Somatic maturation refers to physical growth patterns, particularly height development, with the athlete reaching maturity when their height growth slows to less than one centimetre per year.^[3]

Each of these approaches has advantages and limitations.

Skeletal maturity assessment is often the reference standard in research studies that compare maturation methods.^[4] However, skeletal age assessment requires radiographic imaging, which unnecessarily exposes individuals to radiation and can be costly in terms of both resourcing and staffing.^[5] These factors limit its practical application in many sporting environments.

Sexual maturation assessments have extensive historical use, but are often considered intrusive and invasive, which can raise safeguarding concerns in modern sport science practice.

As a result, practitioners frequently rely on somatic maturity indicators, which estimate maturation using anthropometric measurements such as height and body mass.

Maturity offset equations estimate the timing of peak height velocity.^{[6, 7]} They are widely used but have important limitations.

Chronological age strongly influences predictions: the method is most accurate within the narrow age range of 13–15 years.^[1] This method also tends to produce systematic error in estimates of early and late maturing individuals. For early maturing athletes, these equations may predict age at PHV later than actual age at PHV. The reverse holds for late maturing athletes: the equations estimate age at PHV as earlier than it actually occurs.^[8]

Studies in youth football have also questioned the accuracy of maturity offset equations, finding that maturity offset methods did not improve the prediction of PHV compared with chronological age alone.^[9]

Percentage of predicted adult height

An alternative somatic indicator is expressing a child’s stature as a percentage of their predicted adult height (PPAH). Two children of the same height may differ in maturity status depending on the proportion of adult height they have already attained. The individual who has reached a greater percentage of their adult height is considered more biologically mature.^[1]

This method shows moderate agreement with skeletal maturity assessments, making it a useful practical alternative when radiographic methods are not feasible.^[10] In youth football populations, the percentage of predicted adult height provides better classification of PHV timing compared with maturity offset methods.^[9]

A key step in calculating the percentage of predicted adult height is estimating the athlete’s adult stature. Earlier prediction methods required skeletal age information, which limited their practical use in applied sport environments.

The Khamis and Roche ^[11] method estimates adult height without requiring skeletal age.

This method predicts adult stature using a chronological age- and sex-specific intercept (), and coefficients for height (), body mass (), and mid-parent height ().

The standard error associated with the Khamis and Roche method is slightly larger than earlier skeletal age-based models, but it remains sufficiently accurate for many applied settings.

Accurate estimation of mid-parent height is an important component of the Khamis and Roche method. Ideally, practitioners will directly measure the height of both biological parents. However, in many cases, parent heights are self-reported. Self-reported stature is often overestimated, so practitioners should adjust their calculations using correction equations when relying on these values.^[12]

If the height of one parent is unavailable, Khamis and Roche ^[11] recommend estimating the missing value based on the reported stature of the available parent. Alternative approaches include using population averages, although this may introduce additional error.

Once we’ve estimated predicted adult height, the percentage of predicted adult height follows via a simple formula:

This value provides an estimate of somatic maturity. Certain percentage ranges correspond to key maturation events. For example, the adolescent growth spurt typically takes off at approximately 85% of predicted adult height, while PHV typically occurs around 90% of predicted adult height.^[13] PHV may occur slightly later, around 91.5% of predicted adult height.^[14]

Based on these observations, the circa-PHV period may be broadly defined as approximately 88%–96% of predicted adult height.

Converting PPAH to maturity timing

The percentage of predicted adult height can also be converted into a maturity timing classification using z-scores relative to age-specific reference values derived from the Berkeley Longitudinal Study ^[15] or the UK 1990 dataset ^[16].

Using these reference values, individuals can be classified as early maturing, on-time maturing, or late maturing.

However, it is important to consider that the reference population will limit the application. Therefore, more clubs and federations are identifying or creating the most appropriate reference populations for comparison.

Common issues when using PPAH

Although the PPAH method has been widely used in both research and applied sport settings, inconsistencies have emerged.

Common errors in spreadsheet calculations include inappropriate units of measurement, the use of incorrect coefficients, and inappropriate electronic spreadsheet formulas. These are typically accidental or through a lack of awareness, but they present issues in maturational assessment.

Self-reported parental heights must be adjusted to account for over-prediction stemming from footwear, or the relative perceptions of a partner.

Regarding units of measurement, the Khamis and Roche formula uses regression coefficients established from stature measured in inches and body mass measured in pounds. In contrast, practitioners in European nations typically record height and mass in centimetres and kilograms, respectively. Before calculating predicted adult height (PAH), all observations need to be converted to the imperial system.

The original work from Khamis and Roche provides both the calculation to predict adult height and a set of coefficients required for the calculation.

However, due to a typographical error, some coefficients are incorrect, and an updated table of coefficients was published as an erratum.^[17] Notably, several columns of the original coefficients are out by a decimal place. Unfortunately, this error is not clearly signposted, and practitioners may unknowingly use erroneous coefficients, resulting in incorrect prediction calculations.

Improper rounding of decimal age is a common source of “operator error”.

The age‑specific coefficients in the KR method are calibrated to the child’s current age band. If decimal age is rounded up or rounded to the nearest age band (which implies the potential to be rounded up), the calculation will incorrectly pull coefficients from an older, future age group.

This skews the results because you should not model maturation using reference values for an age the child has not yet reached, consequently inflating and distorting prediction outcomes.

Therefore, decimal age must always be rounded down to ensure the prediction uses the correct, age‑appropriate coefficient set.

Lastly, practitioners should be mindful of using confidence intervals when reporting height predictions.

Prediction accuracy increases the closer a youth is to PHV. Therefore, while confidence intervals are derived from PAH, their use is encouraged to provide a range over a precision of height.

There is a clear need for standardised protocols that clearly outline the correct steps to calculate predicted adult height, percentage of adult height, and maturity classifications. Inconsistencies in implementation can lead to both random and systematic errors, ranging from minor discrepancies to substantial misclassifications. While many of these errors are subtle, or require greater awareness of their existence, failure to address these will result in inconsistent and incorrect assessments of maturation, including incorrect PPAH, and potentially inaccurate maturational status classification.

In practical terms, this may result in inappropriately grouping athletes for training and games (e.g., biobanding), false perceptions of growth status that may bias against early or late maturing athletes, and inflated assessments of PAH that potentially impact talent identification processes.

The large variation in maturation timing among individuals means that chronological age alone is often insufficient when evaluating young athletes. Practitioners should adjust performance evaluations based on maturation and provide developmentally appropriate training based on maturity status.

By reliably measuring maturity, practitioners can better understand an athlete’s journey towards adulthood, including the identification and development of talent, as well as implementing an appropriate training plan.