Comparing Accuracy of the Final Height Prediction Models for Elite Football Players and Developing a New Model

Area of Science:

• Sports Medicine
• Human Growth and Development
• Biostatistics

Background:

• Accurate prediction of final height is crucial in sports, particularly for elite athletes.
• Previous height prediction models have varying accuracy across different populations.
• Understanding factors influencing final height in athletes can aid in talent identification and development.

Purpose of the Study:

• To evaluate the accuracy of existing height prediction models in male Japanese football players.
• To develop and validate new, more accurate height prediction models for this specific demographic.

Main Methods:

• Compared two conventional models: Model 1 (current height, calendar age, bone age) and Model 2 (current height, calendar age, parental height).
• Developed Model 3 by optimizing coefficients for Model 1.
• Created Model 4 by adding parental height to Model 3 and optimizing coefficients.

Main Results:

• Model 2 (parental height) showed higher prediction accuracy (R=0.52) than Model 1 (R=0.33).
• The new Model 3 equation was derived based on current height, calendar age, and bone age (R-square=0.49).
• The final Model 4, incorporating parental height, demonstrated the highest accuracy (R-square=0.61).

Conclusions:

• Parental height is a significant predictor of final height in male Japanese football players.
• A novel height prediction model (Model 4) integrating current height, age, bone age, and parental height offers superior accuracy.