Integrated anthropometric correlates of planned change-of-direction performance (T-test) in male badminton players: a partial least squares regression study.

PURPOSE: Based on the sport-specific demands of badminton for planned change-of-direction (COD) ability, this study used the T-test to examine the joint association structure between anthropometric indicators and planned COD performance in elite male badminton players, and to identify the high-contribution indicators most closely related to T-test completion time.

METHODS: A cross-sectional design was adopted, including 38 elite male badminton players classified as National Level 1 or Level 2 athletes. Anthropometric assessment included length, breadth, and girth variables, together with derived height-normalized ratios/indices and selected segmental proportion indices. Planned COD ability was assessed using the T-test. Exploratory bivariate correlation analyses were first performed. Subsequently, all anthropometric indicators were jointly entered into an integrated partial least squares (PLS) regression model. The optimal number of latent components was determined by cross-validation using the minimum root mean squared error of prediction (RMSEP) criterion. Model performance and the relative contribution of predictors were evaluated using the coefficient of determination (R²), root mean squared error (RMSE), variable importance in projection (VIP), and regression coefficients.

RESULTS: A total of 38 elite male badminton players were included. Correlation analysis showed that hand length (r = -0.448, p = 0.006), sitting height (r = -0.340, p = 0.043), Cormic Index (r = -0.389, p = 0.019), hand length-to-height ratio (r = -0.482, p = 0.003), and forearm length-to-height ratio (r = -0.482, p = 0.003) were significantly negatively correlated with T-test completion time. In contrast, lower-limb length (r = 0.365, p = 0.029), shank length (r = 0.330, p = 0.049), shank length-to-height ratio (r = 0.351, p = 0.036), shank-to-thigh index (r = 0.355, p = 0.033), Manouvrier's index (r = 0.384, p = 0.021), lower-limb length-to-height ratio (r = 0.384, p = 0.021), and brachial-antebrachial index (r = 0.373, p = 0.025) were significantly positively correlated with T-test completion time. Further multivariable analysis using an integrated partial least squares (PLS) model demonstrated that the three-component solution provided the best predictive performance, with the lowest cross-validated RMSEP (0.784), an R² of 0.65, and an RMSE of 0.599. VIP analysis showed that the anthropometric information most relevant to T-test completion time was concentrated in a limited subset of variables, particularly hand length-to-height ratio, forearm length-to-height ratio, lower-limb length, lower-limb length-to-height ratio, Manouvrier's index, shank length, forearm length, hip breadth-to-height ratio, hip breadth, brachial-antebrachial index, and hand shape index. Regression coefficients further indicated that the retained predictors were associated with T-test performance in different directions, rather than contributing uniformly, suggesting a multidimensional structural pattern underlying planned change-of-direction performance.

CONCLUSION: The present study suggests that T-test performance in elite male badminton players cannot be adequately explained by any single anthropometric indicator alone, but is more likely associated with an integrated morphological profile composed of a limited number of high-contribution variables. The anthropometric information represented in planned change-of-direction performance was mainly concentrated in indicators related to distal upper-limb proportions, lower-limb structural proportions, and segmental proportional configuration. These findings provide a preliminary morphological basis for athlete profiling and training monitoring in badminton. However, given the cross-sectional nature of the study, the observed associations should not be interpreted as causal, and further validation in larger and prospective studies is warranted.