Anthropometric Clusters of Competitive Cyclists and Their Sprint and Endurance Performance

Front Physiol. 2019 Oct 9;10:1276. doi: 10.3389/fphys.2019.01276. eCollection 2019.


Do athletes specialize toward sports disciplines that are well aligned with their anthropometry? Novel machine-learning algorithms now enable scientists to cluster athletes based on their individual anthropometry while integrating multiple anthropometric dimensions, which may provide new perspectives on anthropometry-dependent sports specialization. We aimed to identify clusters of competitive cyclists based on their individual anthropometry using multiple anthropometric measures, and to evaluate whether athletes with a similar anthropometry also competed in the same cycling discipline. Additionally, we assessed differences in sprint and endurance performance between the anthropometric clusters. Twenty-four nationally and internationally competitive male cyclists were included from sprint, pursuit, and road disciplines. Anthropometry was measured and k-means clustering was performed to divide cyclists into three anthropometric subgroups. Sprint performance (Wingate 1-s peak power, squat-jump mean power) and endurance performance (mean power during a 15 km time trial, V ˙ O2 peak) were obtained. K-means clustering assigned sprinters to a mesomorphic cluster (endo-, meso-, and ectomorphy were 2.8, 5.0, and 2.4; n = 6). Pursuit and road cyclists were distributed over a short meso-ectomorphic cluster (1.6, 3.8, and 3.9; n = 9) and tall meso-ectomorphic cluster (1.5, 3.6, and 4.0; n = 9), the former consisting of significantly lighter, shorter, and smaller cyclists (p < 0.05). The mesomorphic cluster demonstrated higher sprint performance (p < 0.05), whereas the meso-ectomorphic clusters established higher endurance performance (p < 0.001). Overall, endurance performance was associated with lean ectomorph cyclists with small girths and small frontal area (p < 0.05), and sprint performance related to cyclists with larger skinfolds, larger girths, and low frontal area per body mass (p < 0.05). Clustering optimization revealed a mesomorphic cluster of sprinters with high sprint performance and short and tall meso-ectomorphic clusters of pursuit and road cyclists with high endurance performance. Anthropometry-dependent specialization was partially confirmed, as the clustering algorithm distinguished short and tall endurance-type cyclists (matching the anthropometry of all-terrain and flat-terrain road cyclists) rather than pursuit and road cyclists. Machine-learning algorithms therefore provide new insights in how athletes match their sports discipline with their individual anthropometry.

Keywords: anthropometry; cycling; data science; machine learning; physical performance; sports specialization.