The goal of our study was to discriminate potential genetic differences between humans who are in both endpoints of the sports performance continuum (i.e. world-class endurance vs power athletes). We used DNA-microarray technology that included 36 genetic variants (within 20 different genes) to compare the genetic profile obtained in two cohorts of world-class endurance (N=100) and power male athletes (N=53) of the same ethnic origin. Stepwise multivariate logistic regression showed that the rs1800795 (IL6-174 G/C), rs1208 (NAT2 K268R) and rs2070744 (NOS3-786 T/C) polymorphisms significantly predicted sport performance (model χ(2) =25.3, df=3, P-value <0.001). Receiver-operating characteristic (ROC) curve analysis showed a significant discriminating accuracy of the model, with an area under the ROC curve of 0.72 (95% confidence interval: 0.66-0.81). The contribution of the studied genetic factors to sports performance was 21.4%. In summary, although an individual's potential for excelling in endurance or power sports can be partly predicted based on specific genetic variants (many of which remain to be identified), the contribution of complex gene-gene interactions, environmental factors and epigenetic mechanisms are also important contributors to the "complex trait" of being an athletic champion. Such trait is likely not reducible to defined genetic polymorphisms.