In order to solve the problem of low accuracy of personalized prediction of current sports course model, personalized sports course design based on machine learning is proposed. In this paper, the machine learning algorithm-support vector machine is used to establish the college students’ sports performance prediction model, and the particle swarm algorithm is used to select the model parameters, and finally the model is applied to the sports performance modeling and prediction in a university. The experimental results show that the prediction accuracies of all college students’ sports scores are more than 90%, which are much larger than 85% in the practical application range, which indicates that the model is versatile and can be applied to the prediction of college sports scores in practice.