Aiming at the problem that the current model does not consider the reasonable selection of the evaluation parameters of the personalized sports training effect in the evaluation of the personalized sports training effect， which leads to the increase of the running time of the personalized sports training effect evaluation model and the low accuracy and efficiency of the effect evaluation， a GA-SVM based personalized sports training effect evaluation model is proposed. First of all， according to the principle of selecting evaluation indicators for personalized sports training effect， based on graph theory and matrix tools， the evaluation indicators are selected， and the classification and dimension of the selected evaluation indicators are unified； Then， GA （genetic algorithm） is used to select the optimal value of SVM （support vector machine） evaluation model parameters to optimize the SVM evaluation model； Finally， using MATLAB simulation software， the evaluation model of personalized sports training effect based on GA-SVM is simulated and tested. The simulation results show that the relative error of the evaluation model based on GA-SVM is 0.04%， the time is 1.9 s， the relative error of model evaluation is small， and the time for model evaluation is short， which can effectively improve the efficiency and accuracy of model evaluation.