Prediction of the Swings in Plays Based on Momentum of Tennis Matches

Tennis is an increasingly popular sport worldwide. This paper focuses on the study of "momentum" in tennis matches, which is utilized to predict match fluctuations. In this paper, we quantify the momentum about physical momentum and construct an evaluation model. Using principal component analysis to analyze the long-term influencing factors, four principal components are obtained; considering that the influence of the actual game situation has periodical timeliness, a mathematical model of the short-term factors is constructed. Then, based on the established evaluation model, this paper quantifies the fluctuation, and based on the data of the finals, conducts ARIMA time series prediction of the fluctuation, and uses the neural network model to determine the most relevant factors affecting the fluctuation.