Momentum Study Based on Logistic Regression Models and BP Neural Networks

Abstract

In the context of a game, "momentum" usually refers to a set of events (e.g., consecutive points scored) that create momentum or a trend in a game, which may have a significant impact on the outcome of the game. The purpose of this paper is to investigate how momentum affects the outcome of a game through mathematical modeling. First, an AHP-TOPSIS model is built to calculate the score of each athlete at each moment to quantify and visualize momentum. Second, based on real game data, a Kruskal-Wallis H model test is established to assess the correlation between momentum and score. To predict the fluctuation of players' momentum, logistic regression model was used to predict the momentum fluctuation. Finally, GABP neural network and genetic algorithm are established to visualize the performance data of the players and predict the inflection points during the game. The model proposed in this paper effectively solves the problem of momentum during the game and improves the ability to predict and analyze the game results.