Race momentum analysis prediction based on wavelet neural networks

Tennis is a popular sport around the world and 'momentum' plays a key role in determining the outcome of matches, as discussed in this report. Our team created a momentum model for tennis matches by analysing seven factors, including points scored, serve ends, and using an activity index and an error index as state scores and demerits. The model was based on a sample of the 2023 Wimbledon 1301 Championships and showed changes in athlete form through line graphs. Through Monte Carlo simulation, the experiment found that the momentum distribution of actual match results differed significantly from the simulated random distribution, indicating that momentum had a significant effect on match results. And the change in dominance in the middle of the match is predicted by a wavelet neural network. The innovative approach using wavelet neural networks also allows for more detailed analysis of momentum changes during a race, providing a new perspective on the dynamics of athletic performance.