This study examines the importance of game elements of the first set in Wimbledon matches in deciding the match outcome. We propose a LASSO-induced logistic regression model based on first set data to identify the variables that impact the match outcome. Our findings indicate that winning service points, average distance travelled, and rating points significantly impact match outcome. Additionally, we show the effectiveness of our proposed model in within-match forecasting during the first set, and it frequently performs better than other statistical and machine-learning approaches. We also discuss important managerial applications of our methodology for players, coaches, and other stakeholders.