The days of the classic hit-and-run “Whitey Ball” in baseball are long gone. In today’s game, coaches, scouts, front office staff, and players themselves rely heavily on quantitative analysis, which has transformed the world of sports over the past decade.
But what if artificial intelligence (AI) could be used to push these achievements even further? was shown to increase the effectiveness of
“Basically, we combined a probabilistic (randomly determined) game model with deep neural network learning techniques to compute the optimal pitching sequence at the baseball bat,” says Computer Science and Engineering. Associate professor and collaborator Evgeny Volobeychik said. He is the author of a paper recently presented at the Florida Association for AI Research (FLAIRS) conference, “Computing the Optimal Pitching Strategy in Baseball At-bats.”
Vorobeychik and his co-authors learned how to use deep neural networks to predict the outcome of a pitch every time a hitter swings. We then modeled at bat as a stochastic game and solved it using a combination of a technique known as numerical iteration and linear programming, a mathematical modeling technique.
They focused on data from the 2015-2018 Major League Baseball season to determine an approach to each at-bat that actually helped improve the efficiency of average and below-average pitchers.
The main findings of the study are:
- Dynamic Games: Vorobeychik and his co-authors developed a dynamic game theory model that takes into account the pitcher’s repertoire and control, the batter’s perseverance, i.e., the tendency to swing on pitches outside the strike zone, to determine the optimal pitch. Generated a sequencing strategy. each at-bat.
- Data integration: By integrating comprehensive player and game data, including historical performance and pitch tracking, researchers have developed a framework that can generate personalized pitching strategies for individual pitchers against specific hitters. created.
- Performance analysis: This study evaluated the effectiveness of the optimized pitching strategy by comparing it with the observed pitching efficiency in the data. As a result, it was demonstrated that the batter’s on-base percentage, especially for low-ranked pitchers, decreased significantly.
“I think basically every pitcher who goes to the major leagues has something great,” Vorobeychik said. “Part of what separates good players from just good players is how they use their arsenal against specific hitters in a game setting. By officially solving this as a game, even pitchers with less experience and ability may be able to understand it.” Realize the best pitch sequence and make better use of its elements. ”
But could this model be used to help major league pitchers in their game settings?
“I believe so,” said Volobeychik. “Certainly, there’s still work to be done. For example, we assume that each at-bat is independent, which obviously isn’t the case. That’s actually what we’re working on right now. ”
/ Open to the public. This material from the original organization/author may be of the nature of its time and has been edited for clarity, style and length. Mirage.News does not take any organizational positions or positions and all views, positions and conclusions expressed herein are those of the authors only. Read the full article here.
