Artificial Intelligence in Sports: An Analysis of Lacrosse Shooting Technique

Abstract

Artificial Intelligence (AI) can transform athletic training as it offers advanced methods of shooting analysis and personalized feedback. While 2.5 million Americans play lacrosse, about 1.4 million play recreationally and could utilize artificial intelligence systems to gain an advantage on the field.1 This review paper investigates how AI can be an effective training tool by improving the technique of lacrosse shooting. Through the review of mechanical analysis and AI tools in sports, including transformer architectures, Large Language Models (LLM)  and reinforcement learning based algorithms, this paper explores how AI can identify flaws in form and provide corrective feedback. By evaluating different research papers and testing several generative AI systems, this paper will summarize the effectiveness and the differences between each generative AI. These findings demonstrate how the use of AI can assist traditional coaching by providing data-driven information, helping to accelerate skill development. AI can also cater to the individual and produce personalized training programs to effectively improve balance, power and accuracy in a lacrosse shot. These AI systems can help bridge the gap between human instruction and newly developed technology, thereby training the player in a more effective and efficient manner. However, the generative AI utilized to gather evaluations and recommendations does not have a reliable spatial analysis feature, which may cause improper outputs or false data.