Abstract: In the new media era, a comprehensive examination of online public opinion incidents in commercial sports events has become crucial for promoting high-quality development in the sports industry. Using trending topics on Sina Weibo as data sources and Weibo posts and comments related to the “MX Hong Kong Event” during its peak attention period as samples, this study employs Python, Origin, and ROST EA software to conduct a multi-dimensional analysis of the evolution of online public opinion. Key dimensions include temporal evolution, opinion leaders, focal topics, and emotional tendencies. The study identifies four typical characteristics of public opinion evolution: a multi-factor-induced “superposition state”, a strong attitude-driven “communication state”, a reference-effect-based “comparison state”, and a high-response-demand “dissolution state”. Based on these findings, the study proposes coping strategies from four perspectives: “intensity” regulation, “attitude” regulation, “direction” regulation, and “effectiveness” regulation. These strategies aim to mitigate risks associated with online public opinion in sports events, standardize event operations, and enhance comprehensive governance capabilities.