Machine learning-based optimization of contract renewal predictions in Korea Baseball organization

Area of Science:

• Sports Analytics
• Machine Learning in Sports
• Baseball Performance Evaluation

Background:

• The Korea Baseball Organization (KBO) implemented a foreign player system in 1998, increasing foreign player limits in 2014.
• KBO teams typically sign two foreign pitchers and one foreign batter annually.
• Low contract renewal rates (34-36%) for foreign players highlight the need for improved recruitment and retention strategies.

Purpose of the Study:

• To develop and compare machine learning models for predicting foreign player contract renewal decisions in the KBO.
• To identify key performance indicators and data sources that enhance prediction accuracy.
• To provide data-driven insights for KBO teams to optimize foreign player recruitment and contract management.

Main Methods:

• Utilized machine learning algorithms to predict contract renewal decisions for foreign players.
• Incorporated performance data from Minor League Baseball, Major League Baseball, and the KBO.
• Included player image data as a potential predictive factor.
• Evaluated model performance using accuracy, Area Under the Receiver Operating Characteristic curve (AUC), and precision.

Main Results:

• Performance data within the KBO was found to be the most significant factor in improving prediction accuracy.
• Machine learning models demonstrated effectiveness in predicting contract renewal outcomes.
• Foreign batters who initially succeeded in the KBO showed a performance decline potentially linked to international tournament results.

Conclusions:

• The proposed machine learning approach offers a valuable tool for KBO teams to make informed contract renewal decisions.
• Accurate player performance evaluation can enhance foreign player recruitment, team performance, and the overall competitiveness of the KBO league.
• Further analysis of foreign batter performance trends may offer insights into long-term player development and international competition impacts.