Machine Learning-Based Cricket Player Performance Prediction System Using Django

Abstract

Cricket is a data-intensive sport where player selection, performance evaluation, and match strategy rely heavily on statistical analysis. Traditional methods of assessing player performance often depend on manual observation and basic statistical measures, which may not fully capture a player’s potential or consistency. With the advancement of machine learning techniques, predictive analytics has become a powerful tool in sports analytics. This paper presents a web-based cricket player performance prediction system that leverages machine learning models to estimate a player's performance score based on key statistical inputs.The proposed system is developed using the Django web framework, integrating a pre-trained machine learning model stored using Joblib. The application accepts user input parameters such as batting average, strike rate, bowling average, economy rate, number of matches played, experience level, and recent form. These features are processed and passed into the trained model, which predicts a performance score. The system provides real-time predictions through an intuitive web interface, enabling coaches, analysts, and enthusiasts to evaluate players efficiently.The architecture ensures efficient model loading at server startup to reduce latency during prediction. Data preprocessing is handled using Pandas, ensuring structured input for the machine learning model. The modular design allows easy updates to the model without altering the frontend interface. Error handling mechanisms are incorporated to ensure robustness and user-friendly feedback.This system aims to bridge the gap between traditional cricket analysis and modern data-driven decision-making. By providing accurate and fast predictions, it supports better player selection and performance assessment. The application can be extended with advanced features such as real-time match data integration, deep learning models, and player comparison analytics.The results demonstrate that machine learning models can effectively predict player performance using historical statistics, making this system a valuable tool in sports analytics. The project highlights the potential of combining web technologies with artificial intelligence to create scalable and impactful applications in the sports domain