Sports prediction models have become an essential field within sports analytics, combining statistical reasoning, data mining, and machine learning to forecast outcomes and inform decision-making. The ability to forecast results in sports such as football, handball, cricket, tennis, and basketball is valuable for coaches, analysts, bettors, and fans. The reviewed literature demonstrates a methodological evolution from traditional statistical models—such as logistic regression, Poisson regression, and Gaussian approximations—to modern artificial intelligence techniques including adaptive back-propagation neural networks, ensemble learning methods, and explainable AI integrated with deep learning. Findings highlight that predictors vary across sports: defensive efficiency and shooting accuracy are dominant in football, overs and wickets significantly influence cricket outcomes, momentum predicts tennis performance, and recent win rate is crucial in basketball. Meanwhile, handball research is expanding rapidly, integrating sensor-based inertial measurement units (IMUs) and computer vision approaches. This paper synthesizes contributions from 2016 to 2025, identifies methodological strengths and limitations, and proposes future directions focusing on generalizable models, multi-modal data integration, and transparent explainable systems.