<p>Meta-heuristic optimization methods are widely used to solve optimization problems across various fields. However, many of these methods face common challenges that include high computational burden, premature convergence, sensitivity to parameters, etc. To address these issues, any optimization method needs to achieve an effective balance between exploration and exploitation phases. In this context, this paper introduces a new game-inspired meta-heuristic optimization algorithm called the Kabaddi Optimization Algorithm (KOA). KOA is inspired by Kabaddi, a traditional sport originating from the Indian subcontinent, and mimics the strategies and actions of defending team players attempting to catch the opponent’s raider. Compared to other optimization methods in this category, KOA employs unique grouping and movement strategies that help in maintaining a strong balance between exploration and exploitation. To evaluate the performance of the proposed KOA scheme, a comparative study is conducted using classical benchmark functions as well as the CEC-2020 benchmark functions. The comparative analysis demonstrates that KOA outperforms other competitive algorithms. Based on Friedman’s rank test for the CEC-2020 functions, KOA achieved a rank of 2.115, which is better than the mean ranks of 2.225, 3.75, 2.8, and 4.1 obtained by other methods used for comparison. Furthermore, the algorithm is tested on a real-world optimization problem related to electrical power systems: the combined heat and power economic dispatch (CHPED) problem. Four test cases were considered for evaluating the performance of the proposed method. Among the four test cases, the proposed method provides an average improvement of 0.82% in cost reduction compared to various other optimization methods used for comparison. Overall, by considering the results for the two test cases, it can be concluded that the method provides a satisfactory performance with respect to other methods used for comparison.</p>