Budget allocation for power system reliability improvement is considered among the sophisticated problems because of its nonlinear nature. This nonlinearity makes the problem intractable for large-scale power systems. This paper compares two approaches for budget allocation for power system reliability improvement. The first method is the traditional one, which suffers from the non-convexity and nonlinearity. In the second approach, a linear zero-sum mixed-strategy game is proposed where a limited budget is allocated among the network elements based on the game variables along with an iterative algorithm to find the solution. Both models are applied to the modified RTBS system. The results show that while each strategy adopts a different tactic for reliability improvement, both strategies improve the system reliability to the same level for a given budget.