Accurate load forecasting is essential for modern power systems, especially with the increasing integration of renewable energy sources and the participation of diverse market players. This paper proposes a load forecasting method based on a multi-player generalized Nash game, capturing the strategic interactions among market participants. The players are categorized into four types: temporal, sustaining, curve-based, and variable, according to their distinct characteristics. The Nash equilibrium is solved using the Gauss-Seidel Fixed-Point Iteration algorithm. Then, the market clearing results derived from the Nash game are combined with the baseline load forecast to produce the final total load prediction, effectively incorporating the strategic bidding behaviors of players. Simulations on the IEEE-9 bus system validate the proposed approach, demonstrating significantly improved forecasting accuracy over traditional methods, particularly in scenarios with high renewable energy penetration.