This paper outlines the main features of LREMS, an electricity market simulator based on game theory allowing to simulate the long run evolution of a generation set (investment in new plants, refurbishment or dismission of older ones). LREMS is a hierarchic simulator: a long term "outer" game takes yearly investment decisions based on mid-term price projections provided by an "inner" medium run market simulator. The latter, also based on game theory, calculates the hourly market clearing over a yearly time horizon taking into account the strategic bidding of the market leaders while optimally allocating the hydro resources within each month. A third game is devoted to calculate periodically the unit commitment of the thermal units. An application of LREMS studying the evolution of the Italian electricity market in the next twenty years is also presented