One of the strengths of futures markets is the elimination of counterparty risk, but to accomplish this, it is important to consider the financial guarantees the clearing house requires from market participants. These margins must hedge the risk related to extreme variations in the product price, but they should not be excessive to avoid limiting the number of participants in the market. In this paper we propose a new methodology to provide appropriate margins in the electric power futures market, and we present an application for the Colombian market. We conduct a Monte Carlo simulation to assess the daily changes of the futures price and estimate measures of risk for different scenarios for "El Niño" weather conditions, holding periods, and expiration times. We find that the new methodology substantially modifies required financial guarantee levels compared to the methodology currently used to calculate margins.