Consideration of the eventual forced outage of individual power stations leads to a large number of possible states of the power generating system, all with their own probability. It is possible to design a stochastic method to properly take into account all of these possibilities and to weigh them accordingly. In broader energy models, instead of these stochastic techniques that require a considerable amount of calculation time, mostly approximative static simplified methods are applied. Up till now, these simplified techniques have not been validated. The scope of this paper is to check their validity. Therefore, two approaches are compared: a complete stochastic approach and a method based on the derated power (which is the nominal power multiplied with the average availability) of the individual plants. The conclusion of this comparison is that derated power may be used in energy modelling instead of the complicated stochastic approach. The error made is very small and the correlation between the unserved load probability functions obtained by both methods is excellent. Copyright (C) 2004 John Wiley Sons, Ltd.