This is the final version of the article. Available from Elsevier via the DOI in this record.To calculate the return periods of individual wave or crest heights, the long-term distribution of sea states must be combined with the short-term distribution of individual wave or crest heights conditional on sea state. This is normally achieved using an equivalent storm model to parameterise the distribution of the maximum wave or crest height in a storm. A new equivalent storm model is introduced that generalises the approach of Tromans and Vanderschuren (1995). The generalised equivalent storm (GES) method is significantly simpler than equivalent storm methods that model the temporal evolution of the significant wave height in a storm. The GES method is applied to long time series of wave buoy measurements for deep and shallow water sites and demonstrated to be more accurate than existing methods at representing the statistical characteristics of measured storms. Return periods of crest heights from the GES method are shown to be more robust to uncertainties in the fitted models of the equivalent storm parameters than estimates from temporal evolution methods such as the equivalent triangular storm and equivalent power storm model.This work was partly funded through EPSRC grant EP/R007519/1.