This paper evaluates four different methods that have been proposed for the estimation of the detachment depth beneath detachment folds. Guidelines are presented in order to use the most suitable method in a particular region deformed by detachment folds. These guidelines are constructed considering the assumptions of each method, the influence of different parameters on the estimation of the detachment depth (folding mechanisms, cross-sectional area variations, position and orientation of the regional datum, number of stratigraphic horizons, bed length and thickness, dip and position of the bounding lines, initial thickness of the ductile unit, single anticlines or fold trains) and the available data (ductile unit thickness, regional datum, detachment depth). Moreover, a new, simple method based on two previous methods is presented. The advantages of this new method are that the calculations are extremely simple and it uses information from more than one stratigraphic horizon; therefore, it does not depend strongly on the accuracy of the data at one level. The precision of the proposed methods is tested by their application to a number of natural and experimental single anticlines and fold trains. It appears that better predictions are obtained when analysing a complete detachment fold train than when dealing with a single detachment anticline. The reason might be that ductile material flow beneath the folds along the cross-sectional plane is taken into account when dealing with a long-enough cross section. We also suggest some hypotheses concerning what the shape and dimensions of the detachment folds indicate about the detachment depth in those cases where insufficient data are available to apply the methods proposed.