The article discusses the application of the ideas of evidence-based medicine or empirically supported treatment to the field of psychoanalysis. The author argues that empirical support is gradual rather than categorical, and that the randomized clinical trial (RCT), which is commonly heralded as the appropriate method of support, confounds patient suitability factors with treatment 'as such'. As a consequence of the exclusive reliance on the RCT, empirical validation of psychological treatments has generally favoured cognitive-behavioural treatments, although there is ample evidence - albeit non-RCT - for the beneficial effects of short-term and long-term psychoanalytic treatment. The author argues for a suitability matching approach to empirical validation, where each treatment is tested, as powerfully as feasible, on samples based on selfselection (patient's choice, clinical assignment, suitability judgments etc.). The paper concludes by referring to two recent studies, indicating that there is reason to evaluate alleged empirical support with caution.