The double-blind, placebo-controlled trial is held as the gold standard in medical knowledge, but this tool of investigation has its weaknesses. These include ethical limitations on the types of comparison that can be undertaken, the central conflict between best practice for an individual and trial protocols, problems of applicability to the general population and applicability of work done on one population to another, type II errors, publication bias, misuse and limitation of statistics, fraud, maintenance of blinding, asking the wrong question, and a simplistic, reductionist view of clinical management. The concentration on the randomised, controlled trial devalues information from other sources, such as natural history studies, clinical experience and case reports. The randomised, controlled trial is an important source of information and as physicians we should welcome more well-crafted trials, but they are not the only source of information.