Claims that sperm counts have fallen worldwide are based on meta-analysis of sperm concentration data taken from a variety of published studies that were never designed to answer this question. Since methods of counting sperm in the laboratory over time have changed so markedly, both in terms of the techniques used and in terms of the staff training and quality control methods involved, this debate article argues that it is almost impossible to conclude that the proposed decline is anything other than methodological error. The American cosmologist Carl Sagan said that ‘extraordinary claims require extraordinary evidence’, and at the present time the author of this article argues that the data from these meta-analyses are not extraordinary evidence. To properly answer the question, it is proposed that adequately powered and controlled prospective studies be conducted at various locations around the world. Only then will we truly know whether sperm counts have declined worldwide or not.