The sun is not hurried by early risers.

'If the clinician knows, or has good reason to believe, that a new therapy (A) is better than another therapy (B), he cannot participate in a comparative trial of therapy A versus therapy B. Ethically, the clinician is obligated to give therapy A to each new patient with a need for one of these therapies' (Shaw and Chalmers, 1970). This ethical dilemma would make participation in clinical trials for many of us impossible. It may also hinder, or even thwart, the development of new therapies. For this reason, Ben-jamin Freedman introduced the term equipoise (Freedman, 1987). Equi-poise means that in the medical community at large—so, not in individual clinicians—there is genuine uncertainty about either the benefit of a new therapy or about the superiority of a new therapy over an existing one. Given that randomized clinical trials presume equipoise, very large treatment effects will therefore rarely be found. They usually are found in first, small, observational studies, especially in those that present surrogate outcomes. If a large treatment effect is found in such a first small study, it usually occurs by chance, at one of the extremes of a normal distribution, and regression-towards-the-mean will, by definition , have subsequent studies show smaller effects (Ioannidis, 2008). Be aware that this may concern perfectly valid studies with rigorous, robust methods. Pereira and co-workers (2012) studied over 80 000 forest plots from 3000 Cochrane reviews (the highest quality evidence). They found that 90% of very large treatment effects (i.e. odds ratio [OR] .5.0) in first trials became smaller in meta-analyses that included subsequent trials. The mean OR decreased from 12 to 4. Many trials lost their nominal significance. They conclude: 'Most large treatment effects emerge from small studies, and when additional trials are performed, the effect sizes become typically much smaller', and—referring to the effect of regression-towards-the-mean—'(.. .) most large treatment effects should be considered with caution: many are spurious findings, while the vast majority may represent substantial overestimations'. Ours is the clinical specialty with the smallest, utterly convincing, observational, n ¼ 1 study in the medical literature: the birth of a healthy baby to a woman without Fallopian tubes (Steptoe and Edwards, 1978). Not all our studies are that clear-cut however, and especially many 'firsts' have to be taken with a pinch of salt. We want you to be aware of this. Human Reproduction wishes to publish high-quality reviews of high-quality original …