With this scientific command, researchers illuminate the rest of the world with the most rigorous and reliable type of evidence about the usefulness of therapies and other interventions for patient care and public health. However, randomization poses a mathematical challenge widely known and described in probability more than a hundred years ago as the Bertrand paradox, Count Buffon needle problem (Geoeges-Louis Leclerc) and several others. Therefore randomization “per se” does not imply that the evidence is better, more rigorous, or more reliable. In fact, when the randomization method is not correctly described in probability, a default bias is created. Ultimately, this bias makes a study neither reproducible nor controversial, events that are both indisputable pillars of the scientific method.