Centre of foot pressure (CoP) measures, computed from a force platform, are commonly used to assess individual's postural control during unperturbed stance. The purpose of the present study was to evaluate the effect of the number of trial recordings on the test-retest reliability of CoP measures and to determine the optimum number of trial recordings required to maximise their test-retest reliability. Ten young healthy adults were asked to stand upright, eyes closed, as still as possible on a force platform allowing measuring the CoP displacements. Two sessions of ten 30s trials were performed with 1h rest in between. Intra-class correlation coefficient (ICC) with 95% confidence interval and Bland and Altman analysis were used as statistical method for assessing test-retest reliability of CoP measures. These analyses were conducted on both (1) non-normalized CoP measures and (2) CoP measures normalized relative to the subjects' anthropometric properties (height, weight and body mass index). Results show that ICCs generally increase as the number of trials used to compute CoP measures increases. Interestingly, three 30s trial recordings are sufficient to ensure excellent test-retest reliability of 12 CoP measures widely employed in clinical practice, namely two-dimensional CoP parameters (surface area, range, mean and maximal velocities of the CoP displacements) and one-dimensional mediolateral and anteroposterior CoP parameters (variance, range, mean and maximal velocities). The present findings could have implications in clinical and rehabilitative areas.