Historically, clinical trials of haemophilia with inhibitors (HwI) have been challenged by the small patient population. New approaches to clinical trial methodology and statistical modelling could potentially be used for study optimization. The aim of this work was to evaluate the impact of different trial designs and study conditions on the estimated drug potency and power, and compare traditional statistical methods with repeated time-to-event (RTTE) modelling in terms of power. Bleeding information from a clinical trial of 23 haemophilia patients with inhibitors treated on-demand was used to develop a baseline RTTE model using NONMEM. Clinical trial simulations for a hypothetical anti-haemophilic drug were performed, by adding a drug effect and a literature-derived placebo effect to the baseline RTTE model, using different trial designs (parallel-group, placebo-controlled parallel-group, crossover and placebo-controlled crossover designs) and study conditions, including sample size, study duration and doses. The precision and accuracy of the estimated drug potency (EC50) and power for different trial designs, study conditions and statistical methods (RTTE modelling, t-test and negative binomial regression) were evaluated. The developed baseline RTTE model accurately described the clinical data. The crossover designs displayed up to four-fold higher precision of the estimated EC50 and three-fold higher power relative to the parallel-group trial designs. Furthermore, RTTE modelling provided a higher power relative to the traditional statistical tests. We found that crossover designs in combination with RTTE modelling can reduce the required sample size and study duration, while ensuring high power and precise estimation of EC50, in clinical trials of HwI.