Null (red) and alternative (blue) distributions with simulated trials
Power against patients per arm

Randomised Two-Arm Trial: Power Curves and Repeated Trials (2D)

Run thousands of simulated two-arm clinical trials and see how often a real treatment effect is detected. Change the effect size, patient variability, group size and significance level to move the null and alternative distributions of the test statistic, shade the false-positive and false-negative regions, and find the sample size needed for a target power. A Blender-built stage with checks of power, type I error and sample size against Monte Carlo and brute-force integration.

Made with 🟠 Blender