Get a warrior with boneshatter, preferably hero with boneshatter at 100%. Then spam 200 boneshatters on the same mob with the heavy mace. Spam 200 boneshatters on the same mob with the light mace. Get the proportion of successes (let's call it p). If you've learnt Statistics, you'll know what to do from here... you can get the sample variance from p(1-p)/n, sample standard deviation from the square root of the sample variance. Repeat the sample variance for the other proportion (light mace successes), and call it q or something. Then set up a difference of means, p - q, and get the difference standard error with square root of (sample variance of heavy mace divided by 200 plus sample variance of light mace divided by 200). Use the standard error of the mean difference and then perform a hypothesis test of null hypothesis difference = 0, alternative hypothesis not equals 0, and see if you can reject the null at some significance level (5% is usually used).
Could someone who knows stats who has a hero mace spec test this?