Central Limit Theorem

The central limit theorem says that averages of many independent, similar-sized random influences tend to form a bell-shaped sampling distribution, even when the original data are not bell-shaped. It explains why averages are often more predictable than individual observations.