Regression to the mean

Extremes are mostly noise

Regression to the mean is the tendency of extreme results, which mix skill with luck, to be followed by ordinary ones.

Accounting for regression means interpreting results by:

  1. Separating the repeatable component of a result from the lucky one.
  1. Expecting the next observation to sit closer to the long-run average.
  1. Asking whether an intervention beat that expectation, or merely preceded it.

The outlier was mostly luck, and luck does not repeat on schedule.