Regression to the mean
After an extreme result, the next one is usually closer to average.
What it is
When results depend partly on luck, an unusually good or bad one tends to be followed by one nearer the average. Nothing has to cause the change; the luck simply does not repeat.
An example
A flight instructor notices that cadets he praises fly worse next time and cadets he scolds fly better, and concludes that scolding works. Both changes are what you would expect after an unusually good or bad landing.
Where to use it
Before crediting a fix, a treatment or a new manager for a recovery from a low point.
The question to ask
How much of that result was luck that will not repeat?