Regression

This fallacy occurs when regression to the mean is mistaken for a sign of a causal connection. Also called the Regressive Fallacy. It is a kind of false cause fallacy.

Example:

You are investigating the average heights of groups of Americans. You sample some people from Chicago and determine their average height. You have the figure for the mean height of Americans and notice that your Chicagoans have an average height that differs from this mean. Your second sample of the same size is from people from Miami. When you find that this group's average height is closer to the American mean height [as it is very likely to be due to common statistical regression to the mean], you falsely conclude that there must be something causing Miamians rather than Chicagoans be more like the average American.

There is most probably nothing causing Miamians to be more like the average American; but rather what is happening is that averages are regressing to the mean.

False Cause

Improperly concluding that one thing is a cause of another. The Fallacy of Non Causa Pro Causa is another name for this fallacy. Its four principal kinds are the Post Hoc Fallacy, the Fallacy of Cum Hoc, Ergo Propter Hoc, the Regression Fallacy, and the Fallacy of Reversing Causation.

Example:

My psychic adviser says to expect bad things when Mars is aligned with Jupiter. Tomorrow Mars will be aligned with Jupiter. So, if a dog were to bite me tomorrow, it would be because of the alignment of Mars with Jupiter.