The fallacy of sweeping generalization occurs when an arguer incorrectly treats a generalization as universal, ignoring that there are exceptions and in some cases even a sizable minority for which the generalization does not hold true. The attraction of the fallacy is that the generalization has some truth to it, in that it holds true for most of the cases. The fallacy lies in treating the generalization as if it applies to the entire population.

Here is an example:

"People with degrees from elite colleges do not play in the NFL."

While it may be true that most NFL players did not graduate from an elite school, there have been many who did, such as John Elway (Stanford), Ryan Fitzpatrick (Harvard), Rodney Thomas (Yale), and Andrei Iosivas (Princeton), to name a few. Thus, the above statement is a sweeping generalization.

Note that merely by inserting the word "most" at the beginning of the statement, it would no longer be a sweeping generalization -- though it would be a weaker (less impressive) statement.

Another form of this fallacy occurs when the terms of a generalization are not very specific, such that the generalization may be true in one respect, while false in another, depending on how it is construed.

For example:

Jojo: Bart had the best season of any running back, because he gained the most yards.

Momo: Yeah, but he didn't score as many touchdowns as Mark's best season.

In this example, what counts as the "best season" is not specified by Jojo, and the circumstances make the generalization either true or false, depending on how "best season" is spelled out. This means that the non-specific generalization is a sweeping generalization.