It's listed there as a way that people use it and then calls that usage objectionable and a misunderstanding.

I don't dispute that people use it that way but it's objectively a misuse. The phrase's misuse implies that evidence against a statement supports the statement.

> In many uses of the phrase, however, the existence of an exception is taken to more definitively 'prove' a rule to which the exception does not fit.

> In what Fowler describes as the "most objectionable" variation of the phrase,[1] this sort of use comes closest to meaning "there is an exception to every rule", or even that the presence of an exception makes a rule more true; these uses Fowler attributes to misunderstanding.