[foobarbazqux]

A single counterexample is enough to refute a universal claim.

[vinceguidry]

It certainly isn't a universal claim.

[tomrod]

But multiple examples provide only a skewed dataset.