If one's opponents say that outliers do not exist at all, all you need to do to show that they are wrong is come up with a single example, outlier or no.

The wise opponent will then retreat into more nuanced language of trends, and then the rhetorical battle must be fought on the fields of propensity and majority.

A lot of rhetorical opponents are neither wise in debate nor in belief, though. So it's entirely sensible for Will's first foray to be pointing out exceptions and outliers.