I agree that stereotyping is all too easy to do. Then again, complaining about stereotyping is always easy. As I explain in my articles on epistemology (link below, for example), all concepts are models that simplify. The question is whether a generalization is useful and fair, or counterproductive and misleading.
So if I generalize about Trump's supporters, I'm necessarily leaving out details and exceptions. The same is true of every single generalization that's ever been uttered.
Is the comparison between Trump's supporters and the OJ Simpson jurors fair and useful? Does it capture important, overlooked similarities? Or is it a gross misrepresentation? Is the analogy strong or weak? Those are the relevant questions for critics, I think.