> no small amount of luck, finance, or legacy The problem is that those factors confound measurement so much that we don't actually know if the highest-scoring

> no small amount of luck, finance, or legacy The problem is that those factors confound measurement so much that we don't actually know if the highest-scoring students are really the best. Lowering the bar in the face of absolute certainty and precision might be bad (that itself is a debatable point) but that's not what the authors are proposing.