Those statistics show where the students actually go, not where they apply and are accepted. Why do you assume that a baseline admit rate of 50% would imply that 50% of prep-school alums go to Harvard, which is clearly a false implication?