It depends on the policy. Is the height of a basketball hoop a policy that is unintentionally discriminating against me because I can't jump that high? Is the Nobel Prize racist because Jews are massively overrepresented among winners?
The debate here is over whether any policy that results in racial disparity is racist. To me, that argument is obviously wrong.
> A good example is in the (historical) [entrance exams for the UK's Grammar School systems which are sat at age 11[(https://en.wikipedia.org/wiki/Eleven-plus#Controversy); while the grammar-school system was ostensibly universal and open to all, the entrance exams with questions predicated on a familiarity with middle-class culture naturally disadvantaged working-class students.
To you, tests like the one you mention are racist (or classist) because you assume that certain races or classes will be more likely to know certain things and others less likely. But, ironically, to me, your assumption is racist (and classist) because we have different definitions of racism.
Finally, I think there's a serious flaw in your thinking. If "studies show" that rich kids test better in math, does that mean math tests are classist? Is it even possible to create a test so that every group you consider (ethnic or socio-economic) will achieve the same average score? And which groups shall we consider? Isn't it common knowledge these days that there is no canonical way to divvy people up into "races"? Why is it that, for the purpose of college admission, "Asian" is an ethnicity but "Jew" isn't? Furthermore, imagine I am born to rich parents and my parents hire a math tutor from the age of 3 until I graduate high school. By the time I take the SAT, I will probably be much better at math than the average high school student. Is there anything wrong with that? Why is it correct for tests to attempt to discern innate ability rather than current ability?