I think the author did the right thing here: pick an arbitrary cutoff like top 20, and assume that within that bucket they’re all pretty good, without worrying

I think the author did the right thing here: pick an arbitrary cutoff like top 20, and assume that within that bucket they’re all pretty good, without worrying too much about the specific ranking of each school.