Each school has a list of admitted and a list of enrolled. So, theoretically to get the exact cross-admit numbers instead of abusing stats like what Parchment does to figure out, all the schools have to do is to exchange those two lists in October, or prior to that when they know for sure. I remembered seeing Stanford’s report on Stanford’s admission Dean calling MIT’'s on this issue, though I am not exactly sure how Stanford does this.
Those numbers are highly guarded and only Stanford sometimes releases them. The cross-admit yield rates seriously affect their decisions on many things, especially the yields among HYPSM, since most cross-admits they lost in the past were among those five schools.
The yield of each school does reflect something regarding the winning/losing cross-admits, and the yield/admit ratio might be a better indicator for which school is “more popular”, since it reduces some of the manipulation effect . Last year, it was the first time Stanford had a higher yield/admit ratio than Harvard did.
So is what the title trying to say, nothing more than this.