International applications continue to rise making up for a lot of the decline in domestic applications.
^^ @doschicos: While that may be true, HYPSM and many selective colleges have NOT raised their cap for international students, currently around 10% to 12% of an incoming freshman class. So, while the competition for international students at those institutions has certainly increased, it has not translated into MORE international acceptances. The exception being many US state universities, which are balancing their budget on the backs of full-fare paying international students. To wit: University of Illinois Urbana: https://www.insidehighered.com/news/2015/01/07/u-illinois-growth-number-chinese-students-has-been-dramatic.
But that’s a digression, as we were discussing YIELD, which is the number of admitted students agreeing to matriculate to a college – and that has reached all time highs at Harvard, MIT, Yale, Dartmouth and probably a few other colleges who have yet to report their yield.
or are the AO’s getting better at really picking out the students who want to come to their college…
^^ 
I have a theory, which unfortunately we don’t have the data to prove or disprove: as financial aid has improved at the top schools and their proportions of first-gen students have increased, these schools are admitting more and more kids for whom they’re both the best and most affordable option, and yield ticks up as a result. I think first-gens, unless they’re academically or otherwise elite URMs that many schools want, don’t tend to be cross-admitted much to other top schools with similar financial aid, both because they don’t tend to apply to as many (for example, if they’re matched through QuestBridge) but also because their profiles are relatively more idiosyncratic and may not stand out to more than one of the top schools.
^^ The generous financial aid provided by HYP is definitely a factor in the higher yield.
“as financial aid has improved at the top schools and their proportions of first-gen students have increased, these schools are admitting more and more kids for whom they’re both the best and most affordable option, and yield ticks up as a result”
A very sensible and logical hypothesis.
On the other side of the same coin: the proportion of non-first-gen students have decreased, and the competition among this population have increased. As a result, in the past, a super strong non-first-gen student might be cross-admitted by 2 or 3 of HYP, but today the chance of being cross-admitted by 2 or 3 of HYP is probably lower, thus leading to a higher yield for all of HYP among this student population.
This is just hypothesizing, but I think the ridiculously high yield across many top schools could very likely be a result of Trump’s election. People are now terrified of the future, so they go with the most secure, sure path to success in their minds (reducing the proportion of students who choose to go elsewhere because the fit is a little better).
@StudiousMaximus, you are making me laugh: “People are now terrified of the future”. :))
Actually, given the rise of SCEA/ED the rise of yield should follow. As a pure hypothetical excise, consider 100 kids who would be cross admitted to all top ten schools. In previous year only ten of them applied EA/ED and ninety did RD. So each school would have issued 1 early admit and 90 during RD. Here, for simplicity I assume the interests in ten schools are even and all kids admitted in early round were satisfied and not applying in RD (but if they did the results would not have changed drastically). So, in this previous year the yield of the ten schools would be 10/(1+90)=11%. Now, this year the same 100 kids all decided to do SCEA/ED, and again for simplicity they were evenly split among ten schools and no further application in RD. Now the yield this year would be 10/10=100% for these 100 kids.
Given that 50% of admits nowadays are from early round I think any uptick in SCEA/ED among the super strong applicant pool would raise the yield.