hopefulincal wrote:I'm continuing this discussion only because I think it's an interesting business case. Obviously depending on the sample you present, you could argue one way or the other. I'd like to think about how this process actually happens:
September comes around, and applications start rolling in. For the sake of the argument let's drop the whole holistic view thing. The school starts getting apps, and decisions need to be made. Every school knows their previous medians and would like to increase those numbers. Thus, you pretty much auto-admit those with both numbers higher than your medians (sure you worry about yield, but let's put that aside for the moment). You also pretty much admit those with one number higher and the other number at median. Yes you can throw money at any of the above students to get them to enroll, but so are the other schools with similar medians. So, pretty soon you have an admitted pool, and you would know what your current medians are from the pool. You now monitor the medians for this pool closely. As you decide whether or not to admit each incoming application, you know how that applicant is going to affect your pool medians, and you weigh the gains and losses for each admit.
Meanwhile, your similarly ranked competitors are doing the same thing (which is why you have similar medians in the past). Now along comes an applicant that every school considers a splitter. Do you admit him/her? You look at how this applicant affects your medians. Most likely if admitting this applicant will help your numbers, it will likely help your competitors' numbers as well, so both of you will end up admitting him/her.
Yes, in the perfect world if you had all the applications laid out in front of you, you could figure out how to maximize your medians from the entire pool, but it doesn't work that way. Especially given the faster turnaround times for ED, you can't just sit and wait to see what your pool would look like. So you couldn't possibly go into the season with a splitter strategy because with every new admit you're risking your other number. If the distribution for both numbers are in fact very different and admitting one pool of splitters would easily raise the medians, you don't think your competitors would have figured that out also?
These rankings mean life and death to the Admissions Office, and if there is a low hanging fruit like admitting splitters, many schools would have done it already.
I'd be shocked if the numbers of your applicant pool varied that much year-by-year. It should be fairly easy to predict close to how many people with whatever numbers you want start applying. There's some risk involved that all of a sudden a lot of applicants you admit won't matriculate, but for this is probably steady year-to-year, as well.
And no, there's plenty of 170+ splitters that routinely get rejected from T-14 schools. The only schools in the T-14 that have a reputation for accepting them are UVA and Northwestern (but at NU only if you have work experience). Outside of that, you're an automatic reject at almost all of them unless you have around a 174+ and really good softs.
If UVA admits these people, they really have nowhere else to go in the T-14. It helps them raise their medians, raise their yield rate, and they really don't have to offer them scholarship money to get them to attend. It's a win-win-win situation for them.