Yield Protection by Location

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

SingleCellPremed

Full Member
Joined
Sep 27, 2021
Messages
16
Reaction score
12
I'm aware that many schools engage in yield protection to some degree, particularly when they don't want to waste resources on extremely qualified applicants who have lots of options to choose from.

Do these schools also engage in yield protection for applicants who seem more likely to matriculate to a school located within their own state? Some states have medical schools that interview and accept a very high percentage of their own premeds. For instance, in the previous application cycle, Indiana interviewed 466/683 of their own state's students, and 291 matriculated. Would a school like Drexel, which received 12000+ OOS applications last cycle, be hesitant to interview a moderately qualified applicant from Indiana?

And conversely, would applicants from places like California have a greater likelihood of receiving interviews from low yield schools, solely based on the fact that Californian premeds have increased difficulty gaining acceptance to schools within their state (and therefore are more likely to seriously consider OOS schools across the country)?

These questions came to mind after talking to someone who received interview invitations from both of her state schools but nothing so far from OOS ones.

What do y'all think of this? Do you believe this is a significant consideration for low-yield schools that are searching for ways of cutting down their massive application piles?

Members don't see this ad.
 
Last edited:
  • Like
Reactions: 1 user
N=1 but I think this may have happened to me! I live 4 hours from a neighbor state school and was swiftly rejected, only to receive an II from my own state school (20 min away) the next morning. I was an excellent mission fit for the neighbor school but also my own state school (public health + social justice focus). I looked at the OOS tuition for the neighbor school and it does seem very unlikely that I would choose to attend and pay it if the option for my in-state school was available.

In short, it’s possible they rejected me because I am likely to get accepted to my nearby state school and matriculate there. I’ll never know if that’s what happened but it makes sense and also matches what I would have actually chosen if the scenario presented itself, ha. It’s also possible they rejected me for any of many possible other reasons, of course!
 
  • Like
Reactions: 2 users
I'm aware that many schools engage in yield protection to some degree, particularly when they don't want to waste resources on extremely qualified applicants who have lots of options to choose from.

Do these schools also engage in yield protection for applicants who seem more likely to matriculate to a school located within their own state? Some states have medical schools that interview and accept a very high percentage of their own premeds. For instance, in the previous application cycle, Indiana interviewed 466/683 of their own state's students, and 291 matriculated. Would a school like Drexel, which received 12000+ OOS applications last cycle, be hesitant to interview a moderately qualified applicant from Indiana?

And conversely, would applicants from places like California have a greater likelihood of receiving interviews from low yield schools, solely based on the fact that Californian premeds have increased difficulty gaining acceptance to schools within their state (and therefore are more likely to seriously consider OOS schools across the country)?

These questions came to mind after talking to someone who received interview invitations from both of her state schools but nothing so far from OOS ones.

What do y'all think of this? Do you believe this is a significant consideration for low-yield schools that are searching for ways of cutting down their massive application piles?
Not sure exactly what you mean by "low yield schools" or "yield protection." What are you trying to say?
 
Members don't see this ad :)
Not sure exactly what you mean by "low yield schools" or "yield protection." What are you trying to say?
These are terms I have seen on SDN and Reddit before. Here is how I would describe them:

By "low yield schools" I mean schools that interview/accept a low proportion of applicants. Think of yield as the ratio of accepted applicants to total applicants. Tufts is a great example of a low yield school. Out of 11,845 OOS applicants who applied last cycle, only 167 of them were interviewed. Tufts must repeatedly turn down highly qualified applicants in place of less qualified ones that are more likely to matriculate.

By "yield protection" I mean the practice of schools rejecting applicants who seem unlikely to matriculate even if they get accepted. This is done to conserve resources like time, money, and interview slots.

Lets say a mid-tier school decides not to yield protect. Maybe they need to interview and accept 50 extremely qualified applicants just for one of them to actually matriculate. The cost is too great. How many people would realistically turn down an acceptance at Harvard or their state school for a random average ranked school located across the country? Now this school has to be less picky with who they accept because a large fraction of their interviews were spent on people who will not matriculate.
 
Last edited:
  • Like
  • Love
Reactions: 1 users
The simple, general answer to OP's question is that resource protection generally does not occur with IS applicants. Geographic proximity and cost advantages usually address concerns regarding the sincerity of the interest of very strong applicants in all MD programs, regardless of tier, or whether they are low yield.

As far as CA applicants having a better than average shot at OOS low yield schools, that's an interesting theory, and it's certainly possible, since CA applicants definitely punch above their weight OOS. There is no way to really know for sure, though, since CA applicants also have higher than average stats, and low yield schools are low yield for everyone. If they are a little less low yield for CA applicants, it would be impossible for us to know whether that's because they are receiving extra consideration because of their IS situation, or because they are just stronger candidates than other OOS applicants.

A school like Drexel is a great example. It's low yield because its stats are attainable for just about everyone and it is in a desirable city, so it attracts a ton of applications. If you are not receiving an IS preference there (IS II rate = 37%; OOS II rate = 10%), I doubt it matters too much whether you are coming from Indiana, or any other state, other than maybe those in the Philly metro area, like NJ and DE.
 
Last edited:
  • Like
Reactions: 1 users
Resource protection is a more accurate term.

If Drexel were to reject someone from Indiana, then they'd also be more likely to reject people from the 40+ states that have state medical schools.

CA is a next exporter of premeds, and one would simply have to look at the school-specific data from AAMC as to where matriculants come from, and go.
 
  • Like
Reactions: 2 users
These are terms I have seen on SDN and Reddit before. Here is how I would describe them:

By "low yield schools" I mean schools that interview/accept a low proportion of applicants. Think of yield as the ratio of accepted applicants to total applicants. Tufts is a great example of a low yield school. 1. Out of 11,845 OOS applicants who applied last cycle, only 167 of them were interviewed. Tufts must repeatedly turn down highly qualified applicants in place of less qualified ones that are more likely to matriculate.

2. By "yield protection" I mean the practice of schools rejecting applicants who seem unlikely to matriculate even if they get accepted. This is done to conserve resources like time, money, and interview slots.


Lets say a mid-tier school decides not to yield protect. 3. Maybe they need to interview and accept 50 extremely qualified applicants just for one of them to actually matriculate. The cost is too great. 4. How many people would realistically turn down an acceptance at Harvard or their state school for a random average ranked school located across the country? Now this school has to be less picky with who they accept because a large fraction of their interviews were spent on people who will not matriculate.
Re: your example w/Tufts, this is just one school with typical applicants/matriculants stats. I don't see what stats you are comparing this to. Tufts is a private school that typically accepts more OOS students anyway. Like Goro said, you should obtain a MSAR which breaks down IS/OOS applicants/matriculants by school, and by state.
Re: your second point, why would students spend thousands of dollars, and hundreds of hours applying only to "seem unlikely to matriculate?" And how would adcoms even know this??
Re: the third bolded point, these numbers are WAY off. Thousands of students apply for 100-150 or so spots per incoming class.
Re: your fourth point, not too many.
Your last sentence just plain makes no sense.
 
A school like Drexel is a great example. It's low yield because its stats are attainable for just about everyone and it is in a desirable city, so it attracts a ton of applications. If you are not receiving an IS preference there (IS II rate = 37%; OOS II rate = 10%),
Interestingly, Drexel is one Pennsylvania school that I would say does practice resource protection with its instate students. If you are a high stat, instate Drexel applicant, I’d be surprised if they sent you an II. If you look at the tracker, it’s heavily skewed towards 513 and lower. In real life that’s what we’ve noticed as well. (*I don’t blame them given most schools have enough matriculation data to guide how they want to spend their finite resources.)
 
  • Like
Reactions: 1 user
Re: your example w/Tufts, this is just one school with typical applicants/matriculants stats. I don't see what stats you are comparing this to. Tufts is a private school that typically accepts more OOS students anyway. Like Goro said, you should obtain a MSAR which breaks down IS/OOS applicants/matriculants by school, and by state.
Re: your second point, why would students spend thousands of dollars, and hundreds of hours applying only to "seem unlikely to matriculate?" And how would adcoms even know this??
Re: the third bolded point, these numbers are WAY off. Thousands of students apply for 100-150 or so spots per incoming class.
Re: your fourth point, not too many.
Your last sentence just plain makes no sense.
1. The data I am using comes from MSAR. I specifically referenced Tufts as an example because it is an outlier in terms of OOS matriculation stats. Tufts must engage in resource protection at the interview level if they have such a high conversion rate between OOS applicants interviewed (146) and OOS applicants matriculated (167). Look at all of the other OOS-friendly schools and you will see that none of them have a conversion rate that comes even remotely close. Most of the other OOS-friendly schools are willing to interview at least ~3x the number of OOS applicants that matriculate. This means that Tufts is extremely selective with who they interview from OOS, basically only choosing applicants who have them as their #1 choice (over their other acceptances).

2. Adcoms can easily spot exceptional applicants by looking at their AMCAS applications. Top-tier applicants with perfect stats, good experiences, and well written essays will definitely stand out. These are the same applicants who have 15+ interview invites and multiple acceptances at highly ranked schools. With so many options, they are unlikely to matriculate at any one given school.

3. Again, number I used was referring to top-tier applicants. Think of the those who have a WARS score of 100+. Maybe 50 applicants with a WARS score of 100+ need to be interviewed at a random mid-tier school for just one to matriculate. The other 49 decide to go to Harvard, Hopkins, or Stanford. On the other hand, maybe only 3 applicants with a WARS score of ~70 need to be interviewed before one of them matriculates at that mid-tier school.

4. If a random mid-tier school uses 50 interviews on highly overqualified applicants with no ties to the region just to have one of them matriculate, then they have essentially wasted the other 49 interviews. The remaining pool of students that the school can accept has now been decreased. This means the school has fewer options to build their desired class with. Those 50 interviews could have been spent on less qualified applicants who probably only had 1 or 2 other schools to choose for matriculation.
 
  • Like
Reactions: 3 users
Re: your example w/Tufts, this is just one school with typical applicants/matriculants stats. I don't see what stats you are comparing this to. Tufts is a private school that typically accepts more OOS students anyway. Like Goro said, you should obtain a MSAR which breaks down IS/OOS applicants/matriculants by school, and by state.
Re: your second point, why would students spend thousands of dollars, and hundreds of hours applying only to "seem unlikely to matriculate?" And how would adcoms even know this??
Re: the third bolded point, these numbers are WAY off. Thousands of students apply for 100-150 or so spots per incoming class.
Re: your fourth point, not too many.
Your last sentence just plain makes no sense.
Actually, I see where @SingleCellPremed is going with this. I'd have said it differently, but he's not wrong.

As a currently applicant, I can tell you from first hand experience that applicants DO spend thousands of dollars, and hundreds of hours, applying to schools they are never going to matriculate at. This is because, at the beginning of the cycle, we are all scared out of our minds by everything we read and hear, so we apply far and wide. Even the high stat URMs, who apply to 25 schools and end up with 20 IIs and 15 As. Believe me, they aren't doing it just to collect trophies and fin aid awards. They do it because they have no idea how desirable they are before results start coming in.

The same goes for mere mortals. We apply all over the place because we have no idea how things are going to shake out for us. Adcoms, OTOH, do this year in and year out. They know exactly how things are going to play out, and they allocate their resources accordingly.

As to the next point, he means that, with no resource allocation, schools would spin their wheels wasting time and interview slots with people where they would maybe have a 5% yield rather than a more normal 30-50% yield, and then be left scrambling in May to fill a class off the WL, and possibly even be left with empty seats in the class as a majority of the WL evaporates, accepting offers at "better" schools.

You're absolutely correct about his last point not making sense, because if the school wastes so many II slots on people who are never going to attend, the risk is not having to be less picky, it's running out of people to pick from at all at the end of the cycle. How embarrassing would it be to have to reopen IIs in May to fill the class??

THIS is why resource protection is a very real, and very legitimate thing, everywhere outside the T10. Even the T20 do it with some T10 caliber applicants. Yes, it stings when you are an applicant who really wants to go to Drexel, or GW, or Tufts, or BU, and you can't score an II. Also, no one knows until the IIs and As starting coming in just how their cycle is going to go, so you have to apply widely. But schools use experience gained over many cycles to screen candidates out for many reasons, and this is one of them. It makes total sense from their perspective.
 
Last edited:
1. The data I am using comes from MSAR. I specifically referenced Tufts as an example because it is an outlier in terms of OOS matriculation stats. Tufts must engage in resource protection at the interview level if they have such a high conversion rate between OOS applicants interviewed (146) and OOS applicants matriculated (167). Look at all of the other OOS-friendly schools and you will see that none of them have a conversion rate that comes even remotely close. Most of the other OOS-friendly schools are willing to interview at least ~3x the number of OOS applicants that matriculate. This means that Tufts is extremely selective with who they interview from OOS, basically only choosing applicants who have them as their #1 choice (over their other acceptances).

2. Adcoms can easily spot exceptional applicants by looking at their AMCAS applications. Top-tier applicants with perfect stats, good experiences, and well written essays will definitely stand out. These are the same applicants who have 15+ interview invites and multiple acceptances at highly ranked schools. With so many options, they are unlikely to matriculate at any one given school.

3. Again, number I used was referring to top-tier applicants. Think of the those who have a WARS score of 100+. Maybe 50 applicants with a WARS score of 100+ need to be interviewed at a random mid-tier school for just one to matriculate. The other 49 decide to go to Harvard, Hopkins, or Stanford. On the other hand, maybe only 3 applicants with a WARS score of ~70 need to be interviewed before one of them matriculates at that mid-tier school.

4. If a random mid-tier school uses 50 interviews on highly overqualified applicants with no ties to the region just to have one of them matriculate, then they have essentially wasted the other 49 interviews. The remaining pool of students that the school can accept has now been decreased. This means the school has fewer options to build their desired class with. Those 50 interviews could have been spent on less qualified applicants who probably only had 1 or 2 other schools to choose for matriculation.
I'm generally supporting you here, so please don't get upset with what I'm, about to say.

Most of what you are saying is valid, but it's important to apply a smell test to data before you use it to support an argument. The MSAR data makes no sense. Is it really possible that ANY school would have such a crazy OOS conversion rate as compared to its IS one?

Before you get to conversion, you first to have an acceptance. The highest acceptance rates are at schools like Michigan and Virginia, public schools where their IS acceptance rates exceed OOS by a fairly significant margin. If Tufts actually only interviewed 167 OOS applicants, it would be unbelievable if they even accepted 146 of them (87%), let alone enrolled that many!

No private school has an acceptance rate approaching that level. After that, it's important to remember that NO ONE has anything close to a 100% yield. Not Harvard. Not NYU. And not Tufts with its OOS candidates.

It is FAR more likely that MSAR is inverting the IS and OOS II numbers. USNWR reported data confirms that is in fact the case.
 
Last edited:
I got rejected by a low-mid tier DO like two weeks after I submitted my secondary. I figured with how fast it was, it was probably due to this, since my GPA would've been on par but my MCAT would've been around 10 points over their median and I was very far away from the school too (as opposed to closer ones with lower stats that haven't rejected me).
 
  • Like
Reactions: 3 users
I got rejected by a low-mid tier DO like two weeks after I submitted my secondary. I figured with how fast it was, it was probably due to this, since my GPA would've been on par but my MCAT would've been around 10 points over their median and I was very far away from the school too (as opposed to closer ones with lower stats that haven't rejected me).
I'm surprised they did this. My school would have given you an II. The mindset of "they won't come here" is just simply not used in our Adcom meetings.
 
  • Like
Reactions: 1 user
Members don't see this ad :)
I'm surprised they did this. My school would have given you an II. The mindset of "they won't come here" is just simply not used in our Adcom meetings.
I'm confused. You just acknowledged the existence of resource protection being a thing a few posts above, and now you're saying your school doesn't do it. Are you saying it's not a thing at DO schools, just at your school, or am I missing something?
 
I'm confused. You just acknowledged the existence of resource protection being a thing a few posts above, and now you're saying your school doesn't do it. Are you saying it's not a thing at DO schools, just at your school, or am I missing something?
I think he might be saying that geographic distance isn't one of the factors they use in yield protection at his school. MCAT 10 points above median also might not be enough for yield protection to kick in at his school..
 
I think he might be saying that geographic distance isn't one of the factors they use in yield protection at his school. MCAT 10 points above median also might not be enough for yield protection to kick in at his school..
Okay, but, if that wouldn't warrant resource protection, what would? 10 points on a 56 point scale, where you really need to be in the top 28 points to even be deemed qualified, is a LOT!! If you're not looking at geography and stats, the only thing really left is old regular screening!
 
  • Like
Reactions: 1 user
Okay, but, if that wouldn't warrant resource protection, what would? 10 points on a 56 point scale, where you really need to be in the top 28 points to even be deemed qualified, is a LOT!! If you're not looking at geography and stats, the only thing really left is old regular screening!
well their gpa is within the normal range. For me, my MCAT is much better than my GPA, so I’m hoping I don’t get screened out at low/mid tier schools…
 
I'm hoping that resource protection may manifest as follows for some applicants/schools: If the med school believes the applicant's mcat and gpa are too high for their school of med., they could merely place a pre-II Hold on that applicant, then offer, say, not until Dec. an II for interview in Jan. 2022. By then, if the applicant already has an A from their local, less expensive state school, they're more likely to simply decline the II from the far away (and often higher tier) school that is practicing resource protection. Idk, but I'm hoping something like this could apply to my case (very high stats, but nearly no IIs from medium and high tier schools that are across the county. Though, it could alternatively be application weaknesses outside of high mcal/gpa, such as inadequate community service hours, or subpar PS.)
 
I'm hoping that resource protection may manifest as follows for some applicants/schools: If the med school believes the applicant's mcat and gpa are too high for their school of med., they could merely place a pre-II Hold on that applicant, then offer, say, not until Dec. an II for interview in Jan. 2022. By then, if the applicant already has an A from their local, less expensive state school, they're more likely to simply decline the II from the far away (and often higher tier) school that is practicing resource protection. Idk, but I'm hoping something like this could apply to my case (very high stats, but nearly no IIs from medium and high tier schools that are across the county. Though, it could alternatively be application weaknesses outside of high mcal/gpa, such as inadequate community service hours, or subpar PS.)
That's just not how this works, especially in a virtual world where IIs cost applicants nothing but a little time, and where the payoff is potential As to stroke their egos and scholarships that could be used to leverage money from other schools. Holds are definitely a thing, but not to play games to see whether or not someone will accept a late II.
 
This thread is really hitting close to home as I'm preparing my list for next year's cycle. ORM, 4.0 GPA, MCAT 52x, HYP, etc. All boxes checked and ready to apply...but don't see any other option but to cast a wide net to secure an A...and throw money down the drain at schools where I will be resourced protected and quickly thrown an R pre-II. No way around it it seems.
You’re from CA?
 
That's just not how this works, especially in a virtual world where IIs cost applicants nothing but a little time, and where the payoff is potential As to stroke their egos and scholarships that could be used to leverage money from other schools. Holds are definitely a thing, but not to play games to see whether or not someone will accept a late II.
Oh, okay thanks; I forgot to consider that detail about leverage that you mentioned here. So then, I'm just not doing well (with this app./cycle). The pre-II holds are a negative sign then in my case.
 
  • Care
Reactions: 1 user
This thread is really hitting close to home as I'm preparing my list for next year's cycle. ORM, 4.0 GPA, MCAT 52x, HYP, etc. All boxes checked and ready to apply...but don't see any other option but to cast a wide net to secure an A...and throw money down the drain at schools where I will be resourced protected and quickly thrown an R pre-II. No way around it it seems.
Ask your health professions advisors for data where previous alum have applied and been accepted. I’m also HYP (with much lower stats), and my advisor gave me a list that details which med schools “like” our graduates and which schools have literally never accepted a single one.
 
  • Like
Reactions: 1 users
This thread is really hitting close to home as I'm preparing my list for next year's cycle. ORM, 4.0 GPA, MCAT 52x, HYP, etc. All boxes checked and ready to apply...but don't see any other option but to cast a wide net to secure an A...and throw money down the drain at schools where I will be resourced protected and quickly thrown an R pre-II. No way around it it seems.
THIS^^^^. The problem is, you really only know in hindsight how strong you really are. The reapplicant pool has many people with your stats (well, maybe not many, but still :)) in it.

Overall acceptance rate last year was around 36%. 16% of people 3.8+/518+ had zero As over the past 3 cycles. Since that number has been going up every single year since I started paying attention, I promise you it was significantly higher than that just last year.

Yes, 84% is far better than 36%, and the odds are very high that you will be successful, given your stats. OTOH, you would be playing with fire if you chose not to risk "throwing money down the drain" and sticking with T20s, which is where you are likely to end up. At least a slice of that 16% had stats like yours and also didn't see the point of throwing away money.

It's not as bad as it seems. Instead of applying to 20 or so schools, just add another 10 to the mix and spread them around. It's $1,500 worth of insurance, which is reduced to less than $500 if you are eligible for FAP.
 
Last edited:
Oh, okay thanks; I forgot to consider that detail about leverage that you mentioned here. So then, I'm just not doing well (with this app./cycle). The pre-II holds are a negative sign then in my case.
It really is far from over, and a LOT of people report success with IIs coming between December-March. It's very unnerving and sucks, but it's not quite the end of the world. The holds also can and do turn into IIs, but, no, they are not great and are not really a good sign. The better sign is everyone you still haven't heard from, because some of them could turn into IIs if they just haven't reviewed you yet, as opposed to putting you on hold or rejecting you and just not letting you know.
 
  • Like
Reactions: 1 users
Top