WedgeDawg's Applicant Rating System (Updated Jan 2017)

[efle]

After seeing many very similar applicants in previous cycles (and this current one) receive interviews at places that I clearly wasn't even one of their top ~700 most desirable applicants, yeah I was surprised I didn't get more interviews. Maybe I got unrealistic expectations from SDN, but I really thought my application was strong enough that more of the top schools would at least want to meet me in person before waitlisting or rejecting me.

For perspective, look at all my similar-stat contemporaries on SDN this cycle.

Damn dude, people always say stats get you interviews and ECs get you acceptances, so how the hell is an 80 LizzyM getting only a handful of Top ~15 interviews despite applying to almost all of them? When you rule out undergrad reputation mattering much what else would cause this

---

Members don't see this ad.

 

[NeuResurgens]

Good idea, just will need some specifics in the future as far as which activities fall under which sub point for each of these categories. Also, I thought undergraduate college did not factor as heavily into admissions decisions as this weighted score places on it? (I was doing well until I hit that one, so naturally I have to challenge it. Haha)

 

[NeuResurgens]

Damn dude, people always say stats get you interviews and ECs get you acceptances, so how the hell is an 80 LizzyM getting only a handful of Top ~15 interviews despite applying to almost all of them? When you rule out undergrad reputation mattering much what else would cause this

I am also a bit confused by your results, to be perfectly honest. There are people with stats slightly below yours with more top 15 interviews.. Medical school admissions are screwy at times.. But hey, you got in though!

 

[justadream]

Some people have a hook that sticks out, maybe their PS or one of their letters. At a certain point, you essentially get to flip a coin to see if you get an interview or not. The undergrad game is similar... only a few people get into all the Ivies every year, but many of them get into a top school.

They might have just liked someone with a similar app to yours over you. It happens. I applied to 45 schools and got 7-8 interviews. I was expecting to get interviews at about a quarter of places I applied to and I got a couple less. The timing of some of those interviews really surprised me as well since it felt like I was on the cusp for a lot of schools and many passed me over till the very last minute, including the school where I will likely be matriculating.

@gettheleadout is saying that his super high stat's (and general well-rounded application) should have warranted more interviews than he received. With his stats, I think most would have expected him to get an interview at almost every school (remember, getting an interview is often a lot easier than getting an acceptance). IIRC, his stats were like 90th percentile level at even Harvard-level schools.

 

[RogueUnicorn]

Thank you very much for your input. I would like to comment on your statement and say that this is just something that I've pulled together in my spare time as an additional tool that might help new applicants start to create a school list. I never suggested that this is an end-all-be-all for any particular applicant in terms of creating a school list. I 100% agree that it is very difficult to determine how med schools evaluate applicants or where applicants should apply - this is very clearly shown by what you and others have said about how this classified you and them and how those results proved to be contrary to what this showed (if I recall correctly, you attend a Category 1 school).

My intention is definitely not to "throw lots of numbers and fancy sounding stuff" at people, my intention is to help people who aren't sure where to start quickly and easily come up with a preliminary list of schools to apply to. If you think that there is no merit to this system at all, I strongly encourage you to recommend against it, but I even more strongly encourage you to help improve it so that applicants might benefit from an additional tool to add to their repertoire in a highly complex and often confusing process.

There is a lot in this system that is arbitrary, I agree - however, many of the arbitrations that can be found in this algorithm seem to elicit results that at least in part are accurate. One of the reasons I've so intently tried to solicit community feedback is to try and fine-tune the algorithm and thus reduce the arbitrary parts and increase the evidence-based parts.

Again, this is just something I came up with in my spare time based on what I've seen throughout my cycle and in the WAMC as well as the pre-allo forum in general. If people do not thing this is a useful tool, I have no problem scrapping it - at worst, it would be an interesting experiment that kept me from being bored on a long bus ride. However, if people see merit in it and wish to devote time and effort to continually improving it for the benefit of future applicants, I am of course all for that as well. It is now up to you to decide whether you would like to see this project scrapped or if is worth investing yourself in. Either way is fine with me.

I very much appreciate your input regardless of whether or not you choose to invest in this project.

I commend you for making a hearty effort, but once again I have to emphasize that this to me seems an overwrought attempt at quantifying the unquantifiable. More importantly, I take exception to your claim of "evidence-based." Other than the gpa/mcat stats, which are hard, verifiable numbers, there is nothing here that is based on evidence to any credible scientific standard. What you have created is a mathematical reduction of the SDN echo-chamber, but (though not your goal) for all intents and purposes look believably scientific. While I subscribe to the notion that perfect shouldn't stand in the way of better, I think misguided is worse than nothing.

I think your efforts are better directed at trying to categorize schools and leaving out applicantion suggestions altogether. There is much information to be found through SDN about what kinds of things a certain school might emphasize. For instance, at my school (Columbia), your extracurriculars are going to impact your application much more than at a similarly ranked school. Just providing a guidance with regards to school preferences, and allowing premeds to decide for themselves independent of a pseudo-algorithm imo is much better.

 

[efle]

@gettheleadout is saying that his super high stat's (and general well-rounded application) should have warranted more interviews than he received. With his stats, I think most would have expected him to get an interview at almost every school (remember, getting an interview is often a lot easier than getting an acceptance). IIRC, his stats were like 90th percentile level at even Harvard-level schools.

Maybe Harvard was like oh snap he's too good better yield protect no II

For instance, at my school (Columbia), your extracurriculars are going to impact your application much more than at a similarly ranked school.

And yet Columbia has stats just like similarly ranked schools

 

[maccu5000]

I think you should start collecting data to evaluate whether the tool "works":

People's feedback is good but you'll only know how good it is if you can see how the model shakes out.

 

[Rainbow Zebra]

I tweaked it a bit, due to 2 MCATs (had an 8 point jump, so didn't want to average, so I just dropped one half point from a 7 to a 6.5 for Stats), and some of my ECs were 2.5 or 3.5, not full integers. So, this is my school list by category (reclassed U of Michigan, as the stats for IS are 1.5 points lower on average for MCAT than OOS). I am applying this next cycle at 24 schools (5 of which are in Michigan), and most other schools are in the midwest.

Category 1 5% 2 8.33%
Category 2 15% 6 25.00%
Category 3 25% 4 16.67%
Category 4, 5 55% 12 50.00% (moved U of M out of Cat 2 for IS)

All in all, pretty darn close predictor. I would put in some multiplier effect for the poor CA residents that discounts their chances instate, and maybe increase the multiplier for folks in Florida and Michigan and New York (states with a lot of of IS options). Hopefully with a 73, I'll get in somewhere. Was a nice validation that my research on schools to apply to was independently and objectively verified. Many thanks for the downloaded spreadsheet.

 

[resiroth]

As someone else that didn't exactly sweep the top 5 with similar stats I think you guys may be in for a treat when you apply. SDN is far from representative. I think at some point (say 3.8/37 and up) adcoms stop caring as much (I'm sure Harvard could have 41 mcat as their median if they really wanted, yet it's actually "only" a 37). Search MDapps for MCATs >38 and you'll find a handful of people like http://www.mdapplicants.com/profile.php?id=29825 and http://www.mdapplicants.com/profile.php?id=29465 and it wouldn't surprise me if mdapps were also weighted toward more successful profiles.

 

[WedgeDawg]

I commend you for making a hearty effort, but once again I have to emphasize that this to me seems an overwrought attempt at quantifying the unquantifiable. More importantly, I take exception to your claim of "evidence-based." Other than the gpa/mcat stats, which are hard, verifiable numbers, there is nothing here that is based on evidence to any credible scientific standard. What you have created is a mathematical reduction of the SDN echo-chamber, but (though not your goal) for all intents and purposes look believably scientific. While I subscribe to the notion that perfect shouldn't stand in the way of better, I think misguided is worse than nothing.

I think your efforts are better directed at trying to categorize schools and leaving out applicantion suggestions altogether. There is much information to be found through SDN about what kinds of things a certain school might emphasize. For instance, at my school (Columbia), your extracurriculars are going to impact your application much more than at a similarly ranked school. Just providing a guidance with regards to school preferences, and allowing premeds to decide for themselves independent of a pseudo-algorithm imo is much better.

I'm not entirely sold on the idea that creating categories with no means of navigating them will be any better than the system that already exists, particularly it seems that a lot of people don't know how to construct a school list. Let's pretend that a 3.8/33 applicant with strong research is trying to come up with a school list. Let's also pretend that we've categorized schools into very well defined groups based on stats and what they're looking for or what they value. This applicant would primarily be looking at schools in the group (for instance) Medium Level Stats with High Emphasis on Research. That group would probably consist of very few schools. So the applicant at that point either has to expand their list by looking into groups that they fit the profile of less or they are stuck with very few schools. Obviously this is a somewhat extreme example.

I think what this tool does is tell you how good of a "cookie cutter" applicant you are - i.e. do you meet the minimum thresholds to be in general, considered a competitive applicant for this very general category of schools? Obviously within categories you will have different schools and even adcoms having their own preferences for particular things, but I would find it hard to believe that a person who is competitive for, say, Columbia, would not be also a competitive applicant to, say, Duke (on an unrelated and incidental note, I will be attending the same school as you next year). These preferences are not always immediately transparent and it's not going to be easily identifiable if the applicant has the precise characteristics that a single particular school is looking for. This is why people apply to high numbers of similar schools where they believe they are competitive - it is first the admissions committee's job (pre-interview) and then both the school's and applicant's job (during the interview) to decide whether or not the applicant would be a good fit for that particular school. This tool helps people decide broadly where to apply and then should be fine-tuned based on the applicant's specific interests and strengths. It provides a basic structure to help set up a school list which can then be continually refined by looking at unique aspects of an application.

I do not pretend that this is an end-all-be-all algorithm, it is merely another resource to help premeds decide how to construct a preliminary school list. If people opted to use this first, then come up with a school list, and then post a WAMC thread, I think that they would be able to get a lot more specific help that is tailored to their own application because the broad general structure will already have been generated beforehand. Right now, WAMC threads are essentially "here's a brief profile of my application make me a school list" and then you have people basically saying apply to some of these schools - which is exactly what this tool does. After that point, the applicant and anyone trying to help out can take a more detailed look at the application and the school list and say "I know you are applying to school x, but I think you might be a better fit for school y based on a, b, and c - perhaps consider swapping them?". This isn't a revolutionary or novel system, just a convenient and efficient way to gauge very generally an applicant's level of competitiveness.

And that's not to say that cookie cutter students can't be successful at schools that value other things - I am one of them, and I'm sure that there are many others. Even if you do have something very unique that happens to be what a specific school ostensibly values, it would be unwise to apply only to that school and ignore similar ones. Who knows, maybe something on your app will catch someone's eye that you didn't anticipate.

 

[WedgeDawg]

Additionally, if anyone would like to contribute actual data, I have a google doc set up to do just that.

Copy and paste the template and enter information only in the green boxes - everything else is automatic.

Thanks!

[link removed]

 

[mehc012]

As someone else that didn't exactly sweep the top 5 with similar stats I think you guys may be in for a treat when you apply. SDN is far from representative. I think at some point (say 3.8/37 and up) adcoms stop caring as much (I'm sure Harvard could have 41 mcat as their median if they really wanted, yet it's actually "only" a 37). Search MDapps for MCATs >38 and you'll find a handful of people like http://www.mdapplicants.com/profile.php?id=29825 and http://www.mdapplicants.com/profile.php?id=29465 and it wouldn't surprise me if mdapps were also weighted toward more successful profiles.

Only if every single 40+ applicant from each year's sitting of the MCAT joined their class...and even then it's close.
Looking at the 2013 stats, if 99.8% of students scored below a 40, that means that <200 people scored 40+ that year (we'll assume that 40+ students typically get in during their first cycle).
HMS has a class size of ~230 students. So at least 15% of the students are 2 points below a 41, and odds are that 40+ group is still weighted towards the 40s and 41s. Is it possible that HMS could end up with a 41 median? I suppose. But they'd have to care only about MCAT and everyone would have to choose HMS over any other option (so, really good finaid I guess?)

Sorry, I just found the thought experiment interesting. It's a little grounding to realize that no, even top schools couldn't fill themselves up with 40+ applicants. Good reminder after a day on SDN, no?

 

[MaybeDr]

Fantastic job!

There definitely needs to be more schools on the cumulative list though.

I think something people are failing to appreciate here is that the process is random. There has been a few posts trying to simplify the process or explain a phenomena of why some with a 80+ didn't do well the last application cycle. To be completely blunt no one knows jack about the process as whole. No everyones application goes through the same pathway. Stray away from making blanket assumptions. 80+ doesn't guarantee you anything if you can't write, you aren't socially aware, the list goes on.

 

[MaybeDr]

@WedgeDawg Also if you could clarify your metric for determining low yield that'd be awesome.

 

[WedgeDawg]

@WedgeDawg Also if you could clarify your metric for determining low yield that'd be awesome.

Low Yield generally means that for most applicants, the time, effort, and money invested in applying to these particularly schools could better be used elsewhere. In general, these are schools with a very high volume of applicants or a very low interview:applicant ratio.

 

[MaybeDr]

Low Yield generally means that for most applicants, the time, effort, and money invested in applying to these particularly schools could better be used elsewhere. In general, these are schools with a very high volume of applicants or a very low interview:applicant ratio.

Out of curiosity how come you decided to use interview ratio over matriculant ratio? I know matriculant ratio isn't exactly accepted but I'd lean to it being some sort of indicator.

 

[WedgeDawg]

Out of curiosity how come you decided to use interview ratio over matriculant ratio? I know matriculant ratio isn't exactly accepted but I'd lean to it being some sort of indicator.

For most, it wouldn't matter, as the volume of applications for these schools are often in excess of 10,000. The interview thing is mostly for Brown where they interview only 3% of applicants, an outstandingly low number.

 

[resiroth]

Only if every single 40+ applicant from each year's sitting of the MCAT joined their class...and even then it's close.
Looking at the 2013 stats, if 99.8% of students scored below a 40, that means that <200 people scored 40+ that year (we'll assume that 40+ students typically get in during their first cycle).
HMS has a class size of ~230 students. So at least 15% of the students are 2 points below a 41, and odds are that 40+ group is still weighted towards the 40s and 41s. Is it possible that HMS could end up with a 41 median? I suppose. But they'd have to care only about MCAT and everyone would have to choose HMS over any other option (so, really good finaid I guess?)

Sorry, I just found the thought experiment interesting. It's a little grounding to realize that no, even top schools couldn't fill themselves up with 40+ applicants. Good reminder after a day on SDN, no?

89k took it in 2012, 95k in 2013 estimated 100k+ 2014.
In 2013 >.3% scored a 41 or higher, >6We can therefore assume that >300 applicants will score a 41 or higher, enough to fill their class .

HMS has 726 students, which would be 181-182 if it were only MD, but it also includes MDPhD which take twice as long to graduate so its likely closer to ~160.

in spite of all this the median person Harvard accepts has a mere 37, which is matched or surpassed by 3000 people get every cycle .

 

[MaybeDr]

@WedgeDawg dawg The OOS interview to applicant ratio for lets say Stony Brook (231/2835=8.1%) and Penn State (532/6905=7.7%) are almost the same yet Penn State accepts 87 versus Stony Brooks 26. Which makes it really hard to define low yield.

 

[resiroth]

Fantastic job!

There definitely needs to be more schools on the cumulative list though.

I think something people are failing to appreciate here is that the process is random. There has been a few posts trying to simplify the process or explain a phenomena of why some with a 80+ didn't do well the last application cycle. To be completely blunt no one knows jack about the process as whole. No everyones application goes through the same pathway. Stray away from making blanket assumptions. 80+ doesn't guarantee you anything if you can't write, you aren't socially aware, the list goes on.

If an 80+ applies broadly and doesn't get in, THEN you can say that maybe something is wrong with their app. Otherwise it's absolutely absurd to say that someone didn't get into a top 20 because they must have had something wrong with them.

Fun exercise. Take someone with a "successful" cycle with 3 top 20 acceptances. Then imagine what would happen if they didn't apply to 3 out of the 30 odd schools they applied to, and instead applied to 3 other random schools. All of a sudden they "must have had something wrong with them".

You haven't gone through a cycle yet, so trust me when I tell you that you'll be rejected from places that were "safeties" and invited (and maybe if you're lucky accepted) to places so good you thought were a waste of a secondary fee.

 

[WedgeDawg]

@WedgeDawg dawg The OOS interview to applicant ratio for lets say Stony Brook (231/2835=8.1%) and Penn State (532/6905=7.7%) are almost the same yet Penn State accepts 87 versus Stony Brooks 26. Which makes it really hard to define low yield.

True, it is a nebulous indicator and the lines are hard to draw. I think this is certainly a category that can be refined.

 

[medic86]

True, it is a nebulous indicator and the lines are hard to draw. I think this is certainly a category that can be refined.

I'd argue that determining which schools are low yield should be up to the applicant (with the exception of the schools that should clearly fit in this category, like 10,000+ applications per cycle).

If someone has the money and is willing to risk a 5% OOS interview rate and a 5% OOS interviewee matriculation rate, then it's on them (though, I'd call this low yield, myself ).

 

[deleted436726]

3.25 cGPA/35 MCAT reporting in; according to my calculated ARS score of 73 it says that I shouldn't have applied to DO schools at all, but I'm glad I did. However, my luck on the MD side has been more or less in line with the recommendation to apply to Category 4-6 schools.

@WedgeDawg , if you're interested, I'd be more than happy to collaborate with you on creating a web app based on your system. I'd be happy to help you take it to the next level. Let me know via PM if you're interested.

 

[MaybeDr]

True, it is a nebulous indicator and the lines are hard to draw. I think this is certainly a category that can be refined.

I thought you'd agree haha. I've always thought that.

 

[OrbitalOverload]

What exactly is a low-yield school? (category 6)

 

[Glandzburg]

What exactly is a low-yield school? (category 6)

Low Yield generally means that for most applicants, the time, effort, and money invested in applying to these particularly schools could better be used elsewhere. In general, these are schools with a very high volume of applicants or a very low interview:applicant ratio.

This exact questions was just answered several lines ago.

 

[sazerac]

True, it is a nebulous indicator and the lines are hard to draw. I think this is certainly a category that can be refined.

I am surprised that Tulane wound up as low instead of low yield. I always felt its peers were the other "national" med schools like Temple, Tufts, the George's, etc. that all get over 10,000 applications and only interview 500ish.

Strong work!

 

[mehc012]

89k took it in 2012, 95k in 2013 estimated 100k+ 2014.
In 2013 >.3% scored a 41 or higher, >6We can therefore assume that >300 applicants will score a 41 or higher, enough to fill their class .

HMS has 726 students, which would be 181-182 if it were only MD, but it also includes MDPhD which take twice as long to graduate so its likely closer to ~160.

in spite of all this the median person Harvard accepts has a mere 37, which is matched or surpassed by 3000 people get every cycle .

The 230ish number came straight from HMS website info on their MD class size for 2014. The 99.8% was the AAMC data for MCAT 2013 for 40+...not sure where your 0.3% or 180 figure came from.

Either way, the point stands...in order for even HMS to have a median that high, they'd have to snag 100% or close to it of the top test takers, and even so it's a close call and that only covers one med school.

Crazy when you think about it, esp in the context of SDN. Something to keep in mind, is all!

 

[maccu5000]

WedgeDog-you'll also need to collect data on the categories to know if one or a few factors are driving the result more than you've accounted for in the model. This will allow you to adjust the multipliers and make it more accurate.

 

[resiroth]

The 230ish number came straight from HMS website info on their MD class size for 2014. The 99.8% was the AAMC data for MCAT 2013 for 40+...not sure where your 0.3% or 180 figure came from.

Either way, the point stands...in order for even HMS to have a median that high, they'd have to snag 100% or close to it of the top test takers, and even so it's a close call and that only covers one med school.

Crazy when you think about it, esp in the context of SDN. Something to keep in mind, is all!

Yeah, .2% got a 41, but .1% got a 42 and >0 got > 42, so > .3% get 41+. I get 726 MD (with MDPHD) students from here: https://hms.harvard.edu/about-hms/facts-figures and 178 of those are MDPHD so MD only matriculants = ~550/4 = ~138. Add in MDPHD applicants (which take twice as long) /2 /4 = ~22 so overall ~160.

Oh wait lol thought you meant class size...yeah they admit 231, but their class isn't 231. So 231 is the # they'd need of >= 41 MCATs and they'd have >300 by my calculations so they should be good. They wouldn't need to snag any more than they do now, at least yield rate wise.

But most importantly, the fact is they could realistically full it with 40/41+ MCATs but actually most frequently admit people with an MCAT of 37, which is ~10x as non-selective stats wise.

I mean hey, I wish you were right. I wouldn't mind having my pick of the top 5 personally but it's not the way things are.

 

[mehc012]

Yeah, .2% got a 41, but .1% got a 42 and >0 got > 42, so > .3% get 41+. I get 726 MD (with MDPHD) students from here: https://hms.harvard.edu/about-hms/facts-figures and 178 of those are MDPHD so MD only matriculants = ~550/4 = ~138. Add in MDPHD applicants (which take twice as long) /2 /4 = ~22 so overall ~160.

Oh wait lol thought you meant class size...yeah they admit 231, but their class isn't 231. So 231 is the # they'd need of >= 41 MCATs and they'd have >300 by my calculations so they should be good. They wouldn't need to snag any more than they do now, at least yield rate wise.

But most importantly, the fact is they could realistically full it with 40/41+ MCATs but actually most frequently admit people with an MCAT of 37, which is ~10x as non-selective stats wise.

I mean hey, I wish you were right. I wouldn't mind having my pick of the top 5 personally but it's not the way things are.

Still not seeing where your >300 number is coming from, which is my biggest discrepancy.
Also disagree with the 231 not being important. If they admit 231, that's the number the median is to be calculated from.
Regardless, as I said in my last post, the point still stands that HMS could barely, if even, fill itself with 41 median applicants, and this is a good perspective to keep in mind

 

[VitaminVater]

@WedgeDawg

This is pretty cool!
Do you think it applies to MD-PhD applicants too or would you need to tweak it a bit?

 

[resiroth]

Still not seeing where your >300 number is coming from, which is my biggest discrepancy.
Also disagree with the 231 not being important. If they admit 231, that's the number the median is to be calculated from.
Regardless, as I said in my last post, the point still stands that HMS could barely, if even, fill itself with 41 median applicants, and this is a good perspective to keep in mind

100k people take the MCAT a year (rising by ~6k a year). over .3% get a 41 or over: https://www.aamc.org/students/download/361080/data/combined13.pdf.pdf They could actually fill their class with a 42 median mcat (.1% get 42, so admit all 100 of those, admit the almost certainly >15 people that score above a 42, and then admit 115 people that get 41).

Anyway, the point is that despite Harvard being able to fill a class with superstar MCATers, they chose to admit people with MCATs that are 10x as "easy" to get (if you want to think about it like that), which illustrates how little stats matters to Harvard. Heck, don't they admit like 25 people with below a 34?

Either way, I think the current evidence points to =>75 LizzyM (balanced) as not mattering all that much, even though I wish it did.

 

[MrLogan13]

I'm a non-trad student who had a rough start when I first began my academic career. Dropped out after failing out of college, joined the military, and came back and killed my undergrad and am about the finish up an M.S. degree. The ARS score was 50 and your sheet determined me to be an E-level applicant. It advised me to not apply to any cat 1-3 schools, only to 4 and 5, and to not apply to any cat 6's.

I applied to a few schools in all categories and received interviews at schools in cat 3, 4, and 6. I have one acceptance right now at a cat 6, and am waiting to hear back from a cat 3. I think what you're doing is great, but it seems that with applicants such as myself, there are things that are still not quantified in your algorithm; perhaps there are some things you just can't take into account with things like this.

 

[Cyberdyne 101]

Other factors that are difficult to quantify:
1. Some schools seem to be more non-trad friendly than others.
2. As Goro often points out, some schools like rewarding reinvention.
3. URM status for certain applicants might vary from school to school.
4. The person who is reading your app.

Edit: I see that @WedgeDawg alluded to 4. in a later post. However, aside from subjective activities, this can also tie into academic reinvention, which is pertinent to one's gpa. And to be fair, Wedgedawg did mention the difficulty of quantifying a disparate gpa in his initial discussion.

Edit part 2: @RogueUnicorn also mentioned "reviewer specific" factors.

Sorry, it's Friday morning

Anyway, I applaud your effort!

 

[Doug Underhill]

I think it overrated me, personally, it said my app should be 60% Tier 1 and 2, while Goro thought it should be more like 25%.

 

[DokterMom]

My suggestions:

• Add a feedback note for 'weakest link' section with accompanying recommendations. Ex. If community service is low, mission-driven schools are ill-advised. List schools known to love research, high-stats, community service, non-trads, reinvention, etc.
  
• For some/many categories, no or minimal involvement is a problem. Instead of scoring only 1 or 0, perhaps score it as a negative number, since that's how it's likely to play out.
  
• On the undergrad schools tier -- I'd suggest that you do need four tiers, but that maybe the multiplier drops from 4 to 3.5
  
  • HYP+MS
    
  • other ivy & equivalent (but not quite HYP-level)
    
  • neutral (most well-regarded LACs and flagship state universities)
  • not the best - including CC, non-flagship state colleges and universities, podunk
• I didn't do the math, but I'm curious what the total points possible balance is between 'Stats' and 'Other'
• And of course, the new MCAT score system will change your Stats table. Maybe the time to transition that is now? - using percentages rather than raw numbers.

 

[Doug Underhill]

Yeah, there's clearly a difference between Top 20 USNWR and the South Harmon Institute of Technology. My own USNWR school has a first-time acceptance rate of about 60% compared to the national average of 40%, but yet the formula treats it the same despite being 150% of the average.

 

[mehc012]

100k people take the MCAT a year (rising by ~6k a year). over .3% get a 41 or over: https://www.aamc.org/students/download/361080/data/combined13.pdf.pdf They could actually fill their class with a 42 median mcat (.1% get 42, so admit all 100 of those, admit the almost certainly >15 people that score above a 42, and then admit 115 people that get 41).

Anyway, the point is that despite Harvard being able to fill a class with superstar MCATers, they chose to admit people with MCATs that are 10x as "easy" to get (if you want to think about it like that), which illustrates how little stats matters to Harvard. Heck, don't they admit like 25 people with below a 34?

Either way, I think the current evidence points to =>75 LizzyM (balanced) as not mattering all that much, even though I wish it did.

Ah, yes...the old "AAMC tables are internally inconsistent" problem, sorry.
In general I take the percentiles over the percents given, as I think there are some rounding issues contributing to the percent column (0.051%+0.051% turning into 0.1% and 0.1%, for example) and the percentile column is less prone to that than adding up multiple rounded numbers.

However, I always forget about the discrepancy between the percent and percentile column in these tables - generally if something is 99.9th percentile, you cannot have 0.2% of the takers scoring at or above that level.

AAMC seems to do some variation of percentile where they include your score in the percentile...aka instead of reading it as "you scored better than 99.8% of people" it should be "99.8% of people scored as well or worse than you"/"you scored worse than 0.2% of people"

So, yeah...reading it that way I'd look at the numbers for a 39 (99.5 percentile) and say that 0.5% scored 40+, which turns into 475 people, which could fill 2 HMS's.

Still, do you think that all people with a 40+ apply to HMS, have a strong gpa, and also don't apply to any other school which they might rather attend? I'm actually curious here. I would not think that, but then, I went to an undergrad full of exactly the kind of people who avoided Harvard as an undergrad (one of my friends even matriculated there, then got into my school and went 'see ya!' after getting her Harvard student email and everything).

 

[treschic124]

I am categorized as a "B" applicant on this and I actually fared the best at the low yield schools of all the categories. My theory is that writing a semi-humorous/non-traditional personal statement that was authentic to my personality made me stand out in a good way at a school with 10k+ applications.

 

[packersfans]

I like this methodology alot, very cool idea. However, it did not work for me at all. It predicts I would have not gotten interviews or acceptances anywhere on par with where I ended up.

Either the academic stats are too heavily weighted or I was too hard on myself when assigning EC points to my various activities.

Edit: I didnt see it else where, but did you happen to use the below report in determining the weighting? I think page 12 would be helpful in calibration particularly on the non-academic factors.

https://www.aamc.org/students/download/267622/data/mcatstudentselectionguide.pdf

 

[resiroth]

Ah, yes...the old "AAMC tables are internally inconsistent" problem, sorry.
In general I take the percentiles over the percents given, as I think there are some rounding issues contributing to the percent column (0.051%+0.051% turning into 0.1% and 0.1%, for example) and the percentile column is less prone to that than adding up multiple rounded numbers.

However, I always forget about the discrepancy between the percent and percentile column in these tables - generally if something is 99.9th percentile, you cannot have 0.2% of the takers scoring at or above that level.

AAMC seems to do some variation of percentile where they include your score in the percentile...aka instead of reading it as "you scored better than 99.8% of people" it should be "99.8% of people scored as well or worse than you"/"you scored worse than 0.2% of people"

So, yeah...reading it that way I'd look at the numbers for a 39 (99.5 percentile) and say that 0.5% scored 40+, which turns into 475 people, which could fill 2 HMS's.

Still, do you think that all people with a 40+ apply to HMS, have a strong gpa, and also don't apply to any other school which they might rather attend? I'm actually curious here. I would not think that, but then, I went to an undergrad full of exactly the kind of people who avoided Harvard as an undergrad (one of my friends even matriculated there, then got into my school and went 'see ya!' after getting her Harvard student email and everything).

Harvard has a stupidly high yield rate (70%). The vast majority of people with a 39 or more have a strong GPA. From what I've heard from Harvard graduates they aren't huge fans of it, but appreciate the connections it offered.

I expect the medical school has a higher yield than undergrad, because Princeton and Stanford sway a lot of would-be Harvard students. Stanford as a medical school doesn't have the mass gen hospital which entices the type A gunner that usually scores so well.

If Harvard gave free tuition for everyone I think their yield would approach 85-90%. For whatever reason it is VERY hard mentally to turn down Harvard, even if there are options that, disregarding prestige/rep, you would enjoy and maybe perform better at.

 

[DokterMom]

I am categorized as a "B" applicant on this and I actually fared the best at the low yield schools of all the categories. My theory is that writing a semi-humorous/non-traditional personal statement that was authentic to my personality made me stand out in a good way at a school with 10k+ applications.

I suspect you're right about that. Can you just imagine reading 10,000 personal statements? Death to the cookie cutters!

 

[mmmcdowe]

Very interesting scoring system, though it makes me feel a little bit like Im in Brave New World.

 

[NeuResurgens]

Also, in the school tier system, what do you mean by "low yield?" And the asterisks refer to state schools preferring in-state applicants, correct?

 

[MaybeDr]

Low Yield generally means that for most applicants, the time, effort, and money invested in applying to these particularly schools could better be used elsewhere. In general, these are schools with a very high volume of applicants or a very low interview:applicant ratio.

Also, in the school tier system, what do you mean by "low yield?" And the asterisks refer to state schools preferring in-state applicants, correct?

There is the related comment.

 

[WedgeDawg]

Hi everyone!

I've made some changes to the system. I will make these changes in the OP soon and have uploaded a new version of the Excel document (v1.1) to this post.

• Added new MCAT scores to MCAT/GPA table (old scores are 38+, 37-36, 34-35, 32-33, 30-31, 28-29, below 28 which correspond to their respective new scores of 526+, 520-525, 512-519, 509-511, 506-508, 503-505, below 503 based on rough percentile)
• Changed wording of Level 5 Research from "presentation at major conference" to "solo presentation of their own, original work at a major conference" (keep in mind this is an 'and' as well as an 'or' - this by itself may not necessarily make an applicant a Level 5 Research)
• Clinical multiplier has been changed so that Level 3 is +9, Level 2 is +5, and Level 1 is -10
• Shadowing has been reduced to only two levels with Level 2 being +6 and Level 1 being -5
• Changed wording of Level 3 Volunteering to include "May also be working with marginalized or disadvantaged groups or in uncomfortable settings" (keep in mind that 'may' is used because doing this with a low commitment level or for a short period of time may preclude Level 3 status)
• Added military and Peace Corp service to Level 3 Misc
• Added MIT to Level 3 Undergrad and changed working of Level 3 Undergrad to "All other "prestigious" or highly selective schools including other Ivies, Caltech, Duke, etc" - because, as many of you have pointed out, regional perceptions of prestige outside of HYPSM are so varied, this is a 'catch-all' category for other schools; however, I would ask that this level be used judiciously
• Multiplier for Undergrad changed form x4 to x3
• Added GPA Trend category with two levels - Level 2 is Upward trend and adds +4, Level 1 is No upward trend and adds +0
• New total cap of 121 with thresholds remaining the same
• Tulane moved from Category 4 to Category 6

 

[neurotroph]

Wow my school list already lined up almost perfectly with my ARS score suggestions. Makes me feel better about it. I'm applying MD/PhD though.

 

[MrLogan13]

I'm still rated as an E-level applicant, but I got a whole extra 2 points. lol

 

[Cyberdyne 101]

I'm still rated as an E-level applicant, but I got a whole extra 2 points. lol

I don't think his metrics are appropriate for applicants like you.