What constitutes reaches, matches, and safeties

Not necessarily true, since (for example) an open admission community college or a stats-only admission college may not bother to mention that.

No… Maximum GPA is 4.1. Valedictorian is unweighted… exact 4.0. The 60 who has 4.0 or higher are actually 3.95 (the school rounds off) or higher.
Extremely little weighting. There is 0 weighting of Freshman and Sophomore classes. Junior year, you can have 1.04x weighting of AP classes. So if you max out on AP classes your Junior year, then you get end up with an overall GPA of up to 4.1.

So basically, about 10 students per year get a perfect unweighted 4.00… Another 50 end up with a weighted GPA of between 3.95 and 4.1, which is the equivalent of unweighted 3.92-4.00. (depending how he finishes this year, my son is unweighted at 3.88 and weighted at 3.91… so 3.9 whether weighted or unweighted).

“high performing schools” – Not all 24,000 high schools in the US. That’s the point – wide variety.

But I’m not talking about them. 1500 and “solid” rigor and “high GPA” is not enough to make a T10 school into a match.

I’m disagreeing with the notion that “nobody is a match.”

1500 and “high” GPA and “solid” rigor is only enough to leave someone as a reach.

Yes— The odds are slim as to MOST of their applicants.
Most of their applicants – the odds are around 0%

The question is, do they get ANY applicants where the odds actually are closer to the 50% range?
And my answer is yes – A student who is valedictorian, 1600, 10 AP exams, URM, 1st gen, award winning scientist – Such student’s chances are actually much much higher than 50%.

I’ll answer that – The odds are near 1% or less for the T5. Those schools don’t take “typical.”
For T10-T20… that’s where you get those 10-20% odds.

1 Like

I was referring to private colleges. Nearly all public schools are need blind, at least to their in-state residents. Only a few of them meet full need for OOS students.

1 Like

Of course, both “need blind” and “meet need” do mean less than what most people seem to think they mean.

  • “Need blind” may be narrowly defined as the admission readers having no access to the applicant’s financial aid application or any “need” calculations from it. However, many colleges do consider correlates of need (legacy and first generation are two obvious examples in opposite directions), so that admissions weighting of such correlates of need can be used to target the class’ overall “need” without using any financial aid information for individual applicants (even though some applicants and admits will be outliers in the correlation, such as poor family legacies and wealthy family first generation).
  • “Meet need” may be very flexibly defined, since the college itself defines “need”. This can result in a college that claims to “meet need” offering worse need-based financial aid than a college that makes no such claim, since the “meet need” college may have a very stingy definition of “need”.

Yes, meeting “full need” has different meanings at different colleges. I’d even say that some schools are more “need blind” than others.

1 Like

Adding a little levity, or at least my personal amusement with emojis…

:pig2::pig: :pig_nose:Animal Farm :pig_nose: :pig: :pig2: for the high school reading list! Actually, maybe that was middle school…

As for the weighting system at your school you’ve changed from 60 students at “4.0 or better” to 50 at a weighted 3.95 or better, but given that colleges look at rigor for themselves, and that half the kids from the top 20% gpa were admitted to "Ivy or equivalent,” that’s still a pretty good indication that the rigor was there for the kids with top grades. Likewise for the 100% admission rate for the 9 with an unweighted (and unrounded) 4.0.

There is variety, but applicants are viewed in the context of what their schools have to offer, and at at almost all schools (even those that don’t weight for APs) kids who do well generally take the more difficult courses. Of course there are exceptions, but not to the extent you have suggested.

Above you wrote that for students from your high school applying ED to Cornell,

"3.9/1470 mostly gets you in. (a few rejections above them, but not many. And several acceptances below that, but acceptance rate of over 50% with 3.9 and 1470).”

So for ED at Cornell, at least, you are talking about an acceptance rate of over 50% for kids with a 3.9/1470. Were that accurate, it would make Cornell a match. But this is a good example of what I mean when I mention the overestimation of the odds of admission for unhooked but qualified kids.

When you look at the Naviance stats for kids admitted to Cornell, those stats include legacy and other boosted admissions, and it seems a pretty safe assumption that your school (NY, Cornell is very popular, over 10% of kids go to Ivy or equivalent) has contingent of Cornell/Ivy equivalent legacies. If so, then your conclusions about what it takes to get into Cornell (“ED – 3.9/1470 mostly gets you in”) are likely to be way off.

For example, a high performing Ca high school used to provide statistics which could be culled by legacy and other hooks (ALDC, etc):

  • Cornell accepted 38 students in the most recent 3 year period where the stats were available, but 21 of these 38 students were legacy or otherwise hooked. The odds of admission for hooked students were 4 times higher than that of unhooked students, and a significant number of hooked students were being accepted with much lower grades (and, presumably, test scores, since test scores correlate with grades.)
  • During the same three year period, Harvard accepted 28 students, but 20 of the 28 were legacy or otherwise hooked. The odds of admission were over 5 times higher for hooked students. This matches the findings from the Harvard lawsuit where stats show that legacies received a 5x admissions boost.

If that is the question, then the question and answer are of little use to most of those students/families seeking help here. Telling some family of a high achieving high school student that they’d have a better chance if they were being recruited? Interesting, perhaps, but unhelpful.

But telling an unhooked kid with a 3.9 and a 1470 that they have a better than 50% chance of getting into Cornell if they apply ED? That kid is being mislead into believing their chances are much better than they really are.


It’s easy to overlook how big a factor where you went to HS plays into your admission chances. Are you attending a private schools that sends 30% plus of their students to Ivy+ schools? Or does your local school send 2-3 a year? or 1 in the past decade!!


The issue when we look at results in retrospect is that there are high GPA/Test/Rigor students who individually had better than 50% chance of getting into a T10/T20, and we see that in hindsight based on their multiple acceptances. However, no one can predict those probabilities in advance based on those data points alone. If we go back to data uncovered from the Harvard litigation, we see the following admissions rates for each decile of Academic Index ratings (AI’s are determined by a formula based on GPA and test scores) from Table 5.2 of the Arcidiacano Report:

AI Decile Admit Rate
1 0.01%
2 0.39%
3 1.45%
4 2.83%
5 3.91%
6 4.79%
7 5.62%
8 6.85%
9 8.77%
10 11.70%

I recall when Stanford and Brown also published admit rates by GPA and test score ranges, and they similarly were in the 8-12% range for the top tier if I remember correctly.

Perhaps if we know other factors (ALDC, URM/SES status, winner/high placement in major national/international contests) we can bump up our probability assessment, but as @ucbalumnus pointed out above, what we cannot see and IMO puts a candidate with just great academic stats into a multiple cross admit territory are the essays and LoR’s. We can also get a “list” of EC’s, but what will matter is the quality of the participation which often is backed up in the LoR’s/essays.


I think one of the fundamental problems with categorizing colleges as “reaches”, “matches”, and “safeties” is that we don’t really know the admit probabilities of any given applicant. We only know the admit rates of the applicant pool (or in some cases, a slice of that pool that best matches the applicant’s basic stats). The two aren’t the same, and they can be very different. We simply don’t have enough information to make the determination for most applicants (except for a few applicants who are either the most impressive or who are the least qualified).


This isn’t a conspiracy. 60 students got 4.0 or better – Correct. But if you go down to the next decimal point, it’s 60 with 3.95 or better.
And no, it’s not 50… it’s 9 valedictorians with 3 years of perfect 4.0 (never a single 3.9), AND another 50 with 3.95 (rounded to 4.0) or better. Thus – in total, 60.

It’s proof of the opposite! It means HALF those kids had great rigor – And THAT HALF got into Ivy or Equivalent.
And the other half – that didn’t get into Ivy or equivalent, didn’t have such great rigor.

So yes… half the students with top grades had great rigor. For half, it was more mediocre.
For the students that got into Yale and MIT… they certainly had the top rigor and the top grades.

As I said… this isn’t without rhyme or reason.

Not really… While schools claims they view each student in the context only of their own school, it’s well known that some schools are “feeder” and others aren’t. Schools that don’t offer any rigor generally aren’t sending their students to Harvard.

No… Please stop misquoting. I said that’s the acceptance rate out of MY HIGH SCHOOL FOR EARLY DECISION.

But yes… Cornell is NOT Harvard. (Just ask anyone who went to Ivy). 3.9/1470 is not a match for Harvard – It would be an extreme reach for Harvard. And unweighted 3.9/1470, with a fantastic essay, with fantastic LORs, with fantastic ECs, and with tons of rigors – Yes, that is a match for ED at

Cornell. (with some variation based on which college at Cornell, which program… Some of the programs have overall acceptance rates of nearly 20%)

Nice twisting of my words. But no, telling a kid that has a fantastic resume, that has unweighted 3.9 and a 1470 on the SATs, and have 10 AP exams, fantastic extracurriculars, etc… Telling them that they might have around 50/50 chance at Cornell if they put together a great application … That would be a pretty fair assessment.

1 Like

Exactly. Important to note that not all HS’s are created equal in the eyes of admissions officers.

1 Like

That’s what we try to do when we chance – determine the probability of getting into a school – and it’s difficult to do with any confidence because there are opaque (as ucbalumnus so accurately put it…) variables in play – like LORs and the needs and priorities of the school. Because of that, we lack perfect knowledge, which renders chancing imperfect. We just can’t know everything. I suppose in many ways this reflects the human condition.

But, I suppose an educated guess based on the factors we do know is better than nothing.

1 Like

That would be 100% admission for the kids in the top 20% with rigor, and 0% for those in the top 20% without rigor. Who knew that admissions to an Ivy was so clear cut?

“Feeders” is a misnomer, but some high achieving, rigorous high schools consistently send a relatively high number of kids to Ivy schools. That usually means two things:

  • The school is rich in hooks like legacy, athletes, diversity, deans list, etc., and
  • The school has an overabundance of kids who are extremely well qualified by measures like SAT, grades, ECs, essays, recs, etc.

Taken together, these two factors often mean that AO’s will mostly admit the hooked kids (many of whom are also highly qualified academically), and will perhaps pick and choose a few truly extraordinary kids from the surplus of kids with excellent grades, test scores, ECs, essays, etc.

In my observation, unhooked kids at such schools who have 1470 would be strongly advised to submit test optional and to add some actual matches, because that score will pale in comparison to a surplus of classmates applying to the Ivy’s with much higher scores.

Here again is what you wrote:

Your statement was based on examining the Naviance scores from your school, without the factors you are now adding in. More importantly, unless the ECs rise to the level of truly extraordinary, this tranche of students (unhooked, high performing high school, rigor, 3.9 gpa, 1470 SAT, excellent ECs) . . . face long odds at Ivy League schools, not better than 50%.

Added: this is going nowhere, so I am bowing out.

My advice to any parents or kids from high performing high schools: Be wary of acceptance rates from Naviance and/or schools’s matriculation numbers. When comes to top schools, these are often distorted by legacies, athletes, etc.

1 Like

For applicants who are either exceptional or obviously unqualified, we may be able to estimate their chances, but I’m not sure how we’re going to estimate their probabilities of admission for the vast majority of applicants in the middle. Is the probability 10%, 20%, or even 40% for a particular school? We don’t know enough not only about the applicant in question, but also the relative strength of the applicant pool to that particular school in that particular year.

1 Like

It’s a spectrum. In total, of the top 20% of the class… the top 60, 30 are going to Ivy or Ivy equivalent. I’m not saying 30 kids took 10 APs each and 30 took 0. And certainly, some of those top 60 weren’t even interested in Ivy schools.
But yes – Absolutely, overall, the ones who got into their preferred Ivy/equivalent had more rigor than those that did not.
Schools consider rigor to be MORE important than GPA.

Yes, it is fairly clear cut which kids are likely to get into an Ivy and which aren’t. No, it doesn’t mean that they get into every Ivy they apply to. But yes, I can look at a kid with 4.0, great ECs, 1550 SATs, 10 APs, other factors… and say, “if you apply to every T20 school, you are very very likely to get into 1 or more.”

Since we were talking Cornell:
The Cornell Director of Admissions told our school that failure to submit SATs is essentially a negative – they will assume you aren’t submitting because your school is below their range.
Per their most recent CDS, Cornell’s mid-SAT range is 1410-1530. In other words, 1470 is ABOVE the median for Cornell. It would be a score absolutely totally to submit.

And virtually every school has a much lower acceptance rate for test optional submissions.
Here is the reality – You can get away with going test optional if you have great rigor and GPA AND have a significant hook. If you don’t have a significant hook, if you don’t have a SAT in range, then your chances of acceptance are very very low. If your SAT is ABOVE the median for the school, then definitely submit it.

1 Like


The Cornell director of admissions is a major and vociferous supporter of test-optional, and admissions to three of its undergraduate colleges are TEST BLIND.

“Jonathan Burdick, Cornell’s vice provost for enrollment, said fears of grade inflation are overblown. His university is running a two-pronged experiment. Scores are optional for entry to schools of arts and sciences; engineering; human ecology; and industrial and labor relations. They aren’t considered at all for schools of agriculture and life sciences; architecture, art and planning; and business.

“Results so far are encouraging, Burdick said. Among his admission team, he said: “I have yet to find one who says even for a second that they miss having the test. That’s not a surprise to me. Well-trained readers are able to discern a lot about students’ academic readiness without the extra validation of the test.” His leanings are clear. “We’re better off paying more attention to the transcript,” he said, “and being less impressed with the test score.”


You are also looking at the class of 2024 test scores, not 2025. The range for 2025 was 700–760 RW and 750–800 Math; 33–35 ACT. 40.8% of students submitted SAT scores and 19.7% ACT, so at least 39% of accepted students did not submit scores. 22% of the accepted and 24% of the enrolling students were in the test-blind colleges, meaning that many applicants were accepted into the other 4 colleges without scores.


The Director of Admissions is Shawn Felton, not Jonathon Burdick. I am citing what he told our high school at a presentation just a few months ago. I have no doubt they love test optional because it increases the number of applications, and makes it easier for them to accept hooked test-optional candidates, where a poor standardized score would have otherwise been a demerit. But where all else is equal, they are more likely to admit a student who submitted a score within their general admission range.
But it is very clear – Acceptance rates without test submission are lower than with submission.

I’m looking at the most recent CDS, which is the most direct evidence. I am not doubting the accuracy of the scores you cited – Which would still suggest submit. A 1470 still places an applicant in the median 50th percentile.
And when AOs are reading your application, if you don’t submit, then they will assume you fell below the 50th percentile. (And this was implicit in what the Cornell Admissions Director told us)

You can easily find stats like:

  • University of Virginia – The overall admit rate was 20.57%, but 72% of the 9,875 students who gained admission submitted test scores.
  • University of Georgia (Early Applications) – There were 20,900 applications with an overall admit rate of 39%, but when you break it down, students who submitted scores saw an admit rate of 49%, whereas non-submitters were admitted at only 26%.
  • Georgia Tech – 37% of the applicants did not submit scores, but of those who were admitted, only 21% had not submitted scores.
  • Emory – About 50% of the 33,780 applications included test submissions. Of the 6,892 students who were admitted, however, 69% did submit test scores, giving submitters a 55% advantage.
  • Tufts – Early admissions was reported as an equal split between submitters and non-submitters, but when all of admissions are considered, submitters had a 33.3% advantage.
  • Vanderbilt: 56.3% of applicants submitted test scores, but 61.1% of the admission offers went to students who submitted scores, giving submitters a 36.33% advantage.
  • Boston College Of the 7,536 admitted students, 39% did not submit scores and those who did submit had an 85% advantage in admissions.

The evidence is clear — Your chances of admission go DOWN if you do not submit a score. Of course, submitting a bad score isn’t going to help you. And if you have a hook, maybe you can get in without submitting a score. For unhooked applicants, they are still expecting to see a good score that adds to your overall picture.


Correlation is not causation though.

Not submitting may correlate with a lower admit rate, but may not be the cause of that lower admit rate. It may be that, overall, the nonsubmitters had weaker applications.


What I’ve been told by admissions officers and advisors— If you do not submit a score, they have one less piece of information to determine whether you’re a match.

And generally, they will assume you didn’t submit the score because it wasn’t in their range.

Nobody is going to get rejected because they submitted a score in the median range for the school. May not be a plus, but it avoids a minus.

When an applicant doesn’t have a score in the range, it is best not to submit. But then, everything else must be better to earn admission.
This is great for someone with a hook but a weak score. For example, URM with a subpar SAT score. Schools can now admit that person without fear of reducing their average SAT score.

But if you have 2 applicants who are otherwise totally identical, but one submits a score within median range, the other goes test optional — the first applicant has one extra reason to admit, the first applicant has a better chance of admission.