How did this guy get rejected from every school?

[PhilippMainlander]

Undergrad Institution: Local American State University
Major(s): BS Mathematics, BA Secondary Education
Minor(s): None
GPA: 3.833 overall, 3.91 major

Graduate Institution: Local American State University
Major(s): MS Pure and Applied Mathematics
GPA: 3.975 overall, 4.0 major
Type of Student: Domestic Hispanic Male

GRE Revised General Test:
Q:160 (74%)
V:151 (52%)
W:5 (92%)
GRE Subject Test in Mathematics:
M:640 (46%)

Program Applying: Pure math programs, interested in algebra and combinatorics.

Research Experience: Proved tight bounds and limiting behaviors on certain classes of iterative processes acting on integer labeled vertices in simple directed graphs using results about the convergence of Markov Chains.
Awards/Honors/Recognitions: Department Award for Excellence and Service in Mathematics among GPA based scholarships.
Pertinent Activities or Jobs: Tutoring Center Head at undergraduate university, graduate assistant at graduate university, currently an adjunct professor at both universities.

Applying to Where:

Reach
Ohio State University Rejected 3/06
Rutgers State University, New Brunswick Was told they would not be making an offer via email 2/28
University of California, San DiegoRejected 3/07

Fit
Rutgers State University, Newark
Carnegie Mellon UniversityRejected 4/9
Northeastern University
University of Connecticut Waitlisted 3/8
Virginia TechRejected 4/24

Safety
Stevens Institute of Technology Rejected 2/20

He seems to have a decent profile. I’m confused. People who were less stronger than him got into into some schools.

[mishania1996]

1) GPA is pretty much inflated in the non-top schools.
2) His GRE Q is low, however getting 90% should be super easy for for pure mathematicians because GRE General is taken by almost all majors applying to grad schools in US.
3) We don't see what level of classes he has in his transcript.
4) LoR's are the most important in an application( it matters not just what is written there but also by whom). It is not reflected in his post at all.

[PhilippMainlander]

1) GPA is pretty much inflated in the non-top schools.
2) His GRE Q is low, however getting 90% should be super easy for for pure mathematicians because GRE General is taken by almost all majors applying to grad schools in US.
3) We don't see what level of classes he has in his transcript.
4) LoR's are the most important in an application( it matters not just what is written there but also by whom). It is not reflected in his post at all.

Point 4 kind of worries me. My school is ranked 70 or so on the US News ranking. I don't know how strong the credentials of my letters writers are. Is there any way to tell? They are all full tenured professors.

[mishania1996]

You can check how much they are active in research these days (years actually), who they are collaborating with, which journals they do get published, how much they travel for seminars/workshops, if they are organizing/participating at any.

[dujiahan]

I always wonder: how did someone prove a result on the convergence of something, but only scored 70+ percentile on the GRE quantitative(the General GRE, not GRE math subject)? This makes no sense, because GRE quantitative is a joke, unless he didn't write his proof but let his professors write the proof, and he just put his name down.

Unless he just didn't care about the GRE, in which case it makes no sense either. GRE quantitative is so easy, that a math PhD student will easily score a high score even if he doesn't care about it. For example, you don't need to care to count, right?

[Integreat]

...
He seems to have a decent profile. I’m confused. People who were less stronger than him got into into some schools.

To add onto mishania1996's response, we also don't know how good their writing is. Some amazing mathematicians can be horrible writers, and I feel that a large part of the application depends on selling yourself (i.e., how well you can write)

Point 4 kind of worries me. My school is ranked 70 or so on the US News ranking. I don't know how strong the credentials of my letters writers are. Is there any way to tell? They are all full tenured professors.

This matters 100%. A lot of professors are going to say that the student is stellar, best in the class, etc. So now you need to compare candidates' letter writers. One from a top geometer, saying that so and so is a promising mathematician matters way more than a nobody from a nobody university saying the same thing. It's hard to escape from. As harsh as it is, that's reality -- personal experience from someone in that position.
Some guides: What is their position (assistant/associate/distinguished)? Who do they collaborate with? How many citations/publications have they had recently? You can also ask your advisor/someone in the department for their opinion on how well established your letter writers are.

I always wonder: how did someone prove a result on the convergence of something, but only scored 70+ percentile on the GRE quantitative(the General GRE, not GRE math subject)? This makes no sense, because GRE quantitative is a joke, unless he didn't write his proof but let his professors write the proof, and he just put his name down.

Unless he just didn't care about the GRE, in which case it makes no sense either. GRE quantitative is so easy, that a math PhD student will easily score a high score even if he doesn't care about it. For example, you don't need to care to count, right?

There's a significant difference between doing math in a timed, stressful environment versus doing math without a time limit. Even though the GRE Q isn't too difficult (conceptually), it's still a test, and not everyone is a good test taker. And (disregarding this particular applicant's knowledge base), not everyone is good at stats/combinatorics, which I feel was basically all of the GRE Q. I haven't seen a box and whisker plot in six years! Of course I didn't study for it, and that's my problem. But I don't think it's fair to say that every mathematician, barring preparation, should be scoring 90th percentile plus on a test whose material may be largely irrelevant to their field of study.

With that said, I imagine their advisor helped along the way. Also, things are often exaggerated/inflated so they seem better. It's all about how you present yourself.

[dujiahan]

...
He seems to have a decent profile. I’m confused. People who were less stronger than him got into into some schools.

To add onto mishania1996's response, we also don't know how good their writing is. Some amazing mathematicians can be horrible writers, and I feel that a large part of the application depends on selling yourself (i.e., how well you can write)

Point 4 kind of worries me. My school is ranked 70 or so on the US News ranking. I don't know how strong the credentials of my letters writers are. Is there any way to tell? They are all full tenured professors.

This matters 100%. A lot of professors are going to say that the student is stellar, best in the class, etc. So now you need to compare candidates' letter writers. One from a top geometer, saying that so and so is a promising mathematician matters way more than a nobody from a nobody university saying the same thing. It's hard to escape from. As harsh as it is, that's reality -- personal experience from someone in that position.
Some guides: What is their position (assistant/associate/distinguished)? Who do they collaborate with? How many citations/publications have they had recently? You can also ask your advisor/someone in the department for their opinion on how well established your letter writers are.

I always wonder: how did someone prove a result on the convergence of something, but only scored 70+ percentile on the GRE quantitative(the General GRE, not GRE math subject)? This makes no sense, because GRE quantitative is a joke, unless he didn't write his proof but let his professors write the proof, and he just put his name down.

Unless he just didn't care about the GRE, in which case it makes no sense either. GRE quantitative is so easy, that a math PhD student will easily score a high score even if he doesn't care about it. For example, you don't need to care to count, right?

There's a significant difference between doing math in a timed, stressful environment versus doing math without a time limit. Even though the GRE Q isn't too difficult (conceptually), it's still a test, and not everyone is a good test taker. And (disregarding this particular applicant's knowledge base), not everyone is good at stats/combinatorics, which I feel was basically all of the GRE Q. I haven't seen a box and whisker plot in six years! Of course I didn't study for it, and that's my problem. But I don't think it's fair to say that every mathematician, barring preparation, should be scoring 90th percentile plus on a test whose material may be largely irrelevant to their field of study.

With that said, I imagine their advisor helped along the way. Also, things are often exaggerated/inflated so they seem better. It's all about how you present yourself.

Still, the GRE quantitative is just basic high school stuff, there shouldn't be any stress at all when a PhD applicant takes it, since people major in English also take the same test.

[joriandres]

he wrote: "graduate assistant at graduate university, currently an adjunct professor at both universities.". I have the theory almost proven (I have seen most of this type of profiles rejected), that if you are already in a graduate program, most universities will see this disrepectful or something like that.

[asterac]

“Local state university” says to me that his programs are probably very unknown/low ranked (AMS Group III or maybe M) and chosen out of convenience rather than program quality—he says he’s now an adjunct at both, so they’re in the same area.

I have heard from multiple people, including the Director of Graduate Studies at my school (AMS Group I Public), that GRE scores become way more heavily weighted in evaluating candidates from unknown schools. (This is one reason why higher GRE scores are often emphasized for international applicants.)

If I’m right, rank 70 isn’t really the same situation (tho beware overall of taking USNews too seriously).

This is just a guess, because his description doesn’t give much detail, but his research sounds like it could be very basic for masters level work. Using Markov chains to prove stuff about integer-indexed digraphs could be a fancy way to say something as simple as, “I did a basic reading on graph theory, took some finite digraphs, and used basic linear algebra to do a proof about some algorithm no one’s really curious about.”

Another thing is that his list of schools is sort of odd. For someone who is from an unknown school and doesn’t have great test scores, I’d consider Northeastern and CMU more like reaches. While some of the others are lower ranked, he still hasn’t applied to that many actual matches or safeties.

Further, even if his own institutions are similar in rank to those matches, there are so many unknown factors. Here is my wildest guess yet, really more an example of something that could have gone wrong: he might not have done a great job at selling himself as someone who’s sure he wants to be a research mathematician. The fact that he double majored in math *and* math education, then did his masters in pure *and* applied math (unusual) hint at that a little to me. An “and” like this early in a career could mean overflowing excitement about multiple, related things, or it could mean he lacks mathematical maturity or seems to have lukewarm/unfocused feelings about the field. Or maybe he comes across as someone whose actual passion is teaching and who is only pursuing a PhD because his degrees so far haven’t netted him a stable enough teaching job.

I feel like this rarely gets said on this kind of site, so... None of these things even necessarily say something about his worth as a possible future researcher. The thing to remember is that most admissions committees will knowingly be handing out rejections to possible “false negatives” and even just “not-as-clear-positives.” If he has the capability to succeed but just didn’t make a compelling enough case that he’d do *better* than other applicants that year, that’s enough for a rejection.

[asterac]

he wrote: "graduate assistant at graduate university, currently an adjunct professor at both universities.". I have the theory almost proven (I have seen most of this type of profiles rejected), that if you are already in a graduate program, most universities will see this disrepectful or something like that.

He *was* a graduate assistant in his masters program. Doing a masters doesn’t put you at a disadvantage at all—it’s not really a terminal degree in math, particularly in the US. It’s more like a glorified post-bacc. Also, transfers in grad school are a thing that happens. Someone from my school transferred somewhere in Australia. A professor of mine transferred from Brandeis to Columbia after the first year of his PhD.

I guess it could come off disrespectful if you badmouthed your current institution.

[lambert]

Double majoring doesn't mean anything, many universities offer a double major and doing that shouldn't imply anything about your level of commitment in mathematics. There's a bunch of profiles here that majored in math and did a minor on something unrelated, like music. Most likely grad committees won't give any attention to your minor.

Likewise "MS in pure & applied mathematics" might just be the name of the program, I don't believe people do simultaneous masters in both pure and applied mathematics. When you do stuff like combinatorics, the line there is hazy anyway.

This is all speculation, many people fixate on test scores and grades but these matter the least, and the most important factor (letters of recommendation) tends to be unknown to the applicants. Everyone will tell you they have stellar letters of rec, and though that may sometimes be true, you'd be surprised at how often people overestimate the quality of their LORs.

[ponchan]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

[lambert]

Not really, no professor's going to say "Oh wow this one's been called thoroughly average in his letters but he got a 90% on the mGRE, let's hook this guy up!"

Research isn't that important for grad admissions. Professors know most undergrad research tends to be shallow and kinda trivial, it's up to the rec letters to vouch for whatever research you've done, and show to the committee you hold promise for future research potential. Chances are if you have any papers they won't be read thoroughly cause they can only devote so much time in one applicant.

Obviously if you cruised by undergrad without doing anything other than getting semi-good grades in classes, your LORs won't save you, particularly because they won't be that good by definition. "So-and-so got an A in my real analysis class and they're a good kid. - Prof" isn't going to impress graduate admissions.

[asterac]

Double majoring doesn't mean anything, many universities offer a double major and doing that shouldn't imply anything about your level of commitment in mathematics. There's a bunch of profiles here that majored in math and did a minor on something unrelated, like music. Most likely grad committees won't give any attention to your minor.

Likewise "MS in pure & applied mathematics" might just be the name of the program, I don't believe people do simultaneous masters in both pure and applied mathematics. When you do stuff like combinatorics, the line there is hazy anyway.

Totally agree—the bare facts don’t really tell you how committed he is, but that’s kinda my point. The OP asked, “What’s wrong with this guy? His profile seems fine,” so I suggested some things that might fit, as examples of what else the admissions committees may have seen or thought, maybe in his statement of purpose. I may not have made that clear.

[asterac]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

Yeah, even if they say they’re the strongest student ever in 50 years at x school, the admissions committee still doesn’t really know what that means if the professor and x school are unknown. <- straight from my school’s DGS

Also, a letter that’s nothing but superlatives would be pretty weak, yeah. There’s this thing someone wrote about what makes letters good and bad... I guess it’s kind of a trope for well-intended letters to say something like, “This person’s the next Gauss!” and that’s pretty useless. But a good letter doesn’t have to be effusive. A good letter is one that answers the question, “Has this candidate demonstrated the ability to succeed in our program?” Specifically, another thing our DGS said is, “The best thing professors can do is make a favorable comparison to grad students at their own school. If their school is comparable to yours, that’s the best possible indicator you can get for success in your program.”

I guess this is all a little off topic tho

[ponchan]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

Yeah, even if they say they’re the strongest student ever in 50 years at x school, the admissions committee still doesn’t really know what that means if the professor and x school are unknown. <- straight from my school’s DGS

Also, a letter that’s nothing but superlatives would be pretty weak, yeah. There’s this thing someone wrote about what makes letters good and bad... I guess it’s kind of a trope for well-intended letters to say something like, “This person’s the next Gauss!” and that’s pretty useless. But a good letter doesn’t have to be effusive. A good letter is one that answers the question, “Has this candidate demonstrated the ability to succeed in our program?” Specifically, another thing our DGS said is, “The best thing professors can do is make a favorable comparison to grad students at their own school. If their school is comparable to yours, that’s the best possible indicator you can get for success in your program.”

I guess this is all a little off topic tho

Good point. I think if a professor compares a student favorably to a previous student who now is someone currently well respected in the math community, that would probably make a solid letter.

[Cyclicduck]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

Yeah, even if they say they’re the strongest student ever in 50 years at x school, the admissions committee still doesn’t really know what that means if the professor and x school are unknown. <- straight from my school’s DGS

Also, a letter that’s nothing but superlatives would be pretty weak, yeah. There’s this thing someone wrote about what makes letters good and bad... I guess it’s kind of a trope for well-intended letters to say something like, “This person’s the next Gauss!” and that’s pretty useless. But a good letter doesn’t have to be effusive. A good letter is one that answers the question, “Has this candidate demonstrated the ability to succeed in our program?” Specifically, another thing our DGS said is, “The best thing professors can do is make a favorable comparison to grad students at their own school. If their school is comparable to yours, that’s the best possible indicator you can get for success in your program.”

I still believe lor are the most important factor, at least at the top. There is no such thing as having "everything else right"; realistically even top students' research will be shallow and unimportant. A strong letter of rec doesn't consist of "vapid superlatives". A strong letter of rec doesn't come from an unknown professor at an unknown university. A strong letter of rec comes from well-respected professor who has seen many top students in his/her years and knows the student well enough to confidently judge them to be worthy of a top institution.

[asterac]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

Yeah, even if they say they’re the strongest student ever in 50 years at x school, the admissions committee still doesn’t really know what that means if the professor and x school are unknown. <- straight from my school’s DGS

Also, a letter that’s nothing but superlatives would be pretty weak, yeah. There’s this thing someone wrote about what makes letters good and bad... I guess it’s kind of a trope for well-intended letters to say something like, “This person’s the next Gauss!” and that’s pretty useless. But a good letter doesn’t have to be effusive. A good letter is one that answers the question, “Has this candidate demonstrated the ability to succeed in our program?” Specifically, another thing our DGS said is, “The best thing professors can do is make a favorable comparison to grad students at their own school. If their school is comparable to yours, that’s the best possible indicator you can get for success in your program.”

I still believe lor are the most important factor, at least at the top. There is no such thing as having "everything else right"; realistically even top students' research will be shallow and unimportant. A strong letter of rec doesn't consist of "vapid superlatives". A strong letter of rec doesn't come from an unknown professor at an unknown university. A strong letter of rec comes from well-respected professor who has seen many top students in his/her years and knows the student well enough to confidently judge them to be worthy of a top institution.

For sure

[ponchan]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

Yeah, even if they say they’re the strongest student ever in 50 years at x school, the admissions committee still doesn’t really know what that means if the professor and x school are unknown. <- straight from my school’s DGS

Also, a letter that’s nothing but superlatives would be pretty weak, yeah. There’s this thing someone wrote about what makes letters good and bad... I guess it’s kind of a trope for well-intended letters to say something like, “This person’s the next Gauss!” and that’s pretty useless. But a good letter doesn’t have to be effusive. A good letter is one that answers the question, “Has this candidate demonstrated the ability to succeed in our program?” Specifically, another thing our DGS said is, “The best thing professors can do is make a favorable comparison to grad students at their own school. If their school is comparable to yours, that’s the best possible indicator you can get for success in your program.”

I still believe lor are the most important factor, at least at the top. There is no such thing as having "everything else right"; realistically even top students' research will be shallow and unimportant. A strong letter of rec doesn't consist of "vapid superlatives". A strong letter of rec doesn't come from an unknown professor at an unknown university. A strong letter of rec comes from well-respected professor who has seen many top students in his/her years and knows the student well enough to confidently judge them to be worthy of a top institution.

"realistically even top students' research will be shallow and unimportant." Speak for yourself. Every year there are undergraduates who publish in papers at the level of Transactions of the AMS or higher. I wouldn't call that unimportant. However, I agree that a strong letter must come from a well-respected professor at a good university.

[Cyclicduck]

"realistically even top students' research will be shallow and unimportant." Speak for yourself. Every year there are undergraduates who publish in papers at the level of Transactions of the AMS or higher. I wouldn't call that unimportant.

Speak for yourself? I've participated in multiple REUs that consistently get undergraduate papers into journals of such caliber, and I can guarantee you that it is indeed, unimportant. If you think that most such papers can be reasonably called important, then you are seriously deluded. This is simply how the vast majority of top undergraduate research goes. Students are put together with mentors who know a field well enough to suggest a problem that is reasonable from an undergrad to do, they do it, and it gets published. Sometimes the amount of original work put in is nil, and you see the same result. Occasionally the very best students may produce things which are very impressive and point toward a promising future, but even then the work itself isn't important (I'm not talking about one-in-a-generation exceptions like Drinfeld or Scholze).

[quinquenion]

It's also speculation to assume letters of recommendations are the "most important factor" in evaluating an applicant. If you don't have everything else right -- particularly, strong research -- then what can a letter of rec do beyond fill up a page with vapid superlatives? If a letter says Alice was the strongest student Professor X had in all his 50 years of teaching, then that seems meaningful; unless, of course, there's no solid research or concrete evidence to back that statement up. And if there's solid evidence to back that up, it would speak for itself. A lot of students assume getting a couple A's from a professor with whom they have a cordial relationship means they will get a "glowing" recommendation.

I am on a graduate admissions committee of a well-regarded university. From what I've seen, letters of recommendation are the most important factor in evaluating an applicant. Papers, grades, scores, etc. are all important, but when evaluating a candidate, I read the rec letters and who they're from first.