How to Build a Balanced College List

A balanced college list is one where every school passes three tests at once: you would genuinely be glad to attend it, your family can pay for it, and — taken together — the list contains at least one school that is very likely to admit you. Sort the schools by how likely admission is, and you get the familiar three buckets: reach, target, and likely. The sorting is only useful if you do it against each school's own published numbers rather than against a feeling about how good the school is.

The two mistakes that break a list are almost never "too few schools." They are sorting by prestige instead of by admit rate, and treating affordability as something to figure out after the acceptances arrive. A list of nine schools where the cheapest one costs more than your family can pay is not a list — it is nine ways to end up with no option in April. Build the money constraint in from the start, and build the likely school in as a real preference, not a placeholder.

What do "reach," "target," and "likely" actually mean?

They are statements about probability, and the only honest way to assign them is to compare your own record against the record of the students a school actually admits.

• Likely — your grades and, where relevant, your test scores sit at or above the upper end of the school's admitted range, and the school admits a large share of the people who apply. You would be surprised to be turned down.
• Target — you sit comfortably inside the middle of the school's range, and the admit rate is high enough that a typical qualified applicant has a real chance. You would not be surprised either way.
• Reach — you sit below the middle of the range, or the school is selective enough that being inside the range is not sufficient. Both routes lead to the same bucket.

That second route is where most lists go wrong, and it deserves its own section below.

One clarification, because the labels get used loosely: "likely" is about probability, not about how much you want to go. It is not a lesser school; it is the school most likely to say yes. If you would be unhappy to attend the likely school on your list, the list is not finished — you have not found the right one yet. This is the single most common way an otherwise sensible list fails, because the likely slot gets filled late, from a list of names someone else suggested, and then goes unresearched until it is the only offer on the table.

Where do you find the real numbers for a school?

Not from a ranking site, and not from a forum. Nearly every American college publishes a Common Data Set — a standardized survey that colleges fill out in the same format every year, usually posted by the school's institutional research office. Search the school's name plus "Common Data Set" and you will normally find a PDF or a spreadsheet on the school's own domain.

Three items in it do almost all the work:

• C1 — applications, admits, and enrolled students. The item asks the school to "provide the number of degree-seeking, first-time, first-year students who applied, were admitted, and enrolled." Divide admits by applicants and you have the admit rate, straight from the school, with no rounding by a third party.
• C7 — the relative importance of admission factors. The school rates each factor as very important, important, considered, or not considered: rigor of the secondary school record, class rank, GPA, test scores, the essay, recommendations, interview, extracurriculars, character, first-generation status, alumni relation, and, notably, "level of applicant's interest." This tells you what a school says it weighs, in its own words.
• C9 — the test score profile. The 25th, 50th, and 75th percentile SAT and ACT scores, plus what share of students submitted a score at all.

C9 carries a trap worth understanding, because it changes how you read every score range you will see this fall. The item reports scores for students enrolled in the fall — not students admitted, and not students who applied. Two filters have already been applied before that number reaches you. First, admitted students who chose a different school are not in it. Second, at a school where testing is optional, the percentile range describes only the subset who chose to submit, and students submit when their scores help. So the published range is systematically higher than the range of scores that were actually good enough to get in. Being at the 25th percentile of that range is a weaker signal against you than it looks.

One more piece of context: the most recent Common Data Set describes the class that arrived last fall, which means it describes admission decisions made the cycle before that. The 2025–26 edition, for example, reports on students who enrolled in fall 2025. That lag is fine for sorting a list; it is not fine for assuming a policy still holds. Read policy — testing requirements especially — off the school's current admissions page, which is exactly the discipline behind our guide to which colleges require the SAT or ACT.

Why does being "in range" not make a selective school a target?

Because at the most selective schools, the range stops predicting anything.

Work through the arithmetic for a school that admits a small single-digit percentage of its applicants. The overwhelming majority of the people it turns down are academically qualified — inside the published range, with the grades and the coursework. There are simply many more qualified applicants than seats. So a student who matches the profile is not competing against the unqualified pile; they are competing against a very large group of people who look much like them, on criteria the school does not fully publish.

The practical consequence is a rule that costs nothing to follow: at a school with a very low admit rate, every applicant is a reach. Not because your record is weak, but because the base rate dominates everything a list can account for. Read the admit rate off item C1 and let it override your read of the score range whenever the two disagree.

This is also why a list made of eight schools that all feel like slight stretches is not a balanced list — it is eight reaches with a comforting story attached. If you cannot name the school on your list that is very likely to admit you, you have not built a list yet.

What does an affordable list look like?

Every list needs a financial likely alongside the admissions one: a school you can attend on money you know you will have, without depending on an aid package that may not arrive.

The tool for this is better than most families realize and almost nobody uses it. Under the Higher Education Opportunity Act, every institution participating in federal Title IV student aid must post a net price calculator on its website — the Department of Education describes it as "designed to help current and prospective students, families, and other consumers estimate a student's individual net price at that institution." Net price is the number that matters: sticker price minus the grant aid a family at your income is likely to receive. It is often dramatically lower than the published cost, and it varies enormously between two schools with nearly identical sticker prices.

Run the calculator for every school before it goes on the list, not after. Twenty minutes per school, done in August, is what prevents a spring where the only affordable offer is from a school nobody researched. The estimate is exactly that — an estimate, based on the data you type in — but a rough number now beats a precise number in April.

Two adjacent points, both easy to miss:

• Application costs are real too. Fees add up across a long list, and Common App notes that with either its own fee waiver, "which your counselor must confirm," or a college-specific waiver, "you will not be required to pay the fee to submit your application." If cost is shaping how many schools you apply to, ask your counselor about waivers before you cut schools.
• Merit money and admission are different questions. A school can be a likely admit and still not be affordable, and some scholarships have their own requirements — including test scores at schools where admission itself does not require them.

What does the application system itself constrain?

Three mechanical facts shape the shape of a list, independent of any strategy.

• The Common App caps you at 20 colleges. Common App's applicant guidance is that you may add and apply to up to 20 colleges, and once you submit an application to a school, that school cannot be removed from your list. More than 1,100 colleges are members, so the cap is rarely the binding constraint — but it is a hard ceiling, and it is worth knowing before you treat the list as open-ended. How many of those 20 you should actually use is a separate question, and a real one; it deserves its own answer rather than a number pulled from the air.
• A binding early application removes options from the list. Early Decision is a commitment, not a preference. Penn's admissions site states the terms plainly: an ED applicant "cannot apply to another Binding Early Decision or Restrictive Early Action program at another private institution," and if admitted, "must withdraw all other applications and enroll." Non-binding and rolling applications elsewhere are still allowed. If a binding early school is on your list, decide that before November rather than during it.
• Some systems are one application with one deadline. The University of California is the clearest case: one application covers all nine undergraduate campuses, and the fall filing period runs October 1 to November 30, with no early round to fall back on. Adding UC campuses to a list is a different decision from adding nine separate schools.

What actually goes wrong with a list?

Four failure modes, in roughly the order they show up:

• The list is built last. Every downstream task depends on it — supplemental essays, which teachers to ask, which test dates matter — so a list that is still forming in October pushes everything else late. Our senior year college application timeline puts the list in August for exactly this reason.
• The likely school is a placeholder. Chosen by reputation-adjacent guesswork, never visited, never researched, and quietly resented. Fix it by treating the likely slot as the one that most deserves real research, since it is the school you are most likely to attend.
• Affordability is deferred. No net price calculator was run, so the financial picture arrives with the aid letters, months after the list could have been changed.
• The list is sorted by feeling. Schools get filed as reach or target based on how impressive they sound rather than on item C1. This is the error the Common Data Set exists to fix, and it takes an afternoon to fix it.

How do you build the list, in order?

A sequence that works, and that you can finish in a weekend:

1. Write down what you are optimizing for. Size, distance from home, what you might study, whether you need a specific program. Three or four criteria, honestly ranked. Everything downstream is filtering.
2. Draft a long list. Generous, unsorted, no categories yet. Include schools you know nothing about that meet your criteria.
3. Pull the Common Data Set for each. Admit rate from C1, score profile from C9, and a glance at C7 to see what the school says it weighs. Twenty minutes each, at most.
4. Sort into reach, target, and likely using the admit rate first and your position in the range second. Apply the low-admit-rate rule without flinching.
5. Run the net price calculator on every survivor. Cut anything you cannot pay for that has no realistic path to aid, and make sure at least one affordable likely school remains.
6. Check testing policy per school, off its own admissions page. This determines whether a fall test date needs to be on your calendar at all, and the test itself is not the one most parents sat — what actually changed on the digital SAT covers the format if a score is in play.
7. Freeze it, then work from it. Ask your recommenders with the final list in hand, so you know how many letters each school needs and which deadline is earliest.

The freeze matters more than the sorting. A list that keeps changing through September means supplemental essays that keep restarting, and supplements are where the fall actually gets lost.

The part that is not a research problem

None of the above is intellectually hard. It is a weekend of reading PDFs and typing income figures into calculators — work that is genuinely boring, has no deadline of its own, and competes for attention with a device engineered to win that competition. That is why lists slip to October: not because families cannot do the research, but because nothing forces the afternoon to happen.

It helps to treat the list as a small number of concrete, finishable tasks rather than one vague project. "Look up the Common Data Set for four schools" is a task you can finish before dinner. "Figure out where to apply" is not a task at all, which is why it survives on a to-do list for six weeks. The same principle underlies most of what actually works against an infinite feed: shrink the unit of work until starting is easier than avoiding it.

What to do this week

Two things. First, write down the schools you are already considering and look up the admit rate for each one in its own Common Data Set — not a ranking site — and see how many of them are genuinely reaches once you use the real number. For most students the honest count is higher than expected, and finding that out in August is a gift.

Second, pick the one school on the list you would be happiest to attend among those most likely to admit you, and run its net price calculator tonight. If no school fits that description, that is the gap to close first, and it is the only part of this that gets harder the longer it waits.