Reach, Target, and Safety: Build a Balanced College List

Reach, target, and likely are useful planning labels. They are not personal admission probabilities.

That distinction matters because the same university can represent very different levels of uncertainty for two applicants, and sometimes for the same applicant applying to two different programs. Overall admit data may group colleges or majors together. Some programs admit directly to the major. Residency, application pathway, and capacity can also change the comparison.

The University of California makes this limitation explicit on its own first-year admit data page: published figures are a general guide to selectivity, not a predictor of an individual's admission chances. UC also notes that competition can vary among colleges and majors and from one cycle to the next.

So the job is not to find a magic cutoff. The job is to document what you know, what admission unit you are actually applying to, and what remains uncertain.

Quick navigation

What a college-list category can and cannot tell you
Check the actual admission unit first
Build a classification from evidence, not one cutoff
Read scattergrams with their limits
A worked example of an uncertain "likely"
Add affordability and willingness to attend
What to record in your college-list worksheet

What a college-list category can and cannot tell you

A category can help you distribute application effort and avoid an all-reach list. It can answer a planning question such as:

Does this list contain enough options that are meaningfully less uncertain than our most selective choices?

It cannot answer:

What is this student's exact chance of admission?

That second question usually requires information colleges do not publish, and even published historical data describe prior groups, not a personal probability for the next cycle.

Reach

Use reach when admission is highly uncertain for this applicant at the level that actually makes the decision. That might be the university, a college within the university, a direct-admit major, or another limited pathway.

Common reasons include:

the institution or program is highly selective,
the student's academic record is outside or near the edge of the available historical range,
the intended major is materially more competitive than the university overall,
the applicant is in a different residency or admission category than the headline data,
the school uses a holistic process in which many academically qualified students are denied,
the family cannot verify enough program-level information to justify a less cautious classification.

A reach can still be an excellent application choice. The label means the list should not depend on that result.

Target

Use target when the student's record is plausibly competitive based on current, relevant evidence, but admission still carries meaningful uncertainty.

A target should survive more than a quick comparison to a university-wide GPA or admit-rate number. Check:

the correct applicant type,
the correct campus, college, or major,
the current cycle's published rules,
the student's academic context,
any residency or pathway differences,
whether the program has additional requirements,
whether the historical data you are using actually describe the group you are comparing.

"Target" means reasonable to include as part of the middle of the list, not "we should expect this college to admit the student."

Likely

Use likely only when the evidence supports a clearly less uncertain admission outcome than the rest of the list and you have checked for hidden constraints.

Before using the label, ask:

Is admission to the university the same as access to the student's intended program?
Is the program capacity-limited, impacted, audition-based, portfolio-based, or otherwise separately selective?
Is the student comparing against the right residency and applicant category?
Does the school publish automatic or clearly defined admission criteria, or are we inferring from historical outcomes?
Is the school affordable under a realistic family scenario?
Would the student actually attend?

A likely is not a guarantee. If a major-specific rule is unknown or the program itself is selective, keep the uncertainty visible instead of forcing the school into a reassuring category.

Check the actual admission unit first

The most common classification error is using data from one level of a university to answer a question decided at another.

Before assigning any label, fill in this sentence:

I am applying as a [applicant type] to [campus/university], for [college/major/program/pathway], in the [cycle].

Then ask where admission is actually decided.

Examples:

A university admits first, and the major is declared later.
A university admits directly to a college, such as engineering or business.
A major is impacted or capacity-limited and uses additional criteria.
A student is admitted to the university but not guaranteed the intended major.
An alternate-major policy may exist, but access depends on remaining capacity.

If you cannot answer this question from a current official source, the uncertainty belongs in your list notes.

Build a classification from evidence, not one cutoff

Use a short evidence stack for every school.

1. Official admission structure

Find the current page that explains who is being admitted and to what. Prefer the university, system, college, or program source that owns the rule.

2. Historical data for the closest comparable group

Use campus, college, or program data when available. Record the cycle. Do not quietly mix data from different years or applicant types.

3. Student academic context

Compare the student's record with the information the institution actually publishes. Do not reduce context to one GPA number if the institution evaluates course rigor, school context, grade trajectory, or other academic information.

4. Known selection constraints

Record major capacity, portfolios, auditions, prerequisite coursework, residency treatment, local preference, or another rule that can materially change admission.

5. Unknowns

Write them down. "Program-level data not published" is more useful than inventing precision.

What scattergrams can and cannot establish

A school scattergram can be useful historical context. It can show how students from the same high school with certain academic records fared in prior years.

It still does not tell you why a particular student was admitted or denied.

Scoir, one of the major platforms that displays scattergrams and predictive information, says its scattergrams and predictive chances are intended as guideposts, not predictions of admission results. Scoir also notes that schools configure access differently, the amount of local data can vary, and counselors should help interpret factors that the models do not capture.

When you use a scattergram, record:

the years represented,
approximate sample size if visible,
whether application round is separated,
whether major or college is separated,
whether test-optional applicants are mixed with score submitters,
what the graph does not show.

Do not convert a cluster of green dots into "our high school has an advantage" unless the institution itself documents a relevant policy.

A worked example: the uncertain "likely"

Consider a hypothetical Bay Area student, Maya.

Maya's academic record looks strong compared with the university-wide information published by a large public institution. A school scattergram also shows several recent admits with similar grades.

At first glance, the family wants to call the university a likely.

Then they check the intended program.

Maya is applying directly to a capacity-limited computer science program. The university explains that program access is competitive, but it does not publish a current program-specific admit rate or a student-level probability tool. The scattergram does not identify applicants by major.

The family now has two pieces of useful evidence:

Maya appears academically competitive for the university overall.
The intended program introduces a material uncertainty the university-wide data do not resolve.

The correct response is not to invent a computer-science acceptance probability. The family can classify the school more cautiously, note the program uncertainty, and add another option where both university admission and access to the intended academic path are better documented.

That is what a useful list category does. It changes a planning decision when the evidence changes.

Affordability is part of whether a school belongs on the list

Academic admission is only one constraint.

A school should not function as the family's reassuring likely if the student could be admitted but the family would not be able to use the offer.

For every likely and target, record:

current estimated family cost from the college's official net price calculator when applicable,
whether the estimate depends on competitive merit aid,
known additional program costs,
the family's workable annual range,
whether the student would attend at that cost.

A college can be academically accessible and financially unrealistic. Keep those two questions separate.

Willingness to attend is also part of the list

A likely that the student would never choose is not doing the job the list needs.

Before finalizing the category, ask:

If every more selective option said no, would I be willing to enroll here at the cost we currently expect?

If the answer is no, research a different school rather than treating the application as insurance.

What to record in the Reach / Target / Likely Assignment Sheet

For each college or program, add these fields:

Field	What to record
Applicant type	First-year, transfer, international, or another category
Campus / university	Exact institution
College / major / pathway	The admission unit that matters
Cycle	The year or term the rule applies to
Official evidence	Current official URL and date checked
Historical data	Closest comparable published group
Program or pathway constraint	Impaction, direct admit, audition, capacity limit, alternate-major rule, etc.
Scattergram note	What it shows and what it does not show
Affordability	Current estimate and assumptions
Willing to attend?	Yes / uncertain / no
Unresolved risk	The question still needing verification
Working label	Reach / target / likely, with a short reason

The label should be the last field you fill, not the first.

Use the Reach / Target / Likely Assignment Sheet to keep the evidence beside the category.

A better list-level check

Instead of asking whether you have exactly a certain number of reaches, targets, and likelies, ask:

Do we have multiple colleges where the student has a well-supported, less uncertain path to admission?
Are those colleges affordable enough to be usable options?
Would the student attend them?
Have we separated university admission from access to the intended major?
Have we documented the biggest unknown for every school?
Does the list still work if every highly uncertain application is denied?

That produces a resilient list without pretending admissions can be reduced to a fixed formula.

Related resources

If a classification feels reassuring only because one number makes it look precise, keep researching. The strongest college lists make uncertainty visible and still give the student several options they would be glad to use.