Apply flat curves, square root curves, scale-to-highest, custom multipliers, and more to any set of student scores. See grade distributions, before/after comparisons, letter grade assignments, and export results — all free and private.
Enter scores, choose your curve method, and instantly see curved grades with full distribution analysis
| # | Student | Original | Curved | Change | % | Grade |
|---|
A complete grading toolkit built for educators — powerful curve methods, detailed analytics, and export-ready results.
Add a fixed number of points to every student's score. The most transparent method — every student benefits equally regardless of their original score. Set any point value from 1 to 30.
Scales all scores proportionally so the highest score in the class becomes 100. Students who scored highest benefit most; every score is adjusted proportionally relative to the top performer.
Multiplies the square root of each score by 10. Benefits lower scores significantly more than higher scores — a common choice when low scores are very low but high scores are already near the maximum.
Multiplies every score by a constant factor. A multiplier of 1.05 adds 5% to every score proportionally — students with higher scores benefit more in absolute terms than students with lower scores.
Enter any mathematical expression using "s" as the score variable. Supports all standard math operators and functions: multiply, add, subtract, Math.sqrt(), Math.min(), Math.pow() and more.
Visualise how the curved scores distribute across score ranges with a before/after bar chart — instantly see how your curve shifts the class distribution toward higher grades.
Automatically assigns letter grades (A, B, C, D, F) based on your chosen scale — standard, plus/minus, lenient, strict, or fully custom cutoffs you define yourself.
Define your own A/B/C/D percentage cutoffs to match your institution's grading policy. Your custom scale is applied instantly across all curved scores and shown in the results table.
Optionally cap all curved scores at the maximum possible score (no student can exceed 100%). Optionally round curved scores to the nearest integer for clean, presentable gradebook entry.
Paste an entire class's scores at once — comma-separated or one per line — and the calculator adds all students instantly. No tedious row-by-row entry for large classes.
Download a complete CSV file with student names, original scores, curved scores, point changes, percentages, and letter grades — ready to import into any gradebook or spreadsheet application.
All calculations happen locally in your browser. Student scores never leave your device. No account required, no data stored, no analytics — completely private for sensitive educational data.
What grading on a curve actually means, when it is appropriate, how each method works, and how to choose the right approach for your class
The phrase "grading on a curve" is used so loosely in everyday conversation that it has come to mean almost anything that involves adjusting grades upward. Students use it to mean "the teacher made the test easier after the fact." Parents use it to mean "the grading is subjective." But in its most precise sense, grading on a curve means adjusting a set of scores so that they conform to a predetermined statistical distribution — typically a normal distribution — regardless of what the raw scores are. In practice, most teachers who curve grades are not applying statistical bell curves at all; they are using one of several simpler adjustment methods to bring an unexpectedly low set of scores up to a more appropriate level.
Understanding the distinction matters because different methods have very different effects on different students. A flat addition curve treats every student identically — adding five points to a 55 brings it to 60, just as adding five points to an 89 brings it to 94. A square root curve, by contrast, has a much larger proportional effect on low scores than on high ones. A scale-to-highest curve rewards the top performer by making their score 100 and proportionally adjusts everyone else. These are not just technical differences — they encode fundamentally different philosophies about fairness, effort, and the purpose of grading.
The most common reason a teacher curves grades is that an exam turned out to be harder than intended. This happens for a variety of reasons: ambiguous question wording, topics covered more briefly in class than the test demands, time pressure that most students could not manage, or simply a test that was calibrated for a stronger cohort than the current class. When the class average on a 100-point exam comes back at 58, curving is a reasonable corrective mechanism — the test failed to accurately measure the range of knowledge in the class, so the scores need adjustment.
A second reason is alignment with grading expectations. Many institutions and departments have informal norms about what a class average should look like — somewhere between 70 and 80 for a reasonably challenging course, for example. If a professor's exam produces a 62 average and they know the class performed at roughly the expected level based on homework, participation, and office hours interactions, curving the exam scores to bring the average up to 73 or 75 reflects their judgment that the test was the outlier, not the students' preparation.
A third, less common reason is norm-referenced grading — intentionally grading students relative to each other rather than against an absolute standard. In large university lecture courses, some professors deliberately set tests that very few students can ace, then curve aggressively to produce the desired distribution. This approach has significant equity implications and is increasingly scrutinised by education researchers, but it remains common in certain disciplines.
Each curve method works differently and produces different outcomes for different score distributions. Choosing the right method requires understanding what each one does mathematically and what values it reflects about fairness in grading.
The flat addition curve adds a fixed number of points to every student's raw score. If the teacher decides to add seven points, every student gets seven points — the student who scored 45 gets 52, and the student who scored 91 gets 98. This method is maximally transparent and equitable in the sense that every student benefits identically. It does not change the relative ranking of students or the distribution shape at all — it simply shifts every score up by the same amount.
The flat addition method is most appropriate when the entire test was uniformly too hard — when every student's score is depressed by roughly the same number of points relative to where it should be. It is less appropriate when the difficulty was concentrated in specific questions that some students answered correctly and others did not, because those students are not equally "owed" the same adjustment.
The scale-to-highest method takes the highest score in the class and scales all scores proportionally so that the highest becomes 100. If the highest score was 88, each student's score is multiplied by 100/88 (approximately 1.136). A student who scored 72 would receive a curved score of approximately 82. This method is grounded in the philosophy that the hardest-working or most capable student in the class set the de facto ceiling for the exam, and everyone else should be measured relative to that ceiling.
The scale-to-highest method rewards higher-scoring students more in absolute terms than lower-scoring students, which some view as a compounding of advantage. However, it is mathematically clean and easy to explain to students. It is most appropriate when you believe the test was well-designed but simply too long or too difficult, and that the top scorer's performance represents what 100% should look like given the time and conditions.
The square root curve computes each student's curved score as the square root of their raw score multiplied by 10. A student who scored 64 gets a curved score of 80 (√64 × 10). A student who scored 81 gets a curved score of 90. A student who scored 100 stays at 100. The defining characteristic of this method is that it helps low scores dramatically more than high scores. A student who scored 49 gets curved to 70; a student who scored 81 only gets curved to 90. The curve is non-linear and compresses the score range from the bottom upward.
The square root curve is most appropriate when the score distribution is highly skewed — when many students scored very low but a few scored well. It provides meaningful help to struggling students without dramatically changing the standing of students who already performed strongly. It is a common choice for standardised test preparation contexts and large introductory courses where the difficulty level varies significantly across student populations.
The multiplier method multiplies every score by a constant factor. A multiplier of 1.10 adds 10% to every score. Unlike flat addition, the multiplier gives higher absolute gains to students who scored higher — a student who scored 90 gets 9 bonus points, while a student who scored 50 gets only 5. This method preserves the relative ranking perfectly and widens the gap between high and low scores in absolute terms, while narrowing it in percentage terms. It is most appropriate when you want proportional adjustment — when you feel the test was uniformly calibrated too hard by a consistent percentage factor.
For teachers who want precise control over the curve shape, a custom formula using the student's score as a variable allows any mathematical transformation imaginable. Common custom formulas include combinations of addition and multiplication (score × 1.05 + 3), more complex expressions that cap gains (Math.min(score + 8, 100)), or even formulas that treat score ranges differently. The custom formula option is the most flexible and is particularly useful for teachers who have specific curve outcomes in mind and want to define the transformation precisely rather than relying on one of the standard methods.
The letter grade scale applied after curving is as consequential as the curve itself. The standard American grade scale — A for 90 and above, B for 80–89, C for 70–79, D for 60–69, F below 60 — is deeply familiar but is by no means universal. Many educators use modified scales to reflect their course difficulty or institutional expectations.
| Scale | A | B | C | D | F |
|---|---|---|---|---|---|
| Standard | ≥ 90% | ≥ 80% | ≥ 70% | ≥ 60% | < 60% |
| Lenient | ≥ 85% | ≥ 75% | ≥ 65% | ≥ 55% | < 55% |
| Strict | ≥ 95% | ≥ 85% | ≥ 75% | ≥ 65% | < 65% |
| Plus/Minus | A+ ≥ 97% | B+ ≥ 87% | C+ ≥ 77% | D+ ≥ 67% | < 60% |
A lenient scale is appropriate when the course itself is objectively more difficult than average — advanced placement courses, upper-division electives in demanding fields, or courses with prerequisites that filter for motivated students. Setting the A threshold at 85 rather than 90 acknowledges that strong mastery of difficult material, even if imperfect, deserves recognition as excellent work. A lenient scale is also appropriate when using a strict curve method, to ensure the combination does not systematically disadvantage students who performed well by any reasonable standard.
Grade curving is not without controversy. Critics argue that curving creates a competitive rather than collaborative classroom culture — students are implicitly graded against each other, meaning that a classmate's success makes your grade worse. This concern is most valid for norm-referenced curves, where the distribution of grades is predetermined and some students must fail regardless of their absolute performance. It is less applicable to the adjustment methods described here, which are not zero-sum — every student benefits, and no student's score is reduced.
A more substantive criticism is that systematic curving can mask the quality of instruction. If every exam requires a curve to bring grades to acceptable levels, the issue may not be test difficulty but inadequate preparation, unclear learning objectives, or misaligned curriculum. Curving repeatedly without examining why scores are consistently low may be treating a symptom rather than the underlying problem. The most defensible approach to curving is to treat it as an occasional correction for specific, identifiable circumstances rather than a routine element of the grading system.
Transparency is also essential. Students deserve to know in advance that curving may occur, how it will be applied, and what the curving policy is for the course. Applying a curve retroactively and differently from how it was described — or not describing it at all — erodes trust in the grading process. Our calculator provides a clear, exportable breakdown of exactly how every score was transformed, making it straightforward to share the curve details with students in a transparent and documentable way.
Before applying any curve, look at how your scores are distributed. If most students cluster near the top with a few very low outliers, a flat addition or multiplier curve is appropriate — it helps everyone without dramatically changing the class shape. If the distribution is wide and scores are spread across a large range, the square root method's non-linear compression may be more equitable. If you have a single student who clearly outperformed everyone else on what was otherwise a fair test, the scale-to-highest method rewards that performance while adjusting everyone else proportionally.
A useful approach is to decide what you want the class average to be after curving — say, 75 — and then work backward to find the curve that achieves it. With our flat addition method, you can adjust the points slider until the "Average After" statistic in the results panel matches your target. This goal-oriented approach to curving is more defensible than applying an arbitrary fixed curve and accepting whatever distribution results.
After calculating your curve, use the Export CSV feature to generate a complete record of every student's original score, curved score, and letter grade. This documentation is valuable for your own records and for responding to student grade inquiries with precise, factual information rather than approximations. Sharing the curve method and the specific transformation applied — along with the class statistics before and after — demonstrates the kind of transparency that builds student trust in the fairness of the grading process.