What a grade curve actually does
When a test turns out harder than intended — the class average lands at 58 instead of the expected 75 — many instructors adjust the scores afterwards rather than letting one brutal exam define the term. That adjustment is a curve, and despite the folklore around "grading on a curve", in practice it almost always means one of the three mechanical methods on this page: take square roots, add flat points, or rescale so the top score becomes 100. This calculator applies all three to your raw score at once, so you can see what each would do to you and to scores across the range — and it is honest about the limit of the exercise, which is that only your instructor decides whether a curve happens at all. For the uncurved arithmetic of scores and course grades, start with the Grade Calculator and the Weighted Grade Calculator.
The three curve formulas
The square-root curve is the most elegant: because the square root grows steeply at small numbers and flattens at large ones, it helps low scores most and perfect scores not at all — a raw 64 becomes 80, since 10 × √64 = 10 × 8, while 100 stays 100. The flat curve is the bluntest: everyone gains the same points, so gaps between students are preserved exactly and only the level shifts. The scale-to-top curve treats the class's best raw score as the new 100 and stretches everything proportionally beneath it: with a top score of 88, a raw 64 becomes 64 ÷ 88 × 100 ≈ 72.73. Same raw score, three different gifts — 80, 74 and 72.73 with the defaults on this page.
Who each curve helps, and by how much
The comparison table under the calculator is the heart of the matter, because the methods distribute generosity very differently. The square-root curve is progressive: it adds about 50 points to a raw 25 (which becomes 50), 16 to a 64, and nothing to a 100, compressing the class toward the top. The flat curve is neutral: ten points is ten points whether you scored 30 or 85 — though at the very top the 100 cap quietly confiscates part of the gift, since a 95 can only rise to 100. The scale curve is proportional: everyone gains the same percentage increase, so absolute gaps widen slightly even as relative standing is perfectly preserved — a student at 44 gains about 6 when the top is 88, while a student at 80 gains about 9. Which is "fairest" depends on what went wrong with the test: a uniformly brutal exam argues for flat or scale, a test whose hardest questions sank the bottom of the class argues for the square root's rescue of low scores.
Curves you may have heard of that are not these
Two other uses of the word deserve separating out. Grading on a bell curve — forcing fixed proportions of A's, B's and C's regardless of absolute scores — is a different and much harsher practice, rare in ordinary classrooms precisely because it can lower grades as well as raise them; nothing on this page does that. And norm-referencing to the class average, where scores are shifted so the mean hits a target like 75, behaves like a flat curve whose points are set after the fact by the gap between the actual and desired averages. If your instructor announces a curve, the productive questions are which method, with what settings, and whether there is a cap — the three inputs this page needs. Until then, treat any curved figure here as a well-defined "what if", and keep planning your course grade on the raw numbers, which are the only ones guaranteed to exist.
Worked example
Your raw score is 64%. Square-root curve: 10 × √64 = 10 × 8 = 80%. Flat curve: 64 + 10 points = 74%. Scale curve with a class top score of 88%: 64 ÷ 88 × 100 ≈ 72.73%. With the square-root method selected, your curved score is 80% — the method choice alone is worth more than six points here, which is why the table under the calculator shows all three at once.
How to use this calculator
- Enter your raw score as a percentage.
- Choose the curve method your instructor announced — or try each one to compare.
- For the flat method, enter the points being added; for the scale method, enter the highest raw score in the class.
- Read your curved score in the headline panel, with all three methods compared in the breakdown.
- Scroll to the table to see what the same curve does to scores across the whole range, from 0 to 100.
Why instructors curve — and why some refuse to
The respectable case for curving is measurement honesty: a test is an instrument, and when an instrument reads low for an entire class — because two questions were ambiguous, or a topic was taught in half the usual time — adjusting the readings is more truthful than pretending the class suddenly got worse. Curves also protect comparability between sections and terms taught by different people with different exams. The case against is equally real: routine curving can inflate grades until they stop signalling anything, it can reward a class for collective under-preparation, and a curve announced in advance changes incentives in ways not every instructor wants. Most settle in the middle — reserving the curve for demonstrably faulty tests — which is why students should experience it as occasional relief, never as a study strategy.
Practically, never bank a curve you have not been promised, and never assume its shape. A rumoured "ten-point curve" might be flat points, might be the square-root method (which gives a score of 64 a sixteen-point lift, not ten), or might be a rescale worth less than either. Run the actual announced method through this page, fold the result into your weighted course grade, and plan the rest of the term on the verified number.