Grade Curve Calculator
Paste your class's test scores, pick a curve, and see every student's new score and letter grade — in the order you pasted them, ready to copy back into your gradebook.
Paste a spreadsheet column or type scores separated by commas, spaces or new lines. Lines like Name, 78 keep the name.
Leave at 100 if your scores are already percentages.
Scores stay in your browser. Nothing you paste is uploaded or saved.
Add points to everyone
Everyone gained 5 points. The class average moved from 73.5 to 78.5. 3 of 6 letter grades changed.
Curved scores copy one per line in the same order you pasted them, ready to paste back into a gradebook column.
| # | Name | Original | Curved | Change | Letter |
|---|---|---|---|---|---|
| 1 | Ava | 55 | 60 | +5 | F to D |
| 2 | Ben | 62 | 67 | +5 | D |
| 3 | Chloe | 70 | 75 | +5 | C |
| 4 | Diego | 78 | 83 | +5 | C to B |
| 5 | Emma | 85 | 90 | +5 | B to A |
| 6 | Farah | 91 | 96 | +5 | A |
| Statistic | Before | After |
|---|---|---|
| Students | 6 | 6 |
| Average | 73.5 | 78.5 |
| Median | 74 | 79 |
| Std. deviation | 12.6 | 12.6 |
| Lowest | 55 | 60 |
| Highest | 91 | 96 |
- AA: 1 before, 2 after
- BB: 1 before, 1 after
- CC: 2 before, 1 after
- DD: 1 before, 2 after
- FF: 1 before, 0 after
Compare all six methods
Your scores under each method, with the same cap, protection and rounding settings. Pick one to use it.
| Method | New average | Letters changed | F grades |
|---|---|---|---|
| 78.5 | 3 | 0 | |
| 82.5 | 4 | 0 | |
| 80.7 | 3 | 0 | |
| 85.3 | 5 | 0 | |
| 75.5 | 1 | 1 | |
| 75.5 | 1 | 0 |
Which curve should you use?
Every curve answers a slightly different question, so the right one depends on why the scores came out low.
- Add points to everyone (flat). Best when the whole test was too hard by roughly the same amount — a confusing question, a topic you ran short on. Everyone gets the same bonus, so the gaps between students stay exactly as they were.
- Highest score becomes 100 (add the gap).A flat curve where the test sets the size: if the best paper scored 91, everyone gets 9 points. Fair when your strongest student's miss tells you what was unreachable. Weak when one outlier aced it (no curve at all) or the top score was itself low (a huge one).
- Scale so the highest = 100 (multiply). Every score is multiplied by 100 ÷ top score. Higher scores gain more points than lower ones, which rewards students who were already close.
- Square root. 10 × √score. Lifts the bottom of the class hardest — a 36 becomes 60 — while a 95 gains only 2.5 points. Useful when a test was badly pitched for students who were struggling, but it compresses the class: a 49 and a 100, 51 points apart, become 70 and 100, only 30 apart.
- Target average.You pick the class average you think the test should have produced, and everyone gets the same bonus to reach it. With the 100 cap on, students at the cap can't take their share, so the tool adds a little more to everyone else until the average actually lands on the target.
- Bell curve. Converts each score to a z-score and rebuilds the class around a mean and standard deviation you choose: new = target mean + z × target SD. It is the only method that controls spread as well as level, and the only one that can push scores down.
Worked example: one class, six curves
Six students scored 55, 62, 70, 78, 85 and 91. The class average is 73.5, the median 74 and the standard deviation 12.6. On a 10-point scale that is one F, one D, two Cs, one B and one A. Here is what each method produces with the default settings (capped at 100, never lower a score, whole numbers):
| Method | Curved scores | New average |
|---|---|---|
| Original | 55, 62, 70, 78, 85, 91 | 73.5 |
| Flat +5 | 60, 67, 75, 83, 90, 96 | 78.5 |
| Top → 100 (add 9) | 64, 71, 79, 87, 94, 100 | 82.5 |
| Scale (× 100/91) | 60, 68, 77, 86, 93, 100 | 80.7 |
| Square root | 74, 79, 84, 88, 92, 95 | 85.3 |
| Target average 75 | 57, 64, 72, 80, 87, 93 | 75.5 |
| Bell curve (mean 75, SD 10) | 60, 66, 72, 79, 85, 91 | 75.5 |
A few things stand out. Flat and top-to-100 move everyone by the same amount; scaling gives the 55 five points and the 91 nine. The square root curve gives the 55 nineteen points and the 91 four, and it is the only method here that leaves nobody below a C. The exact target-average shift is 1.5 points, but whole-number scores can only move by whole points, so the tool picks +2 and the average lands on 75.5 (set rounding to one decimal and it hits 75.0 exactly).
The pitfalls worth knowing
- A bell curve can lower scores.In the example the class spread (SD 12.6) is wider than the target (SD 10), so the 85 would fall to 84 and the 91 to 89 — an A turned into a B by a “curve.” That is why “Never lower a score” is on by default; turn it off only if lowering is what you intend.
- Square root helps low scores most. That is its purpose, but it narrows the distance between your strongest and weakest students. If the grade has to reflect a meaningful difference in mastery, a flat curve preserves it better.
- One outlier can switch a curve off. Both top-score methods do nothing if a single student scored 100. Look at the second-highest score before deciding the test was fair.
- Small classes make bell curves noisy. With fewer than about 20 students, the mean and standard deviation move a lot from one paper to the next, so a bell curve mostly reshuffles noise.
- Decide before you look at names. Choose the method based on the test, then apply it to everyone. The comparison table makes it tempting to shop for the curve that rescues a particular student.
Getting scores in and out
Copy a column of scores from Google Sheets, Excel or your gradebook export and paste it straight in; tabs, commas, spaces and new lines all work, and a header row like “Score” is skipped and listed so you can see it. If you paste two columns — names and scores — the names are kept. Copy curved scores gives you one score per line in the original order, so it pastes back into the same column without re-sorting. Grading the stack first? Use the easy grader to turn wrong answers into percentages, then curve them here.
Frequently asked questions
How do I curve grades?
Paste the class's scores (one per line, or straight from a spreadsheet column), choose a method, and read off the curved column. The simplest fair curve adds the same number of points to everyone; the most common is raising the highest score to 100 and giving everyone that same gap.
What is the square root curve?
Each score becomes 10 × √score. A 49 becomes 70, a 64 becomes 80, an 81 becomes 90 and a 100 stays 100. It lifts low scores far more than high ones and never pushes anyone past 100.
How does a bell curve work for grades?
Each score is turned into a z-score — how many standard deviations it sits above or below the class average — and then placed at the same z-score on a new distribution you choose: new score = target mean + z × target SD. Students keep their rank, but the spacing between them changes.
Can curving lower a student's grade?
A bell curve can. If you pick a smaller standard deviation than the class actually had, or a target average below the real one, scores at the top move down. The 'Never lower a score' option, on by default, keeps anyone who would drop at their original score.
Does this upload my students' scores?
No. Everything is calculated in your browser and nothing you paste is sent or stored. Only your chosen method and options are remembered on this device, so the tool opens the way you left it.
My test was out of 40 points, not 100. Does that work?
Yes. Set 'Test is out of' to 40 and paste raw points. Scores are converted to percentages for curving, and you can copy the curved scores back as points or as percentages.
Which standard deviation does the bell curve use?
The population standard deviation of the scores you paste, because the class is the whole group being curved. A spreadsheet's STDEV function gives the slightly larger sample figure, so the two will differ a little for small classes.