How the bell curve works at NUST and LUMS, why it makes marks-based prediction impossible, and how to actually strategise under each system.
Every grading system answers one of two questions.
Absolute grading asks: how much of the material did this student master? Your marks map to a letter through a fixed published table. At COMSATS, 85% is an A. It is an A if the whole class scored 90%, and it is an A if the whole class scored 40%. Your grade is a function of your performance alone.
Relative grading asks: how did this student perform compared with their peers? Your letter depends on where you sit in the distribution of your cohort, because typically measured against the class mean and standard deviation. At NUST, 85% might be an A in one section and a B+ in another, depending entirely on what everyone else scored.
These are not minor variations. They produce different incentives, different strategies, and different answers to the question "what grade am I going to get?"
The mechanics vary, but the common approach positions each student relative to the class mean (μ) using the standard deviation (σ) as the unit of measurement. Roughly:
NUST describes its approach as a distribution-gap method on a 1.00–4.00 scale. The crucial consequence is that NUST publishes no university-wide percentage-to-letter table for most undergraduate courses, because no such fixed table exists. Any website showing you "85% = A at NUST" has invented it.
This is the single most common factual error we found while building this site. It matters because a student who believes 85% guarantees an A will make bad decisions about which courses to prioritise.
You cannot predict your letter from your marks. Under relative grading, 78% tells you nothing until you know the class mean. Students who calculate "I need 82% for an A" are solving the wrong problem.
A hard exam is not necessarily bad news. If an exam is brutal and the class average collapses to 45%, your 62% could be well above the mean and earn a strong letter. Under absolute grading, that same 62% might be a C. Relative grading absorbs badly calibrated assessments in a way absolute grading does not.
Your classmates' performance affects your grade. This is uncomfortable but true, and it is why relative grading is sometimes criticised for discouraging collaboration. It also means cohort composition matters: the same performance in a stronger cohort yields a lower letter.
Grade points remain fixed. This is the part that confuses people. The assignment of letters is relative, but once you have an A it is worth 4.00, the same as everyone else's A. The relativity is entirely in how the letter is chosen, not in what it is worth.
Work backwards from the published table. If A− begins at 80% and you are sitting at 76% with a 40%-weight final, our final exam calculator tells you the exact mark you need. Grade boundaries are knowable, which makes effort allocation a solvable optimisation: find the courses where a small push crosses a boundary, and spend your time there.
Your objective is to beat the mean, so gather information about the mean. Pay close attention to how the class found each assessment. Marked components that many students neglect (participation, small assignments, lab reports) are unusually valuable. The mean on those components is often low, and the cost of doing them well is small.
Group study becomes genuinely double-edged: helping classmates raises the mean you are measured against. Most students still find the net effect positive, because understanding improves faster in discussion than in isolation. But be honest that the incentive exists.
Some universities do not pick one. FAST-NUCES permits both, with the grading scheme set per course and announced by the instructor in the course outline at the start of the semester.
This means a FAST student can be simultaneously enrolled in an absolutely graded course, where 86% reliably earns an A, and a relatively graded one, where the same 86% depends on the cohort. Managing both mindsets in one semester is genuinely harder than managing either alone.
At a mixed-system university, do one thing in week one. Read every course outline and write down which scheme applies to each course. If an outline does not say, ask. It changes how you should approach the entire semester, and instructors are required to tell you.
| University | Mode | Fixed percentage bands published? |
|---|---|---|
| NUST | Relative | No, distribution-based |
| LUMS | Relative (most courses) | No |
| FAST-NUCES | Mixed, declared per course | Yes, for absolutely graded courses |
| COMSATS | Absolute | Yes |
| UET Lahore | Absolute / modified | Yes |
| University of the Punjab | Absolute | Yes |
| IITs | Relative | No |
Check your own institution rather than assuming from this table, and policies are revised, and faculty-level exceptions exist. NUST's School of Art, Design & Architecture, for instance, is documented as using absolute grading in some courses even though the university is otherwise relative.
Understanding the institutional reasoning helps you predict how your own university will behave when a course goes badly.
The case for relative grading is calibration. Assessment difficulty varies enormously between instructors, semesters and disciplines. An exam that one professor considers routine may be punishing from another. Under absolute grading, that variation transfers directly onto students as luck. Relative grading normalises it: whatever the paper's difficulty, the distribution of letters stays roughly stable. This is why research-intensive universities with large, varied teaching staff tend to prefer it, and it is a genuinely student-protective feature when an assessment is badly calibrated.
The case for absolute grading is transparency and fairness of a different kind. Students know in advance exactly what earns each letter. Effort maps predictably to outcome. There is no scenario where you improve and your grade falls because your classmates improved more. It is also far easier to defend to an external body, and professional accreditation councils and employers can read an absolute transcript without knowing anything about your cohort.
The honest objection to relative grading is that it converts education into a zero-sum competition among classmates, which is corrosive in exactly the collaborative settings that produce the best learning. Universities are aware of this. Most that use relative grading soften it, guaranteed absolute floors, capped proportions of low grades, or instructor discretion to shift the whole distribution upward when a cohort genuinely performs well.
Myth: "the curve means only a fixed number of As are available." Usually false. Most modern relative systems position students against the mean using standard deviations rather than allocating fixed quotas. If a cohort is genuinely strong and tightly clustered above the mean, a large share can earn top grades. Hard quotas exist but are less common than folklore suggests, and where they do exist, they are normally published.
Myth: "the curve can only help you." False, and it is the more important direction. If the class average is 82% and you scored 79%, you are below the mean and the curve moves your letter down relative to what an absolute table would have given you. Curves cut both ways.
Myth: "one very strong student ruins the curve." Largely false. A single outlier moves the mean of a 100-student cohort by a fraction of a mark and slightly widens the standard deviation. Distribution-based systems shrug off individual outliers; they are sensitive to the shape of the whole cohort.
Myth: "I can calculate my curved grade if I know my percentage." False, and this is the one that costs students real effort. You need the class mean and standard deviation, which are usually not published until grades are released. Any tool claiming to predict a curved letter from your raw mark alone is guessing. This is precisely why our university pages show no percentage table for relatively graded institutions, because the honest answer is that the mapping does not exist in advance.
These questions are entirely reasonable, and instructors expect them. Asking in week one costs nothing and changes how you should spend the semester.
Write the answers down. In a mixed-system semester you will not remember which course used which scheme by the time finals arrive, and by then the information determines how you allocate your revision time.
Run your own numbers with a calculator built on your university's actual grading scale.