💡 Key Takeaways
- Your raw score is simply correct answers × marks, minus wrong answers × penalty.
- Negative marking decides whether a guess helps you or hurts you — and the maths is easier than it looks.
- Cut off marks depend on vacancies, paper difficulty, applicant numbers and category.
- You can estimate a cut off range using previous-year trends plus a difficulty adjustment.
- In multi-shift exams, normalization can move your final score up or down.
You have written the exam. The answer key is out. Now comes the part that stresses almost everyone — the waiting. You want to know one thing: am I in or am I out?
You do not have to guess blindly. With the answer key in hand and a little arithmetic, you can calculate your score accurately and build a realistic estimate of the cut off. This guide walks through the exact method, step by step.
1. Raw Score vs Official Score — Know the Difference
Before calculating anything, be clear about which number you are working out. These two are not always the same.
- Raw score — what you compute yourself from the official answer key. It is the number of correct answers minus penalties, and it is the number you can work out on result day.
- Official score — what the exam body finally publishes. It may differ from your raw score if there was normalization across shifts, if questions were dropped, or if the final answer key changed after objections.
For a single-shift exam with a final answer key, your raw score is usually your official score. For multi-shift exams, treat your raw score as a close estimate rather than a certainty.
2. The Basic Scoring Formula
Every objective exam score, no matter how complicated the rules look, comes down to one simple formula:
If there is no negative marking, simply drop the second half of the formula.
Knowing your penalty rate
The penalty is almost always written as a fraction of the marks a question carries. A question worth 2 marks with a "1/3 negative marking" rule costs you 0.67 marks when wrong. Here is how the common schemes behave when you guess blindly among four options:
| Negative marking rule | Penalty per wrong answer (question worth 1 mark) |
Net gain from a blind guess (4 options, 1 correct) |
|---|---|---|
| No negative marking | 0 | +0.25 |
| 1/4 negative | −0.25 | +0.0625 |
| 1/3 negative | −0.33 | 0 (break-even) |
| 1/2 negative | −0.50 | −0.125 |
Based on a 4-option question where only one option is correct. Values assume you are guessing with no idea which option is right.
3. Worked Examples You Can Copy
Numbers make this much clearer. Here are two examples using the patterns seen most often in recruitment and entrance exams.
Example A — SSC-style paper
- Total questions: 100
- Marks per question: 2
- Negative marking: 0.5 marks per wrong answer
- Attempted: 92 | Correct: 80 | Wrong: 12
Example B — Bank-style paper
- Total questions: 100
- Marks per question: 1
- Negative marking: 1/4 (0.25) per wrong answer
- Attempted: 88 | Correct: 76 | Wrong: 12
Notice how the penalty in Example A (6 marks) is larger than in Example B (3 marks). A higher marks-per-question value makes every wrong answer more expensive. Always multiply the penalty by the question's marks before subtracting.
4. Negative Marking: Should You Guess?
This is where most candidates lose marks they did not need to lose. The answer is not "never guess" or "always guess" — it depends on how many options you can confidently eliminate.
The idea of expected value
A guess has an expected value: the average marks you would gain if you made that same guess hundreds of times. If the expected value is positive, guessing helps. If it is negative, it hurts.
| Options you can eliminate | Real choices left | Chance of being right | Expected value (1/3 negative, 1 mark) |
Should you guess? |
|---|---|---|---|---|
| None | 4 | 25% | 0.00 | Break-even — skip it |
| 1 option | 3 | 33% | +0.11 | Yes |
| 2 options | 2 | 50% | +0.33 | Definitely yes |
| 3 options | 1 | 100% | +1.00 | That is not a guess |
Expected value per question = (chance correct × marks) − (chance wrong × penalty).
5. Accuracy vs Attempts: The Real Trade-off
Many candidates believe that attempting more questions always means a higher score. With negative marking, that is simply not true. Here is what 100 attempted questions look like at different accuracy levels, assuming 1 mark per question and a 1/3 penalty.
| Accuracy | Correct | Wrong | Penalty | Net score (out of 100) |
|---|---|---|---|---|
| 100% | 100 | 0 | 0 | 100.00 |
| 90% | 90 | 10 | 3.33 | 86.67 |
| 80% | 80 | 20 | 6.67 | 73.33 |
| 70% | 70 | 30 | 10.00 | 60.00 |
| 60% | 60 | 40 | 13.33 | 46.67 |
| 50% | 50 | 50 | 16.67 | 33.33 |
Attempting all 100 questions at 50% accuracy gives you a lower score than attempting only 70 questions at 90% accuracy (63 marks). Accuracy beats volume.
6. What Exactly Is a Cut Off?
The cut off is the minimum score a candidate needed in order to move to the next stage — the next tier, the interview, or the final merit list. It is not fixed in advance. It emerges from the competition itself.
Four factors decide where the cut off lands:
- Number of vacancies — fewer seats means the cut off moves up, because the same number of candidates is competing for less.
- Difficulty of the paper — a tough paper pulls scores down, and the cut off usually falls with it.
- Number and quality of applicants — more applicants, or a better-prepared pool, pushes the cut off higher.
- Category and reservations — cut offs are declared separately for each category, and reserved categories usually have lower cut offs.
7. How to Estimate the Cut Off
No estimate is perfect, but a structured method gets you far closer than a random guess. Use all three approaches below and see whether they point to a similar range.
Method 1 — Previous-year trend adjustment
Collect three years of data
Note the cut off for your exact exam, your category and your post or state, for the last three cycles. Ignore other categories and other posts — they move differently.
Check for pattern changes
Did the number of questions change? Did the marking scheme change? Did a section get added or removed? If yes, convert old cut offs into percentages of total marks so they are comparable.
Adjust for difficulty
If this year's paper felt noticeably easier than last year's, add 2–5% to the previous cut off percentage. If it felt tougher, subtract 2–5%. Be honest here — most candidates rate their own paper as harder than it was.
Adjust for vacancies
A large increase in vacancies usually lowers the cut off. A sharp reduction raises it. If vacancies changed by more than about 25%, treat your estimate as a wide range rather than a single number.
Build a range, not a point
Finish with something like "the cut off should land between 148 and 158". A range is more useful and far more honest than pretending to know the exact figure.
Method 2 — The percentage-of-total-marks rule
Most objective recruitment exams settle in a fairly predictable band when expressed as a percentage of the total marks. Use this as a sanity check on your trend-based estimate.
| Type of exam | Typical cut off band (general category, % of total) |
Why it sits there |
|---|---|---|
| Tier-1 screening exams | 55% – 70% | Large candidate pool, moderate difficulty |
| Banking prelims | 50% – 65% | Speed-heavy paper, sectional cut offs apply |
| Railway CBT-1 | 60% – 75% | Very high applicant numbers |
| State PSC prelims | 45% – 60% | Tougher papers, wider syllabus |
| Final merit-stage exams | 60% – 80% | Smaller, stronger candidate pool |
Indicative bands only, based on patterns commonly seen in recent years. Actual cut offs vary widely by year, post, state and category — always verify against official data.
Method 3 — The attempt-and-accuracy survey
This works surprisingly well for large exams. In the days after the paper, thousands of candidates post their attempts and expected scores in online groups and forums.
- Note the range of attempts most candidates are reporting for your shift.
- Note the accuracy levels they claim — usually between 70% and 90%.
- Calculate the net score for the middle of that range.
- Remember that people posting online tend to be more prepared than average, so treat the result as the upper end of your estimate.
Skip the manual maths
Use our free Cut Off Calculator to enter your attempts and accuracy and get your score instantly.
8. Normalization in Multi-Shift Exams
When an exam runs across several shifts, each shift gets a different paper. One shift's paper may be easier than another's. Comparing raw scores across shifts would be unfair, so exam bodies apply normalization.
In simple terms, normalization adjusts your score based on how difficult your shift was relative to the others:
- If your shift's paper was harder than average, your normalized score is usually higher than your raw score.
- If your shift's paper was easier than average, your normalized score is usually lower than your raw score.
- The cut off is then applied to the normalized score, not the raw score.
Normalization formulas are published in some exams and kept confidential in others. Either way, you cannot calculate the exact adjustment yourself — so treat your raw score as a reliable estimate with a small margin of uncertainty.
9. Category-Wise Cut Off Relaxations
Cut offs are announced separately for each category, and the differences can be substantial. Here is how the relaxations typically work:
| Category | Typical relaxation vs general | Notes |
|---|---|---|
| General / Unreserved | — | Baseline cut off |
| EWS | Slightly lower | Usually close to the general cut off |
| OBC | Moderately lower | Often 3–8% below general |
| SC | Noticeably lower | Varies widely by exam and state |
| ST | Lowest band | Varies widely by exam and state |
| PwBD | Separate relaxed cut off | Usually declared independently |
Relaxation levels differ significantly between exams and between years. Always check the official notification for your specific exam.
One important detail: in many exams, reserved category candidates must still clear a minimum qualifying score (often around 30–40% of total marks). A relaxed cut off does not mean there is no floor at all.
10. Common Mistakes to Avoid
- Forgetting to multiply the penalty by the question's marks. With 2-mark questions and 1/3 negative marking, the penalty is 0.67 per wrong answer — not 0.33.
- Calculating from the provisional answer key. Revised keys routinely change three to five answers, which can shift your score by several marks.
- Assuming your raw score is final in a multi-shift exam. Normalization can move it up or down.
- Anchoring to a single predicted cut off. Predictions vary wildly across sources. Use a range instead.
- Ignoring sectional cut offs. Many banking and some other exams require a minimum score in each section, even if your overall score is high.
- Overestimating your own accuracy. Count from the answer key, do not estimate from memory.
- Panicking at an estimate. A predicted cut off is not a result. Wait for the official figure before making decisions.
11. Quick Checklist
Run through this every time you calculate a score:
- Is the answer key final, or still provisional?
- How many marks is each question worth?
- What is the exact penalty per wrong answer?
- Have I counted correct, wrong and unattempted separately?
- Did I apply the penalty as marks, not as a number of questions?
- Is this a multi-shift exam where normalization applies?
- Are there sectional cut offs I also need to clear?
- Have I compared my score against a range, not a single predicted number?
12. Frequently Asked Questions
Is negative marking always one-third of the question marks?
No. It varies by exam. Common penalties are one-fourth, one-third or one-half of the marks allotted to a question. Some exams have no negative marking at all. Always check the official information bulletin before you calculate your score.
Can the cut off marks be higher than last year?
Yes. If the paper was easier than last year, if the number of vacancies dropped, or if the number of applicants increased sharply, the cut off can rise. It can also fall for the opposite reasons.
What happens if my score equals the cut off marks?
In most recruitment exams, candidates who score exactly the cut off are included in the qualified list. When several candidates tie on the same score, the tie is usually broken using the exam body's own rules, such as higher marks in a specific section or older age.
Does normalization change my raw score?
Normalization does not change the answers you got right or wrong. It adjusts your raw score up or down based on the difficulty level of the shift you wrote, so that candidates across shifts are compared fairly.
How accurate are cut off predictions?
Predictions are estimates, not guarantees. They are usually most accurate when the exam pattern, vacancies and applicant numbers are close to the previous year. Treat any predicted cut off as a rough range rather than an exact number.
Should I start preparing for the next stage before the cut off is out?
If your calculated score is comfortably above your estimated range, it is sensible to begin preparing for the next stage. Preparation is rarely wasted — the same syllabus and skills carry over to other exams.
Calculating your score is not about predicting the future with certainty. It is about replacing anxiety with information. Once you know your number and a realistic cut off range, you can decide what to do next — start preparing for the next stage, or plan your next attempt properly.