Tools
Score and negative marking calculator
Two questions this answers. What did I actually score, once wrong answers are deducted? And to reach the score I want, how many questions do I need to attempt at the accuracy I currently have?
100 questions, 100 marks, 60 minutes with sectional timing. A quarter mark is deducted for a wrong answer.
Score a completed paper
Count your correct and wrong answers from the official answer key, and this applies the marking scheme.
- 57
- Net score
- 100
- Maximum
- 83%
- Accuracy
- −3
- Lost to penalty
72 attempted, 28 left blank. The 3 marks lost to the penalty is the cost of 12 wrong answers, which is worth comparing against what those attempts gained you.
Plan a target
Work the other way round: decide how many you intend to attempt and how accurate you are in practice, and see where that lands.
61.25out of 100
That is 64 correct and 11 wrong, costing 2.75 marks in penalties.
Try moving accuracy down by five points and attempts up by ten. In most configurations the score falls, which is the whole argument for selection over volume.
Is a guess worth taking?
Expected marks from attempting one question, after ruling out options. Positive means attempting is right on average.
| Options you are choosing between | Expected marks | Verdict |
|---|---|---|
| 4, a blind guess | +0.06 | Attempt |
| 3, after eliminating 1 | +0.17 | Attempt |
| 2, after eliminating 2 | +0.38 | Attempt |
These are averages over many questions, and they assume your guess is genuinely random. In practice distractors are written to attract plausible wrong reasoning, so treat a blind guess as worse than the table suggests. The real cost of a hard question is usually the minutes, not the fraction of a mark.
Why the second question is the useful one
Most candidates track their score and nothing else. The score is a result, and by the time you see it the decisions that produced it are already made. The two numbers you can actually control during a paper are how many questions you attempt and how carefully you pick them, and those two combine into the score through fixed arithmetic.
Run the planner and a pattern shows up quickly: past a certain point, extra attempts stop helping. Someone attempting 90 of 100 questions at 70 per cent accuracy scores 58.25 in a banking paper. The same person attempting 70 questions at 85 per cent scores 56.5, and does it with twenty questions worth of time returned to the ones they were actually sure about. Accuracy is a cheaper lever than volume for nearly everybody, and it is the one almost nobody pulls.
The two marking schemes, side by side
| Exam | Penalty | Wrong answers per mark lost | Blind guess on 4 options |
|---|---|---|---|
| IBPS PO, prelims and mains | 1/4 mark | 4 | +0.0625 marks |
| RRB NTPC, CBT 1 and CBT 2 | 1/3 mark | 3 | 0 marks, exactly |
That last column is worth sitting with. In the railways, a blind guess across four options is mathematically pointless: over many questions it returns precisely nothing, while consuming time you could spend on a question you can solve. In banking it is very slightly positive, which is not the same as being a good idea, because the marks gained are dwarfed by what the time costs elsewhere in the paper.
The picture changes completely once you eliminate options. Ruling out two of four turns a break-even guess into an expected gain of a third of a mark in the railways and 0.375 in banking. This is why the instruction is not "never guess" but "never guess blind".
Frequently asked questions
How much is deducted for a wrong answer in IBPS PO?
One quarter of the marks assigned to that question. In prelims, where every question carries one mark, four wrong answers cost one mark. In mains the marks per question vary by section, so the deduction scales with them: a question worth 2 marks costs 0.5 for a wrong answer.
How much is deducted in RRB NTPC?
One third of a mark per wrong answer, in both CBT 1 and CBT 2. Three wrong answers cost one mark, which is a heavier penalty than banking exams apply.
Is there any penalty for a question left blank?
No. An unattempted question scores zero in both exams. This is why leaving a question is always better than a genuinely blind guess, and why the arithmetic of eliminating options matters.
Should I guess when I have eliminated two of four options?
Yes, in both exams. With two options remaining you have a fifty per cent chance of a mark against a half chance of losing a quarter or a third of one, so the expected value is clearly positive. With no options eliminated a blind guess is exactly break-even in the railways and marginally positive in banking, which in practice means it is not worth the time it costs.
Does this calculator predict my cut-off?
No, and nothing can. A cut-off is decided after the result by how many candidates a board needs to shortlist, so it is an outcome rather than something computable in advance. This tool only converts your attempts into the marks the published scheme produces.
Related reading
- Negative marking strategy, in full— the habits that produce the accuracy figure you just typed in.
- Percentile calculator— what that score looks like against other candidates.
- Free mock tests— where the scoring happens automatically, section by section.