NEET Negative Marking: How Many Questions Should You Attempt?

NEET awards four marks for a correct answer, deducts one for an incorrect answer and gives zero for a blank, so any guess with a better than one-in-five chance of being right has a positive expected value. That does not mean you should attempt all 180 questions. Attempt what you are sure of, attempt genuine 50/50s, and decide everything else using hit rates measured in your own mocks, not a number you read online.

The question students ask is "how many should I attempt?". The more useful question is "how accurate am I at each level of confidence?". Two students with the same knowledge can finish 20 to 30 marks apart purely because one answers questions she half-knows and the other answers questions he only thinks he knows.

The marking rule comes from the NEET (UG) 2026 information bulletin published on NTA's NEET site. The probabilities in the table are illustrative, chosen to show the arithmetic. Your own numbers come from your mocks.

Is there negative marking in NEET?

Yes. Under the NEET (UG) 2026 information bulletin, a correct answer earns four marks, an incorrect answer loses one mark, and an unanswered question scores zero, with separate bonus rules for questions that are dropped or have more than one correct option. Check the marking scheme in the bulletin for your own cycle, because the 2027 bulletin has not yet been published.

The paper itself, in the 2026 pattern, is 180 compulsory multiple-choice questions with four options and a single correct answer: 45 Physics, 45 Chemistry and 90 Biology. The optional section that existed in some earlier years was removed from NEET 2025 onwards, so every question you leave blank is a question you chose not to answer, not one you were allowed to skip.

What is the expected value of a guess in NEET?

The expected value of a guess is the chance of being right multiplied by four, minus the chance of being wrong multiplied by one. Here is the arithmetic for each case:

• Two options left (50/50): 0.5 × 4 − 0.5 × 1 = 2 − 0.5 = +1.5 marks.
• Three options left: (1/3) × 4 − (2/3) × 1 = 1.33 − 0.67 = +0.67 marks.
• Blind guess, four options: 0.25 × 4 − 0.75 × 1 = 1 − 0.75 = +0.25 marks.
• Break-even: 4p − (1 − p) = 0, so p = 0.2. Any guess with better than a one-in-five chance is positive on average.

So on paper, even a blind guess pays. That is where most online advice stops, and it is why some students attempt every question. The next section explains why the real picture is less generous.

Why can guessing still cost you marks?

Because your real probabilities are not the textbook ones, the results swing widely, and wrong answers have a second cost in the tie-break.

Your odds are often worse than they feel. A blind guess is only a 25 per cent chance if all four options are equally likely to attract you. NEET options are not written that way. Wrong options are built around common misconceptions, so the option that "looks right" when you do not know the topic is often the trap. If your real hit rate on such questions is 15 per cent, the arithmetic becomes 0.15 × 4 − 0.85 × 1 = −0.25 marks per question. Ten of those cost you about two and a half marks on average.

The swing is large. Take 20 truly random guesses at exactly 25 per cent. The average gain is +5 marks. But four correct out of 20 only breaks even (4 × 4 − 16 = 0), you need five to come out ahead (5 × 4 − 15 = +5), and the binomial arithmetic gives roughly a two-in-five chance of gaining nothing or losing marks on those 20 questions. You are taking on a lot of uncertainty for a small average gain.

Wrong answers count in the tie-break. The official order is covered below. It only matters when totals are tied, but it is a real cost that a blank does not carry.

Deliberating costs time. The guess itself is quick. The two minutes spent agonising over a question you cannot solve are not, and they come out of questions you could have got right.

How many questions should I attempt in NEET?

There is no correct number of attempts, only a confidence threshold. Attempt every question you are sure of, plus those where your mock data shows you are right well above one time in five. A student who attempts 158 questions with 22 wrong scores 522, which beats one who attempts 172 with 34 wrong at 518. Accuracy decides the score, not the attempt count.

The arithmetic behind that example:

Student A answered 14 more questions and got two more right, but the 12 extra wrong answers cancelled those gains and more. Each wrong answer sits five marks below a correct one and one mark below a blank. This is why attempt targets such as "always attempt 170" are a poor rule. They push you to answer the questions where your accuracy is lowest.

How do I set my own attempt rule from mock tests?

Tag every answer with your confidence level during the paper, then calculate your hit rate for each tag after several mocks. Four papers give you enough data to start.

Use three simple marks next to each question on the question paper or rough sheet: S for sure, 2 for two options left, 3 for three options left, and G for a guess. After the paper, check the key and count. An illustrative result across four full mocks, 720 questions in total:

What this student should do:

1. Keep answering "2" questions. A 53 per cent hit rate is close to the theory, worth +1.67 marks each.
2. Keep "3" questions under watch. At 29 per cent they are still positive, but only just. If the rate drops towards 20 per cent over the next few mocks, stop.
3. Stop answering "G" questions. Across four papers they lost six marks.
4. Look hardest at the 32 wrong "sure" answers. This is the biggest leak by far: 32 × 5 = 160 marks lost against getting them right, compared with six marks from guessing. Overconfidence costs far more than guessing does, and it is fixed with accuracy work, not an attempt rule.

Log these alongside your normal error categories. The structure in how to analyse mock tests with an error log takes an extra column for the confidence tag without any other change, and the NEET mock test schedule and review protocol shows where this fits in a 90-minute review.

How can I reduce silly mistakes in NEET?

Name them first, because silly mistake is not a repairable category. Most fall into a few types: missing the word incorrect or not, misjudging one statement in a multi-statement question, unit or sign slips in Physics, and OMR bubbling errors. Give each type a fixed physical check, such as underlining the instruction word, and track whether the count falls over three or four mocks.

Physics carries the most calculation-type errors for most students. If yours are concentrated there, the NEET Physics improvement plan separates concept gaps from calculation slips and gives each its own drill.

Does the number of wrong answers matter in a NEET tie?

Yes, though only after marks. The 2026 bulletin breaks ties first by Biology marks, then Chemistry, then Physics, and then by the lower ratio of incorrect to correct answers, overall and then subject by subject, before a random process. Two candidates on the same total are separated partly by how many wrong answers each gave, so careless attempts carry a small extra cost.

Two practical points follow. First, Biology marks come first in the tie-break, which is one more reason Biology accuracy deserves protection, alongside the fact that it carries half the paper. Second, the ratio rule means a pile of low-confidence attempts is not neutral even when the marks happen to balance out. It is a minor effect next to the marks themselves, but it points the same way as everything else in this guide.

How should I split my time in the exam hall?

Use passes rather than working through strictly in order, so that time goes to questions you can actually convert. NEET 2026 was set at 180 minutes for 180 questions, an average of one minute each, though the 21 June 2026 re-exam window was extended to 195 minutes to cover formalities.

An example time plan, to adjust from your own mocks:

Some students prefer to start with Physics while they are fresh. Either order works if it is rehearsed. What does not work is changing the order on the day. If panic hits during a pass, a short reset works better than pushing on; the routine in what to do when your mind goes blank in an exam fits inside a NEET paper, and the time management guide covers per-question budgets in more detail.

How do I build accuracy work into a study plan?

Treat accuracy as scheduled work, not as a resolution made in the exam hall. Three recurring tasks are enough: the confidence-tag count after each mock, one short set each week aimed at your most common mistake type, and a monthly recalculation of your attempt rule.

In MyStudyPlanner you can set this up inside your NEET plan. Start from the NEET UG template (Physics, Chemistry, Botany and Zoology), add your test-series papers as mock sets in the Mock Tests step, and add custom tasks for each confidence-tag review and the weekly mistake drill. MyStudyPlanner does not provide NEET questions or score your papers, so the hit-rate arithmetic stays in your error log. The free mock-test error log template works for it.

What if my accuracy drops in a mock?

Do not rewrite your attempt rule after one paper. Hit rates on 15 or 20 tagged questions in a single mock swing a lot by chance, so judge the rule on the last three or four papers together.

If accuracy has dropped across several papers, check three things in order:

1. Has the "sure" accuracy fallen? That usually means revision has lapsed in a subject, not that your guessing policy is wrong. Go back to recall work in that subject.
2. Has the number of "3" and "G" attempts crept up? Time pressure often pushes students into answering more low-confidence questions. Tighten the rule for two papers and compare.
3. Is one error type rising? A sudden rise in misreads often comes with fatigue. Check sleep and the time of day you sat the paper before blaming knowledge.

Frequently Asked Questions

Is there negative marking in NEET?

Yes. Under the NEET (UG) 2026 information bulletin, a correct answer earns four marks, an incorrect answer loses one mark, and an unanswered question scores zero, with separate bonus rules for questions that are dropped or have more than one correct option. Check the marking scheme in the bulletin for your own cycle, because the 2027 bulletin has not yet been published.

Should I guess in NEET if I can eliminate two options?

Usually yes. With two options left and a genuine 50 per cent chance, the expected value is 0.5 × 4 minus 0.5 × 1, which is plus 1.5 marks per question. The catch is the word genuine: if one remaining option is a designed trap, your real odds may be lower. Check your own hit rate on these questions across several mocks before trusting the rule.

How many questions should I attempt in NEET?

There is no correct number of attempts, only a confidence threshold. Attempt every question you are sure of, plus those where your mock data shows you are right well above one time in five. A student who attempts 158 questions with 22 wrong scores 522, which beats one who attempts 172 with 34 wrong at 518. Accuracy decides the score, not the attempt count.

Is blind guessing worth it in NEET?

Mathematically, a truly random guess among four options has a small positive expected value of plus 0.25 marks. In practice it is rarely worth relying on. Your real hit rate on blind guesses may be below 25 per cent because wrong options are written to attract, the results swing widely across a paper, and incorrect answers count against you in the official tie-break.

How can I reduce silly mistakes in NEET?

Name them first, because silly mistake is not a repairable category. Most fall into a few types: missing the word incorrect or not, misjudging one statement in a multi-statement question, unit or sign slips in Physics, and OMR bubbling errors. Give each type a fixed physical check, such as underlining the instruction word, and track whether the count falls over three or four mocks.

Does the number of wrong answers matter in a NEET tie?

Yes, though only after marks. The 2026 bulletin breaks ties first by Biology marks, then Chemistry, then Physics, and then by the lower ratio of incorrect to correct answers, overall and then subject by subject, before a random process. Two candidates on the same total are separated partly by how many wrong answers each gave, so careless attempts carry a small extra cost.

Your next step

In your next full mock, tag every answer S, 2, 3 or G as you go. Afterwards, spend 20 minutes building the hit-rate table for that paper alone, then repeat for the next three mocks. When you have four papers' worth, write your attempt rule in one sentence, for example "S and 2 always, 3 only in Biology, never G", and keep it where you will see it before every paper.