JEE Main percentile and normalisation, explained
How NTA turns raw marks from different shifts into one fair ranking, why 99 percentile is not 99% of the marks, and how ties are broken.
Everything on this page comes from Appendix VII and the result sections of NTA’s JEE (Main) 2026 Information Bulletin.
Why does JEE Main need normalisation?
JEE Main Paper 1 runs over many shifts. In 2026, Session 1 had 10 shifts and Session 2 had 9. Each shift gets a different paper, and no two papers are exactly equally hard.
If NTA ranked everyone on raw marks, a candidate who drew a harder paper would be at a disadvantage. So NTA converts raw marks into NTA scores, which are percentiles worked out within each shift. You are compared only with the people who sat the same paper as you.
How is the JEE Main percentile calculated?
NTA’s formula is:
Percentile = 100 × (number of candidates in your shift with a raw score equal to or less than yours) ÷ (total number of candidates who appeared in your shift)
A few rules follow from it:
- The topper of every shift gets 100, because everyone in that shift scored equal to or less than them.
- Percentiles are calculated to seven decimal places, to reduce ties. That is why cutoffs look like 93.4123549.
- NTA calculates a percentile for your total and for each subject.
- Your total percentile is not the average of your three subject percentiles. It is calculated from your total raw score.
A worked illustration
Illustration only. These numbers are made up and rounded to keep the arithmetic simple. They are not real JEE Main data.
Suppose 1,00,000 candidates appear in a shift, and you score 150 out of 300.
- Say 92,000 candidates in that shift, including you, scored 150 or less.
- Your percentile = 100 × 92,000 ÷ 1,00,000 = 92.
Now suppose a friend also scores 150, but in a different shift where the paper was harder. In that shift, 96,000 of 1,00,000 candidates scored 150 or less.
- Your friend’s percentile = 100 × 96,000 ÷ 1,00,000 = 96.
Same raw marks, different percentiles. Your friend’s 150 was a better performance relative to the paper they sat, and the percentile reflects that.
Why is percentile not percentage?
In the illustration, 150 out of 300 is 50% of the marks. The percentile was 92 in one shift and 96 in the other.
| What it measures | Illustration value | |
|---|---|---|
| Percentage | Your marks as a share of 300 | 50% |
| Percentile (NTA score) | The share of candidates in your shift who scored equal to or less than you | 92 or 96 |
So “99 percentile” does not mean 99% of the marks. It means you scored at least as well as 99% of the candidates in your shift. NTA’s bulletin is explicit that percentile is not the percentage of marks.
How are both sessions combined?
After Session 2, NTA merges the NTA scores of all shifts from both sessions into one merit list.
If you sat both sessions, NTA uses your better total NTA score. It does not mix your best Mathematics score from January with your best Physics score from April.
The All India Rank is released after Session 2. There is no re-evaluation or re-checking.
How are ties broken in JEE Main?
Seven decimal places make ties less likely, but NTA still sets out an order for resolving a tie in total NTA score:
- Higher NTA score in Mathematics
- Higher NTA score in Physics
- Higher NTA score in Chemistry
- Lower ratio of incorrect to correct answers across all subjects
- Lower incorrect-to-correct ratio in Mathematics
- Lower incorrect-to-correct ratio in Physics
- Lower incorrect-to-correct ratio in Chemistry
- If still tied, the candidates share the same rank
The 2026 rules use no age or application-number criterion.
Steps 4 to 7 reward accuracy. Two candidates on the same score are separated by who made fewer wrong attempts, which is one more reason to avoid blind guessing under the +4 / −1 marking scheme.
What this means for you
You cannot control how hard your shift’s paper is, or how others in your shift perform. You can control your raw score. The 2026 cutoffs show the percentile scale, and a steady practice routine is how you move your raw marks.
After each TestGuru mock test, the report names the idea behind each wrong answer and builds a practice set from your own mistakes, so you can see where raw marks are being lost.