Every entrance-exam season, students and parents mix up percentiles and percentages. The difference is not technical trivia: it changes how you read your result and which colleges you can realistically target.
The percentile formula, with an example
Suppose 1,00,000 candidates wrote your session and 95,000 of them, including you, scored the same as or less than you. Your percentile is 100 × 95,000 ÷ 1,00,000 = 95.
Notice that your marks do not appear in the formula. If the paper was very hard, you might have scored only 45% of the marks and still be at the 95th percentile. If it was easy, 95th percentile might need well over 80%.
Why exams use percentiles at all
Large exams such as JEE Main are held over several days and shifts, and no two papers are exactly equally difficult. Comparing raw marks across sessions would favour whoever got the easier paper. By calculating percentiles within each session and then merging them, NTA puts candidates from different sessions on a common scale. This is what “normalisation” refers to.
CUET-UG works on the same principle for subjects held in more than one shift: NTA has used normalisation so that scores from different shifts can be compared fairly. The exact method for each year is set out in that year’s information bulletin.
Estimating your rank from a percentile
A rough estimate of your overall rank is:
rank ≈ (100 − percentile) × total candidates ÷ 100
With 12,00,000 candidates, a 98 percentile gives about (100 − 98) × 12,00,000 ÷ 100 = 24,000. Treat this as an approximation only: the official rank list also applies tie-breaking rules, and category ranks are calculated separately.
Percentile is not a percentage of marks
- Do not convert a percentile into marks by assuming 99 percentile means 99% of the marks.
- Never enter a percentile where a form asks for your Class 12 percentage, or the other way round.
- Board results remain percentage-based. If an admission form asks for a board percentage, calculate it from your marks, for example with the marks percentage calculator.
Why the seven decimal places matter
With lakhs of candidates, many share almost the same score. Reporting percentiles to seven decimal places keeps ties to a minimum, so 99.9876543 and 99.9876512 can represent quite different positions in a very large field.