Exams such as JEE Main and CAT publish percentiles rather than ranks. A percentile tells you what share of candidates scored at or below you, so it can be converted into an approximate rank once you know how many people sat the exam.
Enter your percentile.
Enter the number of candidates who appeared.
Read your estimated rank.
Estimated rank: Rank ≈ (100 − percentile) ÷ 100 × candidates + 1
Several national exams, including JEE Main and CAT, report a percentile rather than a rank or raw score. A percentile tells you what share of candidates scored at or below you. This calculator converts a percentile and the number of candidates into an estimated rank.
Students need an approximate rank to shortlist colleges from previous years' opening and closing ranks in counselling, and to judge their chances before official ranks are released. Parents and counsellors use it the same way.
The result is an estimate for planning. Official ranks apply tie-breaking rules, and category ranks are separate, so they can differ from this number.
1. Note your percentile P and the total number of candidates N who appeared.
2. Find the share of candidates ahead of you: (100 − P) ÷ 100.
3. Multiply by N to get the number of candidates ahead.
4. Take the whole-number part of that figure and add 1 to get your estimated rank: rank ≈ floor((100 − P) ÷ 100 × N) + 1.
5. For a category rank, repeat with the number of candidates in that category only if their percentile distribution is known; otherwise use published tables.
The National Testing Agency computes a candidate's percentile as 100 × (number of candidates in the session with raw score equal to or less than the candidate) ÷ (total candidates in that session). The topper of each session is therefore at 100. Percentiles are calculated session by session and then merged, which is the normalisation step for multi-shift exams. Because the definition counts people 'at or below', the share strictly above you is 100 − P percent.
When lakhs of candidates appear, each 0.01 percentile represents a large group. With 12 lakh candidates, the gap between 99.00 and 99.01 percentile is about 120 people. So near the top, differences in the second decimal place of a percentile matter for college choices. This is also why percentiles are reported to seven decimal places. At lower percentiles the density of candidates per mark is even greater, so rank moves quickly with small score changes.
Candidates with the same raw score get the same percentile, but official ranks break ties using rules such as subject-wise scores, age or fewer wrong answers. The estimate here places you just after everyone strictly ahead, so it is the best rank you could get at that percentile. Official figures can be slightly worse. Also note that the percentile is not a percentage of marks: 90 percentile can correspond to quite modest marks in a difficult paper.
Karthik scored 93.42 percentile in JEE Main, and he read that about 12,48,000 candidates appeared. He wants a rough all-India rank to compare with last year's closing ranks.
Share of candidates ahead of you: 100 − 93.42 = 6.58%
Candidates ahead: 6.58% × 12,48,000 = 82,118.4
Estimated rank: floor(82,118.4) + 1 = 82,119
Answer: Estimated rank 82,119; Candidates ahead of you 82,118
Confusing percentile with percentage of marks.
Using the number of registered candidates instead of those who actually appeared.
Treating the estimate as a category rank; category ranks are much smaller numbers.
Rounding the percentile to one decimal, which can shift the rank by thousands.
Shortlisting colleges before counselling using previous years' closing ranks.
Converting a CAT or JEE percentile into an approximate position.
Comparing performance across sessions of a normalised exam.
Counsellors advising students on realistic options.
Is this my official rank?
No. The official merit list applies tie-breaking rules and category ranks. Use this as a planning estimate.
Where do I find the number of candidates?
NTA and exam bodies publish the number of candidates who appeared along with the results.