CAT 2026 · free tool
CAT percentile predictor
Enter a sectional score and this returns a percentile band, drawn from 45 results logged across 15 real CAT mock papers — not a single confident number, because a single number here is not honest. In this log a DILR score of 15 returned the 65th, 76th, 90th percentile on three different papers, and a VARC 39 scored lower than a VARC 33. Below that, every figure 8 institutes publish for the last four CAT years, side by side — they disagree by up to 30.8% on a result three years old. Every row is sourced, so you can check the working.
Estimate from your sectional scores
midpoint 58.3, from 4 logged results
midpoint 73.4, from 7 logged results
midpoint 87.4, from 8 logged results
Total raw score 60 out of 204. Sectional percentiles do not add up to an overall percentile — the overall is computed on the overall score against the same cohort, which is why the reference table below keeps them apart.
CAT marks to percentile, section by section
The lookup nobody else publishes — most tables run percentile to marks; this runs marks to percentile, because that is how people actually have the number in hand after a mock. Same 15-paper log as the tool above, same band, no fitted curve.
| Marks | Percentile band |
|---|---|
| 4 | 12–37 |
| 8 | 15–37 |
| 12 | 28–61 |
| 16 | 50–61 |
| 20 | 50–71 |
| 24 | 57–86 |
| 28 | 82–89 |
| 32 | 82–92 |
| 34 | 78–92 |
| 36 | 78–92 |
| 40 | 78–92(extrapolated) |
| 44 | 78–92(extrapolated) |
| 46 | 78–92(extrapolated) |
| 48 | 78–92(extrapolated) |
| 52 | 78–92(extrapolated) |
| 56 | 78–92(extrapolated) |
| 60 | 78–92(extrapolated) |
| 64 | 78–92(extrapolated) |
| 68 | 78–92(extrapolated) |
| 72 | 78–92(extrapolated) |
| Marks | Percentile band |
|---|---|
| 4 | 52–68(extrapolated) |
| 8 | 52–68 |
| 12 | 58–90 |
| 16 | 58–90 |
| 20 | 75–87 |
| 24 | 75–85 |
| 28 | 82–93 |
| 32 | 82–93(extrapolated) |
| 34 | 82–93(extrapolated) |
| 36 | 82–93(extrapolated) |
| 40 | 82–93(extrapolated) |
| 44 | 82–93(extrapolated) |
| 46 | 82–93(extrapolated) |
| 48 | 82–93(extrapolated) |
| 52 | 82–93(extrapolated) |
| 56 | 82–93(extrapolated) |
| 60 | 82–93(extrapolated) |
| 64 | 82–93(extrapolated) |
| Marks | Percentile band |
|---|---|
| 4 | 55–83(extrapolated) |
| 8 | 55–83(extrapolated) |
| 12 | 55–83(extrapolated) |
| 16 | 55–83 |
| 20 | 75–87 |
| 24 | 80–91 |
| 28 | 74–91 |
| 32 | 74–97 |
| 34 | 74–97 |
| 36 | 74–97 |
| 40 | 74–97 |
| 44 | 74–97 |
| 46 | 74–97 |
| 48 | 74–97 |
| 52 | 74–97(extrapolated) |
| 56 | 74–97(extrapolated) |
| 60 | 74–97(extrapolated) |
| 64 | 74–97(extrapolated) |
Bands, not single numbers, for the same reason as the tool above: this log's own DILR 15 returned three different percentiles on three different papers. Mock percentiles, not official CAT percentiles.
CAT 2025 score vs percentile
Different data from everything else on this page. These are scaled scores for the actual CAT 2025 paper, compiled from candidate-reported scorecards — not the mock log above, and not official, because the IIMs publish no such mapping to anyone. It is the closest available reference point for CAT 2026, not a cutoff.
| Percentile | Overall | VARC | DILR | Quant |
|---|---|---|---|---|
| 99.9 | — | 53 | 38 | 37 |
| 99 | 84.8 | 44 | 29.8 | 27.3 |
| 95 | 62.3 | 32.5 | 21.5 | 18.5 |
| 90 | 51.5 | 26 | 16.7 | 15 |
The same score, different percentiles
5 scores in this log returned more than one percentile, and 59 pairs are outright inversions — a higher score scoring a lower percentile than a lower one in the same section. This is why the tool above reports a band.
| Section | Score | Percentiles it returned | Spread | Papers |
|---|---|---|---|---|
| DILR | 15 | 65, 76, 90 | 25 | AIMCAT SA2701, SIMCAT 7, SIMCAT 103 |
| DILR | 30 | 82, 93 | 11 | SIMCAT 2, SIMCAT 4 |
| Quant | 27 | 83, 90 | 7 | SIMCAT 1, SIMCAT 105 |
| Quant | 25 | 87, 91 | 4 | SIMCAT 104, SIMCAT 4 |
| Quant | 21 | 80, 83 | 3 | SIMCAT 2, SIMCAT 3 |
CAT score vs percentile, 2021–2025
Every published figure below is somebody else’s, collected from 8 institutes on 10 September 2026 and shown with the source on the row. They are here to be compared, not averaged — because they do not agree, and the disagreement is the useful part.
| Year | %ile | Published range | Spread | Sources | Who says what |
|---|---|---|---|---|---|
| CAT 2025 | 99 | 79–85 | 6 (7.6%) | 6 | Career Launcher 79 · Arun Sharma 82 · 2IIM / Rajesh Balasubramanian 84 · Cracku 84.8 · iQuanta 84.8 · IMS 85 |
| CAT 2025 | 95 | 54–63 | 9 (16.7%) | 5 | Career Launcher 54 · IMS 62 · Cracku 62.3 · iQuanta 62.3 · 2IIM 63 |
| CAT 2025 | 90 | 44–52 | 8 (18.2%) | 5 | Career Launcher 44 · IMS 51 · Cracku 51.5 · iQuanta 51.5 · 2IIM 52 |
| CAT 2024 | 99 | 95–103 | 8 (8.4%) | 5 | MBAUniverse 95 · iQuanta 95 · Cracku 95.13 · IMS 96 · Career Launcher 103 |
| CAT 2024 | 95 | 58–70 | 12 (20.7%) | 5 | Career Launcher 58 · Cracku 70 · IMS 70 · MBAUniverse 70 · iQuanta 70 |
| CAT 2024 | 90 | 47–60 | 13 (27.7%) | 5 | Career Launcher 47 · IMS 57 · Cracku 58 · iQuanta 58 · MBAUniverse 60 |
| CAT 2023 | 99 | 76.02–77 | 0.98 (1.3%) | 3 | iQuanta 76.02 · IMS 77 · MBAUniverse 77 |
| CAT 2023 | 95 | 54.86–55 | 0.14 (0.3%) | 3 | iQuanta 54.86 · IMS 55 · MBAUniverse 55 |
| CAT 2023 | 90 | 44–45.16 | 1.16 (2.6%) | 3 | MBAUniverse 44 · IMS 45 · iQuanta 45.16 |
| CAT 2022 | 99 | 84–109.9 | 25.9 (30.8%) | 4 | Cracku 84 · MBAUniverse 84 · iQuanta 84 · Career Launcher 109.9 |
| CAT 2022 | 95 | 59.75–61 | 1.25 (2.1%) | 4 | Career Launcher 59.75 · Cracku 60 · iQuanta 60 · MBAUniverse 61 |
| CAT 2022 | 90 | 48.44–49 | 0.56 (1.2%) | 4 | Career Launcher 48.44 · Cracku 49 · MBAUniverse 49 · iQuanta 49 |
| CAT 2022 | 85 | 41.32 | — | 2 | Cracku 41.32 · Career Launcher 41.32 |
| CAT 2022 | 80 | 36.02 | — | 2 | Cracku 36.02 · Career Launcher 36.02 |
Slot-specific figures are excluded from the spread: a slot difference is a real difference in the paper, not two sources contradicting each other.
Over half the tables online do not say what they are measuring
Of the 196 published figures collected here, 160 (81.6%) do not state whether they are raw marks or scaled scores. CAT is 204 raw marks; a scaled score is what the IIMs publish after normalising across slots. They are different quantities, and comparing one against the other is most of why “how many marks for 99 percentile” has no stable answer.
Some of these sources are copying each other
41.32 and 36.02 and 84.8 and 62.3 and 51.5 appear identically, to two decimal places, in both Cracku’s and Career Launcher’s CAT 2022 tables. Two independent surveys of different candidates do not produce 41.32 twice.
Which matters, because where Cracku labels 110 as the 99.9th percentile, Career Launcher labels 109.9 as the 99th — the same number, one row out. That single mislabel is why CAT 2022’s 99th percentile is published as anything between 84 and 109.9.
All 45 observations
Published in full so the estimate above can be checked rather than trusted. None of the twelve competing pages surveyed shows its rows.
| Score | %ile | Paper |
|---|---|---|
| 3 | 12 | PreSimCAT 03 |
| 5 | 15 | SIMCAT 104 |
| 9 | 37 | SIMCAT 7 |
| 10 | 28 | SIMCAT 2 |
| 12 | 42 | SIMCAT 3 |
| 15 | 61 | SIMCAT 103 |
| 17 | 50 | SIMCAT 102 |
| 18 | 55 | SIMCAT 101 |
| 20 | 71 | SIMCAT 5 |
| 21 | 57 | PreSimCAT 02 |
| 25 | 86 | AIMCAT SA2701 |
| 27 | 82 | SIMCAT 1 |
| 33 | 89 | SIMCAT 6 |
| 36 | 92 | SIMCAT 4 |
| 39 | 78 | SIMCAT 105 |
| Score | %ile | Paper |
|---|---|---|
| 8 | 52 | PreSimCAT 03 |
| 11 | 68 | SIMCAT 3 |
| 13 | 58 | SIMCAT 6 |
| 14 | 59 | SIMCAT 105 |
| 15 | 65 | AIMCAT SA2701 |
| 15 | 76 | SIMCAT 7 |
| 15 | 90 | SIMCAT 103 |
| 16 | 79 | SIMCAT 5 |
| 17 | 87 | SIMCAT 104 |
| 19 | 82 | SIMCAT 101 |
| 20 | 85 | SIMCAT 1 |
| 23 | 75 | SIMCAT 102 |
| 25 | 82 | PreSimCAT 02 |
| 30 | 82 | SIMCAT 2 |
| 30 | 93 | SIMCAT 4 |
| Score | %ile | Paper |
|---|---|---|
| 13 | 55 | SIMCAT 5 |
| 19 | 75 | SIMCAT 6 |
| 21 | 80 | SIMCAT 2 |
| 21 | 83 | SIMCAT 3 |
| 23 | 87 | SIMCAT 102 |
| 24 | 91 | SIMCAT 103 |
| 25 | 87 | SIMCAT 104 |
| 25 | 91 | SIMCAT 4 |
| 27 | 83 | SIMCAT 1 |
| 27 | 90 | SIMCAT 105 |
| 28 | 85 | AIMCAT SA2701 |
| 28 | 85 | SIMCAT 101 |
| 30 | 74 | PreSimCAT 03 |
| 36 | 97 | SIMCAT 7 |
| 50 | 97 | PreSimCAT 02 |
How the estimate works
For a given score it takes every logged observation within a window either side of it, widening the window until at least three are inside, and returns the lowest, highest and mean percentile of that group. It deliberately does not fit a curve: a regression through fifteen points would produce one confident-looking number and hide the spread, which is the only thing here worth knowing.
Where this is weak
- Fifteen papers is a small sample, and one candidate. A coaching predictor fitted to tens of thousands of attempts has a real advantage this does not.
- These are mock percentiles. The cohort sitting a mock is self-selected and far smaller than CAT’s.
- Scores outside the logged range are extrapolations, and the tool says so when that happens rather than quietly returning a number.
- It cannot model normalisation across slots, which is exactly where the published disagreement comes from.
CAT percentile: common questions
- How accurate is a CAT percentile predictor?
- Less accurate than any of them admit. There is no official raw-score-to-percentile mapping for CAT - the IIMs publish a scaled score and a percentile to each candidate and nothing else - so every predictor is working from crowdsourced scorecards or a coaching institute model. Published figures for 99th percentile in CAT 2025 range from 79 to 112 depending on whose table you read. Treat any single number as one estimate among many.
- What is 46 marks in CAT percentile?
- Depends on the section - CAT doesn't score sections together, so 46 marks means different things in each. From this mock log: VARC 46 (extrapolated) → 78th–92th percentile; DILR 46 (extrapolated) → 82th–93th percentile; Quant 46 → 74th–97th percentile. These are mock percentiles from 15 logged papers, not official CAT percentiles, and the range is the finding, not a rounding error - see the marks-to-percentile table below for every value.
- What is 34 marks in CAT percentile?
- Depends on the section - CAT doesn't score sections together, so 34 marks means different things in each. From this mock log: VARC 34 → 78th–92th percentile; DILR 34 (extrapolated) → 82th–93th percentile; Quant 34 → 74th–97th percentile. These are mock percentiles from 15 logged papers, not official CAT percentiles, and the range is the finding, not a rounding error - see the marks-to-percentile table below for every value.
- Why does the same score give different percentiles?
- Because percentile is computed against the people who sat that specific paper, not against a fixed scale. In this log a DILR score of 15 returned the 65th, 76th and 90th percentile on three different mocks - same candidate, same three-month window, same raw score. The cohort changed, so the percentile changed.
- What is the difference between raw score and scaled score in CAT?
- Raw score is marks out of 204 (+3 correct, -1 wrong on MCQs, nothing on TITA). Scaled score is what the IIMs publish after normalising across slots, so that candidates sitting different papers are comparable. Most tables online do not say which one they are using, and that single ambiguity explains most of the disagreement between them.
- How many marks do I need for 99 percentile in CAT?
- On CAT 2025, a scaled score of roughly 85 overall crossed the 99th percentile - about 44 in VARC, 30 in DILR and 27 in Quant. That figure is compiled from reported scorecards rather than issued officially, and published estimates elsewhere run from 79 to 112, partly because some quote raw marks out of 204 and some quote scaled scores. Use a band, not a target number.
- Are mock percentiles the same as actual CAT percentiles?
- No, and this page never claims they are. A mock percentile is computed against the self-selected group that sat that mock - a much smaller and differently-composed cohort than the roughly 2.5 lakh who sit CAT. It is a useful relative signal for tracking your own progress and a poor absolute prediction of your CAT result.
- Can a higher score give a lower percentile?
- Yes, across different papers. In this log a VARC score of 39 returned the 78th percentile while 36 returned the 92nd and 33 returned the 89th - all in the same section, all by the same candidate. That is the clearest possible demonstration that a single conversion table cannot be right.
- What is the CAT exam pattern?
- 68 questions for 204 marks in 120 minutes, stable since 2024: VARC 24 questions, DILR 22 and Quant 22, each worth 3 marks. Sections are time-locked at 40 minutes each and you cannot return to a completed one. Marking is +3 correct and -1 wrong on MCQs, with no negative marking on TITA questions.
- Where does the data on this page come from?
- Forty-five sectional results logged personally across fifteen IMS SIMCAT, PreSimCAT and TIME AIMCAT papers between May and August 2026. Every row is shown on the page with the paper it came from, so you can check the working rather than trust the output. Nothing here is scraped or modelled from someone else table.
- Should I use this instead of a coaching percentile predictor?
- Use it alongside one. A coaching predictor is fitted to a much larger cohort than fifteen papers, which is a genuine advantage. What it will not do is show you the spread or tell you which rows produced its number, and that is what this page adds.
- What was the CAT score vs percentile in previous years?
- It depends entirely on who you ask, which is the finding. For CAT 2022 at the 99th percentile, Career Launcher publishes 109.9 while Cracku and MBAUniverse publish 84 - a 31% disagreement about a result three years old. The full year-by-year table on this page shows every institute figure side by side rather than picking one.
- Why do different websites give different CAT percentile tables?
- Three reasons. Over half the published tables do not state whether they mean raw marks or scaled scores, and those are different quantities. Some are institute predictions made before results, revised later without the old figure being removed. And at least two sources here carry values identical to two decimal places, so they are copying rather than measuring independently.
- Which CAT percentile predictor should I trust?
- None of them alone. Career Launcher was measurably the most generous predictor for CAT 2025 and Cracku the most conservative, with several marks between them at the same percentile. Take a range across sources rather than a number from one, and check whether the table you are reading says raw or scaled before you compare it to your own score.
- Is this CAT percentile predictor free?
- Yes, with no sign-in and no course attached. Karma Yogi is a free study tracker; this tool exists because the same question kept coming up in its own mock logs.
Track your own score-to-percentile spread
The data on this page is one person’s mock log. Karma Yogi records every mock at section level with its score and percentile, so after a dozen papers you have your own version of this table — which is worth more than anyone else’s. Free, no card.