The tool uses question-level data from 18 CAT slots across 2020–2025 (3 slots per year × 6 years). For each slot, we recorded: chapter name, topic, question count, and difficulty label (easy / easy-moderate / moderate / moderate-difficult / difficult). Marks-to-percentile lookup tables come from Tarkashastra for 2020–2022 and Cracku for 2023–2025. We are not affiliated with either source.
For each historical slot, we simulate 400 independent exam attempts. For every question in the slot, we draw against an attempt-probability (does the candidate try this question?) and then, if attempted, against an accuracy-probability (does the candidate get it right?). Both probabilities are set per (preparation rating × difficulty) pair. Marks are tallied with standard CAT scoring: +3 for correct, −1 for wrong, 0 for unattempted. The resulting distribution of 400 marks values is used to compute a median and interquartile range, which are then looked up in the percentile table.
Each chapter is rated on a three-point scale. Solid — you understand the concepts deeply and can reliably attempt and answer questions at this difficulty level. Okay — you have covered the chapter but have gaps; attempt rate and accuracy are moderate. Weak — you have not prepared this chapter thoroughly; most questions will be skipped or guessed. Unrated chapters contribute zero marks to the simulation.
Based on 2020–2025 data, the five most consistently tested QA chapters are: Geometry and mensuration (18/18 slots, Moderate); Application of Time, Speed and Distance (17/18 slots, Easy-Moderate); Quadratic and other Equations (17/18 slots, Moderate); Number systems (17/18 slots, Easy-Moderate); Time and Work (16/18 slots, Easy-Moderate). Arithmetic chapters dominate by frequency. The full breakdown is available on the individual chapter pages.
Arithmetic is consistently the largest topic by question volume in CAT QA. Across 18 slots from 2020 to 2025, Arithmetic chapters contributed approximately 144 questions in total — an average of 8.0 questions per slot. This includes chapters like Application of Time Speed and Distance, Averages, Profit and Loss, Ratio and Proportion, and others.
Based on question-weighted difficulty data, CAT QA difficulty has remained broadly consistent from 2020 to 2025. There is no strong trend in either direction.
This tool is a rough guide, not a precise forecast. Key limitations: (1) VARC is excluded entirely — there is no standardised chapter taxonomy available for the verbal section. (2) DILR is simulated at the set level, not the question level — the model does not capture the strategic decision of which set to attempt first. (3) Percentile lookup tables are aggregated from multiple sources and may not match the cutoff for your specific slot. (4) Attempt and accuracy probabilities are empirically set, not individually calibrated. Use the output as a directional estimate rather than a prediction.
The simulated median marks for each slot are looked up in a (marks, percentile) table sourced from Tarkashastra (2020–2022) and Cracku (2023–2025). Within the table, percentile is interpolated linearly between the two nearest breakpoints. The tool then averages across all slots for which data is available, weighting all slots equally. For DILR, 2020 slot data is excluded because no marks-to-percentile table is available for that year.