CAT 2026 Trend Watch

Every year the CAT 2026 exam pattern question arrives with the conducting IIM, and the speculation starts. IIM Indore is running CAT 2026, so the forums fill with theories about what an Indore paper looks like, which topics it favours, and how the difficulty will move.

Almost none of it survives contact with the record. Look at what has actually changed across the last decade and the picture is the opposite of volatile: the structure has been remarkably stable, and the things that do move are smaller and more specific than the annual speculation suggests.

A wall of pinned charts is useful only if you can tell which ones show a trend and which show a wobble. Most CAT commentary reports wobbles.

What the Record Actually Shows

The interesting story in CAT's structure is how much it settled down, not how much it changes.

The big structural moves happened in 2014 and 2020. Since then the shape has held: three sections, 120 minutes, 40 minutes each in fixed order, and 68 questions split roughly 24 in VARC and 22 each in DILR and QA. Marking has been steady at +3 for a correct answer and −1 for an incorrect MCQ, with no negative marking on TITA questions, giving a 204-mark paper.

That stability is the most useful fact available to anyone preparing, and it is the one that gets least attention because it is not news.

The Pattern Ledger: Three Columns

Sorting what you hear into these three is the whole technique, and it takes seconds once the columns exist.

Locked is the largest column and the one worth building a plan on:

• Three sections in fixed order: VARC, then DILR, then QA. No going back, no carrying unused time forward.
• Forty minutes per section, 120 minutes total.
• Marking at +3 and −1, with TITA questions carrying no penalty.
• Roughly 68 questions, and a total around 204 marks.

Drifting is where genuine year-to-year variation lives, and it is narrower than people expect:

• The exact MCQ and TITA split within each section has moved in both directions across recent years.
• The number of RC passages and the number of standalone VARC questions has varied.
• DILR has trended toward fewer, longer, more reasoning-heavy sets rather than many short ones.
• Within QA, the balance between arithmetic and algebra shifts, though both remain dominant.

Noise is the column most commentary lives in:

• The conducting IIM. There is no reliable public evidence that the institute predicts the paper's character.
• "This slot was harder." Slots differ, which is why normalisation exists.
• Single-year swings in a topic's appearance, which are indistinguishable from chance at this sample size.
• Predictions about specific topics being "due", which is not how question papers are built.

How to Prepare for a Drifting Feature

Locked features are easy: build around them. Drifting ones need a different treatment, because preparing for a point estimate leaves you exposed when it moves.

• Prepare for the range. If DILR has run between four and six sets, be ready for both, and practise both.
• Do not optimise to last year. The most recent paper is one sample, and treating it as the template is the most common overfit.
• Watch direction, not level. DILR trending toward longer sets is a usable signal; the exact set count next November is not.
• Prefer methods that hold either way. A selection strategy that works at four sets and at six beats one tuned to five.
• Rehearse the variant you like least. Drift will eventually hand you the version you did not practise.

That last point does most of the work. Aspirants who have only practised short DILR sets are the ones a long-set paper punishes, and the fix costs nothing beyond deliberately mixing the practice.

What Five Years of Stability Means for Your Plan

The practical consequences of a settled structure are larger than they first appear:

• Past papers are directly usable. Papers from 2021 onwards share the current structure, which makes them proper simulation rather than approximation.
• Sectional strategy transfers. A 40-minute plan built in August still applies in November.
• Selection skill compounds. Fixed order and fixed time mean the same decision problem every time.
• Waiting for news is not a strategy. There is very little left to announce that would change how you prepare.
• The remaining uncertainty is small, and it is best absorbed by practising the range rather than by predicting the value.

Common Mistakes in Reading CAT Trends

Related errors:

• Reading the conducting IIM as a signal. Spending attention on a variable with no demonstrated predictive value.
• Chasing "expected" topics. Preparing for a prediction rather than for the syllabus.
• Treating slot difficulty as unfair. Normalisation exists precisely because slots differ.
• Ignoring direction. Missing a genuine multi-year drift, such as DILR sets growing longer, because it is less dramatic than single-year noise.
• Consuming commentary as preparation. Reading about the paper instead of working through past papers.

A Drill for Reading the Trend Yourself

Twenty minutes with the actual papers beats any amount of commentary.

1. Take the last four CAT papers. For each, record questions per section, MCQ and TITA counts, RC passage count and DILR set count.
2. Put them side by side. Mark anything identical across all four as locked.
3. Mark anything that varies within a narrow band as drifting, and note the range.
4. Anything you cannot measure this way is noise. Write it in a third column and then ignore that column.

What the exercise usually shows:

• Far more is locked than the annual discussion suggests.
• The drifting ranges are narrow, and preparing for the whole range costs almost nothing extra.
• The claims that worried you most are usually unmeasurable, and therefore unusable.
• Twenty minutes of this replaces a season of reading predictions.

The Bottom Line

The most useful thing about CAT 2026 is how much of it is already known. The structure has held for five years, the marking has not moved, and the genuine year-to-year variation sits inside narrow bands that you can simply prepare across. Most of what gets written each August is about the one variable with the least evidence behind it.

The Pattern Ledger, Recap

• Locked: three sections, 40 minutes each, fixed order, +3 and −1, no TITA penalty.
• Drifting: MCQ and TITA splits, RC passage counts, DILR set structure. Prepare for the range.
• Noise: the conducting IIM, slot difficulty, topics being "due". Deliberately ignore.

Build the Habit on Real Papers

Trend-reading only works from primary sources. Work through CAT mock tests and past papers from 2021 onwards, which share the current structure, and use CAT previous year questions for topic-level practice. More analysis sits in the CAT exam updates archive and the CAT strategy archive, and if commentary keeps redirecting your plan it is worth having your strategy reviewed honestly.

The Optima Learn Editorial Team builds CAT preparation content from exam-pattern analysis and Optima Learn's adaptive practice data. This guide is part of our CAT 2026 exam updates series.