What test-day attempt strategy worked across the three CAT sections?

A single cross-section CAT strategy that consistently surfaces in high-scorer Reddit post-mortems targets 65-70% accuracy on VARC, 90% accuracy across 2 full DILR sets, and a floor of 10 correct Quant questions, with the explicit understanding that you will throw these numbers out if the paper demands it.

Why Accuracy Beats Attempt Count

Chasing attempts is how you lose. A 99+ percentile score at IIM Ahmedabad or IIM Bangalore does not require attempting every question. It requires that what you mark is mostly right. One wrong answer in CAT costs you roughly 1.33 marks, so a careless over-attempt in DILR can erase three correct attempts in a single bad gamble. Accuracy targets anchor you; attempt counts are a downstream outcome, not a goal.

Section-by-Section Breakdown

VARC is the section where accuracy targets matter most because the RC passages are designed to feel ambiguous. Attempting all 24 questions and landing at 65% accuracy is a better outcome than attempting 18 with 80% accuracy in most scoring distributions.

The DILR Decision Rule

DILR is where CAT separates the IIM Calcutta admits from the near-misses. The field-tested approach: spend the first 6-8 minutes scanning all 4 sets cold, rank them by solvability, then commit fully to your top 2. If you finish both with time left, scout a third. If you are 10 minutes into a set and stuck, exit without guilt. Sunk-cost thinking in DILR is a percentile-killer. Two clean sets at 90% accuracy outperforms three messy sets at 60% almost every time.

Quant: The Floor Strategy

Quant scores for the IIM ABC shortlists hover around 80-85 percentile as a sectional floor. Getting to 10-12 correct from 22 questions, while skipping the algebra-heavy problems that eat 4+ minutes, is enough to stay in the running.

This is not a section to maximize for most profiles; it is a section to not collapse in. Identify your reliable topics (arithmetic, number systems, geometry basics) and mine them first.

Adapting When the Paper Is a Curveball

Reddit data from multiple CAT cycles confirms that prepared targets regularly get abandoned on test day. A difficult DILR set of 4 sets means 2 good sets may be genuinely hard to find.

A verbal section with 5 dense RC passages changes your timing math. The underlying principle that survives any paper: protect accuracy, adjust attempts.

Rigid attempt targets are pre-exam planning tools, not in-exam rules. High scorers who cracked calls from IIM Ahmedabad, IIM Bangalore, and IIM Calcutta consistently report that mid-exam recalibration was the actual skill being tested.

Build your mock strategy around these benchmarks for the last 6 weeks, but practice abandoning them in at least 2-3 mocks so the pivot feels routine under pressure.