CAT Para Jumbles: The Anchor-Pair Method, Worked Through
Para jumbles in CAT are TITA questions. There is no negative marking on them — which means a jumble you are unsure about should still be answered, and a jumble you have no read on at all is a free guess.
That single fact changes the arithmetic of the question type, and most candidates attempt these as though a wrong answer costs them a mark. It does not. What it costs is time, which is the real constraint and the thing this page is actually about.
Never leave a para jumble blank
Because these are typed-answer questions rather than multiple choice, an incorrect sequence deducts nothing. The only rational reason to skip one is that the ninety seconds are better spent elsewhere — never because you are afraid of being wrong.
Quick Answer (30-Second Read)
Time per jumble
90 seconds
Then answer and move
Core technique
Anchor pairs
Not full sequencing
- Find pairs before finding the order. Two sentences that must be adjacent do more work than any opening-sentence hunt.
- Pronouns, connectors and repeated nouns are what create pairs. Look for those, not for meaning.
- The opening sentence is the one with no backward reference — no "this", "such", "however", no definite article on a first mention.
- Never leave one blank. No negative marking means a guess is strictly better than a skip.
- Read the sentences once. Rereading to "understand the topic" is the commonest way these consume four minutes.
Marking scheme per the official CAT bulletin at iimcat.ac.in.
CAT Mock Test — jumbles under a running clock
The method is easy to follow at leisure and hard under 40 minutes. Practise it where the timer is real.
The anchor-pair method
Most candidates attack a jumble by trying to reconstruct the paragraph's argument. That is a comprehension task, it takes minutes, and it is not what the question rewards.
The anchor-pair method treats a jumble as a linking problem. You are not asked to understand the paragraph. You are asked to find which sentence must follow which — and two confirmed adjacencies usually collapse a four-sentence jumble to a single possible order.
| Step | What you do | Time |
|---|---|---|
| 1 | Read all sentences once, at normal speed, without trying to order them | ~25 sec |
| 2 | Mark every pronoun, connector and repeated noun phrase | ~15 sec |
| 3 | Build one confirmed pair from those markers | ~20 sec |
| 4 | Identify the opener — the sentence with no backward reference | ~15 sec |
| 5 | Assemble, type the sequence, move on | ~15 sec |
Step three is where the question is actually won. A four-sentence jumble has twenty-four possible orders; one confirmed adjacent pair cuts that to six, and a confirmed opener cuts it to two. You almost never need to reason about all four sentences at once.
The signals that create pairs
| Signal | What it tells you | Example |
|---|---|---|
| Pronoun with no antecedent | Something before it named the thing | "It spread quickly" needs a prior sentence naming what spread |
| Demonstrative + noun | Points back at a specific earlier statement | "This assumption", "such delays", "these findings" |
| Contrast connector | The previous sentence stated the opposing position | "But", "however", "yet", "on the contrary" |
| Additive connector | Continues a point already made | "Also", "in addition", "similarly" |
| Definite article, first mention | The noun was introduced earlier | "The scheme" when no scheme has been named yet |
| Repeated or echoed noun | Often adjacent — a concept named, then developed | "…a moral hazard." / "Moral hazard, in this sense…" |
The strongest signal is a pronoun that cannot resolve within its own sentence
If a sentence says "they were unconvinced" and contains no plural noun of its own, some earlier sentence supplied it — and usually the sentence immediately before. That is a harder constraint than any judgement about which idea logically comes first, because it is grammatical rather than interpretive.
Worked examples
Example 1 — pronoun anchor.
(1) The policy was withdrawn within a fortnight. (2) Economists had warned about it from the outset. (3) A subsidy on imported fertiliser was announced in March. (4) Their objections were dismissed as alarmist.
Pairs: "it" in (2) has no antecedent inside (2) — it needs the subsidy, so 3 precedes 2. "Their" in (4) needs "economists", so 2 precedes 4.
Sequence: 3–2–4–1. Only (3) introduces something without pointing backwards, which confirms it as the opener. Assembled in under a minute without reasoning about the economics at all.
Example 2 — contrast anchor.
(1) Yet the effect vanished once household size was controlled for. (2) Larger households consistently reported lower per-capita expenditure. (3) The correlation appeared stable across three survey rounds.
Pairs: "Yet" in (1) contradicts something. What it contradicts is the claim of stability in (3), so 3 precedes 1. (2) introduces the finding with no backward reference.
Sequence: 2–3–1. The contrast connector did all the work; the statistics were never analysed.
Example 3 — echoed noun anchor.
(1) Cities began to price it explicitly. (2) Congestion had long been treated as an unavoidable cost of density. (3) The pricing worked, though not in the way its advocates had predicted.
Pairs: "it" in (1) needs congestion, so 2 precedes 1. "The pricing" in (3) echoes "price" in (1), and the definite article marks it as already introduced, so 1 precedes 3.
Sequence: 2–1–3. Two adjacencies fully determined a three-sentence jumble.
Notice what none of the three required: an opinion about which idea is more important, or any subject knowledge. Every decision came from a grammatical dependency. That is the method working as intended.
How much VARC time are jumbles eating?
The question-level time view shows exactly where the forty minutes went. Most candidates are surprised by this one.
- Time per question
- VARC accuracy
- Sectional percentile
Finding the opener
The opening sentence is defined negatively, and that definition is more reliable than any instinct about which idea "starts" the paragraph.
The opener contains no backward reference. No unresolved pronoun, no "this" or "such", no contrast connector, no definite article on a noun that has not yet been introduced. It typically names its subject in full — "a subsidy on imported fertiliser", not "the subsidy" or "it".
When two candidates for the opener both look clean, the tie-breaker is scope: the broader, more general statement almost always precedes the specific one. A sentence about congestion as a general phenomenon opens; a sentence about what one city did with it does not.
And when you genuinely cannot decide, pick one and type an answer. The absence of negative marking means an uncertain sequence has positive expected value and a blank has none.
Where ninety seconds becomes four minutes
| Habit | What it costs |
|---|---|
| Rereading the sentences to grasp the topic | The topic is irrelevant — the links are grammatical |
| Hunting for the opener first | The hardest sentence to identify, attempted before you have any constraints |
| Trying to order all four at once | Twenty-four possibilities held in working memory, instead of two |
| Re-checking a completed sequence | Verification takes as long as solving and rarely changes the answer |
| Skipping because you are unsure | Forfeits a question that costs nothing to attempt |
The second row is the most common and the most expensive. Candidates open with "which sentence comes first?" because that is how the paragraph will read — but it is the question with the fewest constraints available to answer it. Build a pair first; the opener usually identifies itself once two sentences are locked together.
Key Takeaways
- No negative marking on jumbles. Never leave one blank — a guess strictly beats a skip.
- Find pairs before order. One adjacency cuts a four-sentence jumble from 24 possibilities to 6.
- Links are grammatical, not thematic — pronouns, demonstratives, connectors, echoed nouns.
- The opener is defined by absence: no backward reference of any kind.
- Ninety seconds, then answer. The topic does not need to be understood.
- Do not verify a finished sequence. Re-checking costs as much as solving and rarely changes anything.
People also search for
How do you solve para jumbles in CAT?
Use the anchor-pair method rather than trying to reconstruct the paragraph's argument. Read all sentences once, mark every pronoun, connector and repeated noun, build one confirmed adjacent pair from those markers, then identify the opener as the sentence with no backward reference. One confirmed pair cuts a four-sentence jumble from twenty-four possible orders to six, and a confirmed opener cuts it to two — so you almost never need to reason about all four sentences simultaneously.
Do para jumbles have negative marking in CAT?
No. Para jumbles appear as TITA questions, where you type the sequence rather than selecting an option, and non-MCQ questions carry no penalty for a wrong answer. This changes the arithmetic of the question type completely: an uncertain attempt has positive expected value and a blank has none, so there is never a marking reason to skip one. The only legitimate reason to leave a jumble is that the ninety seconds are better spent on a passage.
How much time should a CAT para jumble take?
About ninety seconds — roughly twenty-five to read, fifty to build a pair and locate the opener, and fifteen to type the sequence. Beyond two minutes the question has stopped being efficient, and the right move is to enter your best sequence and leave. Candidates routinely spend four minutes on a single jumble, almost always because they are trying to understand the paragraph's subject rather than its grammatical links.
What is the anchor-pair method?
Treating a jumble as a linking problem rather than a comprehension problem. Instead of asking what the paragraph is about, you ask which sentence must immediately follow which — a question answerable from grammar alone. A pronoun with no antecedent inside its own sentence, a demonstrative like "this assumption", a contrast connector, or a definite article on a first mention all force an adjacency. Two such adjacencies usually determine the whole sequence.
How do I identify the first sentence in a para jumble?
By what it lacks. The opener contains no backward reference of any kind: no unresolved pronoun, no "this" or "such", no contrast connector, and no definite article on a noun not yet introduced. It usually names its subject in full rather than referring to it. If two candidates both look clean, prefer the broader statement — general precedes specific. And identify the opener after building a pair, not before, since it is the hardest sentence to isolate with no constraints in hand.
Should I understand the paragraph to solve a jumble?
No, and attempting to is the single commonest reason these questions consume four minutes. Every link in a well-constructed jumble is grammatical rather than thematic — a pronoun needing an antecedent, a connector needing something to contrast with, a definite article needing a prior introduction. Worked correctly, you can sequence a jumble about monetary policy or moral philosophy without any view on either subject.
How many para jumbles appear in CAT?
They form part of the non-RC portion of VARC, alongside para summary and odd-sentence-out questions, and the exact count varies between cycles. Because the composition is not fixed, the practical approach is to prepare all three non-RC types rather than banking on a particular distribution. What is stable is that these questions carry no negative marking, which makes them structurally different from the reading comprehension questions in the same section.
Should I verify my answer after sequencing a jumble?
Generally no. Reading the assembled paragraph back to check it flows takes nearly as long as solving it did, and it very rarely changes the answer — when the grammatical anchors are solid, the sequence is already determined; when they were not, rereading tends to produce doubt rather than correction. The time is better spent on the next question. Reserve verification for cases where you built no confirmed pair at all.
Are para jumbles easier than reading comprehension in CAT?
They are faster, which is not the same thing. A jumble takes about ninety seconds against roughly three minutes per RC question once passage reading is amortised, and it carries no negative marking. But accuracy on jumbles is often lower, because a single misplaced sentence loses the whole question — there is no partial credit. The practical implication is that jumbles are efficient to attempt and unwise to rely on as your main source of VARC marks.
How do I practise para jumbles effectively?
Timed, in sets of four or five, with a ninety-second cap per question that you actually enforce. Afterwards, for each one you got wrong, write down which grammatical anchor you missed rather than noting that you were wrong — "did not notice 'their' had no plural antecedent in the sentence" is usable a month later. Practising untimed defeats the purpose entirely, since the difficulty of this question type is almost entirely about how quickly you find the first pair.
Like what you read? Share with a friend
Senior Content Writer
Co-Founder at PrepGrind, working on education content, product development, and student-focused learning resources. I also write articles and guides for MBA/MBS entrance exams, covering preparation strategies, exam updates, and useful resources for students.
