ATMA common mistakes
ATMA · Understand the traps and how to avoid them.
In Verbal Ability, spending 90–120 seconds re-reading a reading-comprehension passage from the top before answering each question, instead of reading once and returning only to the relevant lines.
Candidates trained on school comprehension treat the passage as something to memorise, and anxiety about missing a detail drives full re-reads. ATMA RC passages are short but question-dense, so repeated full reads consume the entire verbal budget and leave grammar and vocabulary items unattempted.
What to do instead
Do one 60–90 second skim to map paragraph function (claim, example, contrast, conclusion), then answer each question by locating the specific sentence and reading two lines above and below it. Practise with a timer capped at 75 seconds per RC question and log how many questions you left blank; if blanks exceed two per section, shorten the first read further.
In Verbal Ability, answering para-jumble and sentence-completion items by 'what sounds right' rather than by identifying the opening sentence and the logical connector.
Spoken English fluency creates false confidence; candidates pick the option that reads smoothly in isolation and ignore pronoun reference, chronological markers, and contrast words that determine order. ATMA jumbles usually hinge on one mandatory opener and one mandatory closer.
What to do instead
Apply a fixed two-step routine: first find the sentence with no backward-pointing pronouns or definite articles (the opener), then find the pair joined by a cause, contrast, or example marker. Eliminate options that violate either anchor before reading the full sequence. Drill 20 jumbles using only this anchor method until it becomes automatic.
In Quantitative Ability, attempting every question in strict serial order and burning six or more minutes on one arithmetic or geometry problem early in the section.
School exams reward completing the paper, so candidates carry a 'finish everything' reflex. ATMA quant is deliberately mixed in difficulty and has no sectional time limit inside the overall clock, so one stubborn problem silently steals time from three easy ones later.
What to do instead
Adopt a two-pass strategy: first pass, solve only questions you can finish in under 90 seconds and mark the rest; second pass, return to marked items. Enforce a hard 90-second stopwatch per question in practice. Track 'easy questions left unattempted' as your key metric, not raw accuracy.
In Quantitative Ability, setting up full algebraic equations for word problems (time-speed-distance, mixtures, percentages) instead of substituting the answer options or picking convenient numbers.
Coaching and school maths condition candidates to show working; under time pressure the equation route adds steps and arithmetic-error surface. Many ATMA quant items are engineered so that back-substitution from options or assuming a base value of 100 resolves them in two lines.
What to do instead
Before writing any equation, scan the options: if they are numeric and the question asks for a single value, test options from the middle outward. For percentage and ratio problems, assign a concrete base (100 rupees, 100 litres, 12 units for LCM-based work rates). Reserve algebra for questions where options contain variables.
In Quantitative Ability, skipping data-interpretation and arithmetic sets entirely because they 'look long', leaving 8–10 marks untouched.
Long tables and multi-line charts trigger avoidance, and candidates assume length equals difficulty. In ATMA, DI sets are often arithmetic-light and reward careful reading; the marks-per-minute on a clean table set is frequently better than on a tricky standalone geometry question.
What to do instead
Open every DI set for 20 seconds and classify it as 'clean' (single table, direct ratio or percentage) or 'layered' (multiple charts needing cross-referencing). Attempt all clean sets first. Pre-learn approximation: round to two significant figures and check whether the options are far enough apart to permit it.
In Analytical Ability, drawing no diagram for seating arrangement, blood relation, or direction-sense puzzles and trying to hold the constraints in working memory.
Candidates believe they can track five or six entities mentally, and under exam stress working memory degrades first. Errors then appear as a single misassigned seat that invalidates the whole set, wasting the time already invested.
What to do instead
Commit to a physical layout every time: a row of boxes for linear arrangements, a circle sketch for circular ones, a family tree with gender markers for blood relations, and a compass cross for directions. Write each clue as a symbol next to the diagram as you consume it. Never answer a puzzle question from memory alone.
In Analytical Ability, treating each question in a puzzle set as independent and re-deriving the arrangement from scratch for every sub-question.
Sets share a single master arrangement; candidates who solve question-by-question duplicate work and also miss the fact that a later question may add one constraint that resolves the whole set. This roughly triples the time cost per set.
What to do instead
Solve the set once to a complete or near-complete arrangement, then answer all its questions consecutively before moving on. If the arrangement is underdetermined, note the two or three viable cases explicitly and check which questions each case satisfies. Answer set questions in a block, never interleaved.
In Analytical Ability, guessing on logical-deduction and assumption/conclusion items by plausibility instead of testing whether the negation of the option destroys the argument.
These items read like general knowledge or opinion, and candidates select the option that feels true in the real world rather than the one the passage actually requires. ATMA critical-reasoning options are written to be individually believable.
What to do instead
For assumption questions, apply the negation test: negate the candidate option and ask whether the stated conclusion still holds; if it collapses, that option is the assumption. For conclusion questions, reject anything that needs outside information or is stronger than the premises (words like 'all', 'never', 'only'). Practise labelling each option as 'stated', 'implied', or 'out of scope'.
Across all three domains, allocating time evenly across sections instead of to the candidate's own strength profile, and refusing to abandon a section once it is clearly going badly.
ATMA gives one overall clock with no sectional cutoffs inside the paper, so candidates default to a fixed 1/3 split out of habit. A candidate strong in quant and weak in verbal then leaves quant marks on the table while grinding through RC.
What to do instead
Set a personal time plan from mock data, not symmetry: for example 30 minutes verbal, 40 quant, 20 analytical, with 5 minutes buffer. Write the plan on the rough sheet before starting and check the clock at each boundary. If a section overruns by more than five minutes, cut losses and move on, returning only if buffer remains.
Taking full-length mocks without a post-mortem, so the same error type (misread question stem, sign slip, wrong option marked on the sheet) repeats in every test.
Candidates equate practice volume with preparation and treat the score as the only output. Without classifying errors, they cannot distinguish knowledge gaps from process errors, and process errors are the cheaper and larger source of lost marks.
What to do instead
Maintain an error log with four columns: question number, error type (concept, calculation, misread, time), and the one-line correction. After each mock, re-solve every wrong and skipped question untimed, then re-attempt the same mock's error list after 48 hours. Review the log before the next mock and confirm the top error type has shrunk.