CAT and MBA Entrance Preparation Module 9 – XAT, NMAT, SNAP, CMAT and MAT: what each adds
CMAT: five sections, including innovation
CMAT's five sections on one 180-minute clock: a time plan from a practice log, +4 and −1 marking, the innovation section's sources, and NTA percentiles.
What you will learn
- Plan 180 minutes across CMAT's five sections
- Apply +4/−1 marking to attempt decisions
- Prepare the Innovation & Entrepreneurship section from official sources
- Compute an NTA percentile from session counts
Before you start
What you’ll learn · Before you start
What you will learn
- Plan 180 minutes across CMAT's five sections
- Apply +4/−1 marking to attempt decisions
- Prepare the Innovation & Entrepreneurship section from official sources
- Compute an NTA percentile from session counts
Before you start
On this page
CMAT 2026 gave 100 questions 180 minutes, 108 seconds each, in five sections of 20, and marked every question +4 for a right answer and −1 for a wrong one. Its fifth section, Innovation & Entrepreneurship, is named in the NTA’s bulletin and not described there.
Here you plan the three hours from a practice log, price a doubtful answer under the marking, prepare the innovation section from the government’s own sources and read the score the way the NTA computes it.
One three-hour clock for five sections of twenty
The CMAT 2026 Information Bulletin lists the paper, computer-based and in English only:
| Section | Questions | Marks |
|---|---|---|
| Quantitative Techniques and Data Interpretation | 20 | 80 |
| Logical Reasoning | 20 | 80 |
| Language Comprehension | 20 | 80 |
| General Awareness | 20 | 80 |
| Innovation & Entrepreneurship | 20 | 80 |
It gives one duration, 180 minutes, for the whole paper. Its procedure for the computer-based test lets you jump to any question by its number and bookmark questions to come back to before you submit. So the plan is yours: rank the sections by the marks a minute your practice gives. Meera is an invented candidate; put your own log into the program:
"""CMAT 2026's pattern, as the NTA's Information Bulletin gives it: 100 questions in 180
minutes, five sections of 20, +4 for a right answer and -1 for a wrong one, nothing for a
blank. The program works out the pace, turns a practice log into a first pass through
the paper and prices five doubtful answers. Meera is an invented candidate; press Edit,
put your own seconds and accuracy into LOG, and run it again.
"""
MINUTES, QUESTIONS = 180, 100
RIGHT, WRONG = 4, -1
# section: (seconds a question on a first pass, share of answers right), from practice
LOG = {
"Quantitative Techniques and DI": (105, 0.70),
"Logical Reasoning": (94, 0.76),
"Language Comprehension": (67, 0.78),
"General Awareness": (25, 0.67),
"Innovation & Entrepreneurship": (30, 0.60),
}
def per_answer(p):
"""Expected marks of one answer that is right with chance p."""
return RIGHT * p + WRONG * (1 - p)
print(f"{QUESTIONS} questions in {MINUTES} minutes: {MINUTES * 60 / QUESTIONS:.0f} seconds a question,")
print(f"or {MINUTES / 5:.0f} minutes a section if the five sections shared the clock equally")
print()
rate = {s: per_answer(p) * 60 / sec for s, (sec, p) in LOG.items()}
order = sorted(LOG, key=rate.get, reverse=True)
print("A first pass, best marks a minute first:")
clock = 0.0
expected = 0.0
for s in order:
sec, p = LOG[s]
minutes = 20 * sec / 60
expected += 20 * per_answer(p)
print(f" minute {clock:5.1f} to {clock + minutes:5.1f} {s:<31} {rate[s]:.2f} marks a minute")
clock += minutes
print(f"First pass done at minute {clock:.1f}: {MINUTES - clock:.1f} minutes left for a second look")
print(f"Expected marks of the first pass: {expected:.1f} of {RIGHT * QUESTIONS}")
print()
print("Five doubtful answers priced: 4p - (1 - p) = 5p - 1 marks")
for p in (0.15, 0.20, 0.30, 0.50, 0.80):
print(f" right with chance {p:.2f}: {per_answer(p):+.2f} marks")Output
100 questions in 180 minutes: 108 seconds a question, or 36 minutes a section if the five sections shared the clock equally A first pass, best marks a minute first: minute 0.0 to 8.3 General Awareness 5.64 marks a minute minute 8.3 to 18.3 Innovation & Entrepreneurship 4.00 marks a minute minute 18.3 to 40.7 Language Comprehension 2.60 marks a minute minute 40.7 to 72.0 Logical Reasoning 1.79 marks a minute minute 72.0 to 107.0 Quantitative Techniques and DI 1.43 marks a minute First pass done at minute 107.0: 73.0 minutes left for a second look Expected marks of the first pass: 251.0 of 400 Five doubtful answers priced: 4p - (1 - p) = 5p - 1 marks right with chance 0.15: -0.25 marks right with chance 0.20: +0.00 marks right with chance 0.30: +0.50 marks right with chance 0.50: +1.50 marks right with chance 0.80: +3.00 marks
Recorded with Python 3.14.8 on macOS 26 arm64. To run it yourself: mise exec python@3.14.8 -- python3 cmat_plan.py
Runs on this device, in your browser. The first run downloads Python (about 13.5 MB), which is kept for the next runs.
Your run, in this browser
Her first pass ends at minute 107 and leaves 73 minutes, which on one clock can go to any section. Walk through it:
- Minutes 0 to 8
General Awareness first fastest marks
Her fastest marks: 25 seconds a question at 67 % right, 5.64 marks a minute. A fact she knows takes seconds, and one she does not know rarely comes back with more time.
- Minutes 8 to 18
Innovation & Entrepreneurship recall section
Thirty seconds a question at 60 % right, 4.00 marks a minute. Each doubtful question gets its first answer and a bookmark, so nothing is left blank by accident.
- Minutes 18 to 41
Language Comprehension reading
Sixty-seven seconds a question at 78 % right, 2.60 marks a minute: the reading sections come next, while she is fresh.
- Minutes 41 to 72
Logical Reasoning reasoning
Ninety-four seconds a question at 76 % right, 1.79 marks a minute.
- Minutes 72 to 107
Quantitative Techniques and DI slowest marks
A hundred and five seconds a question at 70 % right, 1.43 marks a minute: her slowest marks go last, so on a slow day these are the minutes that run short.
- Minutes 107 to 180
73 minutes for a second look second look
Every question now has an answer. The bookmarked ones get a second look, and a doubtful answer stays only if it is right more often than one time in five.
The timeline has 6 events, from Minutes 0 to 8 to Minutes 107 to 180.
Checkpoint
Check: A plan for one three-hour clock
+4 and −1: answer when you are right more than one time in five
An answer that is right with chance p is worth 4p − (1 − p) = 5p − 1 marks on average, and a blank is worth 0, so answering pays when p is above 1 ÷ (4 + 1), a break-even accuracy: The chance of being right at which answering a question has the same expected marks as skipping it: the penalty divided by the sum of the marks for a right answer and the penalty. More in “Negative marking in CAT and other MBA tests” of 20 %. The program’s last lines price five doubtful answers: at 0.30 one is worth +0.50, at 0.15 it costs 0.25. Ten doubtful answers at 0.30 average +5 marks, and they lose nothing whenever two or more are right:
In 10 tries with p = 0.3, the chance of at least 2 successes is 0.8507; on average there are 3.
The numbers as a table
| Successes | Probability |
|---|---|
| 0 | 0.0282 |
| 1 | 0.1211 |
| 2 | 0.2335 |
| 3 | 0.2668 |
| 4 | 0.2001 |
| 5 | 0.1029 |
| 6 | 0.0368 |
The readout gives 0.8507 for two or more right; the other 14.9 % of outcomes lose 5 or 10 marks.
Treating a blind pick as a sure gain
A blind pick among k options is worth (5 − k) ÷ k marks: +0.25 with four options, nothing with five. The bulletin does not say how many options a question has, so read the paper’s own instructions on the day.
Checkpoint
Check: +4 and −1, and the 20 % rule
Innovation and entrepreneurship, from the bodies that run it
The bulletin names the section and describes nothing in it, so this lesson makes no claim about what CMAT asks. It prepares what the name covers from the bodies in charge of it, logged as the general-awareness lesson logs facts:
| Source | What it covers |
|---|---|
| Startup India, run for DPIIT | The definition of a startup and the schemes for recognised ones |
| Atal Innovation Mission, NITI Aayog | Tinkering labs in schools, incubation centres, innovation challenges |
| IP India | Patents, designs, trademarks and geographical indications, and the Copyright Office |
DPIIT recognition: The status the Department for Promotion of Industry and Internal Trade gives an entity that meets the government's definition of a startup. The government's schemes for startups are open to recognised entities. shows how such a fact is used. The Startup India portal treats an entity as a startup for up to 10 years from its incorporation, 20 for a deep-tech startup, so a firm incorporated 12 years ago, outside deep tech, cannot be recognised however small it is.
Checkpoint
Check: The innovation section's sources
The NTA percentile: your share of the session
The bulletin’s Appendix VIII gives each candidate a percentile (test score): A score that says how a candidate did relative to others, as the percentage of test takers at or below (or below, depending on the test's own definition) the candidate's score.: 100 × the candidates in the session with a raw score: The marks a candidate's answers earn before any conversion: in CMAT 2026, 4 for each right answer minus 1 for each wrong one, out of 400. An NTA percentile is computed from it. equal to or below yours ÷ the candidates who appeared in it, to 7 decimal places. Ties share a percentile, and the top score of a session is always 100. CMAT 2026 was held in one shift, and 41,872 candidates appeared. The program checks the formula against the bulletin’s own example; change the toy session:
"""The NTA percentile of Appendix VIII of the CMAT 2026 Information Bulletin:
100 x (candidates in the session with a raw score equal to or below yours)
/ (candidates who appeared in the session)
worked out to 7 decimal places. The program checks the formula against the five
candidates of the bulletin's own example, works it on a toy session of twelve, where ties
share a percentile, and on CMAT 2026's single shift. Press Edit and change SESSION.
"""
def percentile(at_or_below, appeared):
return round(100 * at_or_below / appeared, 7)
# The bulletin's example: Session-3, 41,326 candidates; (candidate, at or below, its figure)
BULLETIN = [("A", 41326, 100.0), ("B", 37244, 90.1224411), ("C", 20851, 50.4549194),
("D", 13102, 31.7040120), ("E", 456, 1.1034216)]
print("The bulletin's five candidates, recomputed:")
for name, count, printed in BULLETIN:
mine = percentile(count, 41326)
print(f" {name}: {count:>6} of 41326 -> {mine:.7f} {'matches' if mine == printed else 'differs'}")
SESSION = [212, 188, 188, 160, 141, 141, 141, 96, 73, 40, 12, -9] # raw scores out of 400
print()
print("A toy session of twelve raw scores:")
for score in sorted(set(SESSION), reverse=True):
at_or_below = sum(1 for s in SESSION if s <= score)
same = SESSION.count(score)
print(f" raw {score:>4}: {at_or_below:>2} at or below, {same} with it -> {percentile(at_or_below, len(SESSION)):.7f}")
print()
appeared = 41872 # CMAT 2026, one shift (NTA's press release on the result)
print(f"CMAT 2026: {appeared} appeared; 40000 at or below a score -> {percentile(40000, appeared):.7f},")
print(f" and {appeared - 40000} candidates scored more")
print(f"27000 at or below of 28012 -> {percentile(27000, 28012):.7f}")Output
The bulletin's five candidates, recomputed: A: 41326 of 41326 -> 100.0000000 matches B: 37244 of 41326 -> 90.1224411 matches C: 20851 of 41326 -> 50.4549194 matches D: 13102 of 41326 -> 31.7040120 matches E: 456 of 41326 -> 1.1034216 matches A toy session of twelve raw scores: raw 212: 12 at or below, 1 with it -> 100.0000000 raw 188: 11 at or below, 2 with it -> 91.6666667 raw 160: 9 at or below, 1 with it -> 75.0000000 raw 141: 8 at or below, 3 with it -> 66.6666667 raw 96: 5 at or below, 1 with it -> 41.6666667 raw 73: 4 at or below, 1 with it -> 33.3333333 raw 40: 3 at or below, 1 with it -> 25.0000000 raw 12: 2 at or below, 1 with it -> 16.6666667 raw -9: 1 at or below, 1 with it -> 8.3333333 CMAT 2026: 41872 appeared; 40000 at or below a score -> 95.5292319, and 1872 candidates scored more 27000 at or below of 28012 -> 96.3872626
Recorded with Python 3.14.8 on macOS 26 arm64. To run it yourself: mise exec python@3.14.8 -- python3 nta_percentile.py
Runs on this device, in your browser. The first run downloads Python (about 13.5 MB), which is kept for the next runs.
Your run, in this browser
So 40,000 candidates at or below your score gives 95.5292319, and the other 1,872 scored more.
Key idea
A CMAT percentile counts candidates, not marks: it says what share of your session scored at or below you.
Checkpoint
Check: The NTA percentile
Not affiliated
Based on public information about the National Testing Agency. MySmartCoPilot is not affiliated with the National Testing Agency, and the National Testing Agency’s real test rules may differ from what this lesson describes.
Practise: twelve questions, untimed
Twelve questions on the plan, the marking, the innovation sources and the percentile. They start with the idea your answers show is weakest, and each one you miss returns once.
Practice that adapts to you
Practise this lesson
12 questions about 7 ideas of this lesson. Each answer is explained at once, and what you get right is kept in this browser.
Innovation and entrepreneurship from official sources: forty facts with their sources Pro: opens with a Pro, Premium or Ultimate pass
Forty facts from Startup India, the Atal Innovation Mission and IP India, each with the page that states it, ready to log and recall. A program then applies the startup test to five firms, the way an applied question would.
In this part:
- Startup India and DPIIT recognition: twelve facts
- Schemes for recognised startups: twelve facts
- The Atal Innovation Mission: seven facts
- Patents, designs, trademarks, GIs and copyright: nine facts
- The startup test applied to five firms
1 runnable example (Python) · about 6 minutes
Opening this part with your pass…
This part opens with any MySmartCoPilot pass.
- Pro pass · 1 week₹79$2.99
- 30 days₹179$5.99
- 1 year₹1,699$45
One payment; a pass never renews. The price shown is the total you pay: no fees or charges are added at checkout.
Passes can’t be bought right now, so this part can’t be opened at the moment. The free part above stands on its own. If you have a pass, enter your licence key.
It also opens every other in-depth part in Learn and every Pro tool. Sales are final, except where the law gives you a right to withdraw, for example in the EU and the UK (refund policy).
A CMAT time plan and twenty mixed questions, fully worked Pro: opens with a Pro, Premium or Ultimate pass
The minutes CMAT leaves after a first pass are worth more than the order of the sections, and a model on Meera’s log prices them. Twenty original questions follow, four a section, each worked, with a program that checks every answer it can.
In this part:
- Three plans for the minutes after the first pass, with numbers
- Quantitative Techniques and Data Interpretation: four questions
- Logical Reasoning: four questions
- Language Comprehension: four questions
- General Awareness: four questions
- Innovation & Entrepreneurship: four questions
2 runnable examples (Python) · about 7 minutes
Opening this part with your pass…
This part opens with any MySmartCoPilot pass.
- Pro pass · 1 week₹79$2.99
- 30 days₹179$5.99
- 1 year₹1,699$45
One payment; a pass never renews. The price shown is the total you pay: no fees or charges are added at checkout.
Passes can’t be bought right now, so this part can’t be opened at the moment. The free part above stands on its own. If you have a pass, enter your licence key.
It also opens every other in-depth part in Learn and every Pro tool. Sales are final, except where the law gives you a right to withdraw, for example in the EU and the UK (refund policy).
Interview questions
Warm-up: what does a CMAT percentile of 95.5292319 tell you that the raw score does not?
It says where you stand among the candidates of your session: 95.5292319 % of those who appeared scored the same as you or less. The raw score, 4 marks for each right answer minus 1 for each wrong one, says how many marks you earned out of 400 but not how hard the paper was or how others did. The percentile is not a percentage of marks, and it is computed to 7 decimal places so that few candidates tie.
-
Fresher to mid
Your practice log shows CMAT's General Awareness at 25 seconds a question and Quantitative Techniques at 105. How would you plan the 180 minutes, and what would you do with the time left after a first pass?
-
Fresher to mid
Ten minutes remain in CMAT. Twelve questions are unread, and five answers you have marked are doubtful, each right about 30 % of the time. What do you do?
-
Fresher to mid
An NTA test is held in two shifts. Two candidates have the same raw score, 160, but percentiles of 91.2 and 88.7. How can that be, and how does the NTA set an eligibility cut-off across the shifts?
Flashcards · 6 cards
Recall it, then check
Say the answer to yourself before you look: recalling it is what makes it stick.
-
What was the pattern of CMAT 2026?
100 questions in 180 minutes, 108 seconds each, in five sections of 20: Quantitative Techniques and Data Interpretation, Logical Reasoning, Language Comprehension, General Awareness, and Innovation & Entrepreneurship.
-
When does a doubtful CMAT answer pay?
Under +4 and −1 an answer is worth 5p − 1 on average, so it pays above 20 %: at 30 % it is worth +0.50, at 15 % it costs 0.25.
-
What is a blind pick worth in CMAT?
(5 − k) ÷ k marks among k options: +0.25 with four, nothing with five. The bulletin does not give the number of options, so read the paper's instructions on the day.
-
How do you prepare CMAT's Innovation & Entrepreneurship section?
The bulletin names it without describing it, so prepare from the bodies in charge: Startup India (run for DPIIT), the Atal Innovation Mission of NITI Aayog and IP India, and claim nothing about the paper.
-
For how long can a firm count as a startup for DPIIT recognition?
Up to 10 years from its incorporation, 20 for a deep-tech startup, so a firm incorporated 12 years ago outside deep tech cannot be recognised, however small it is.
-
How does the NTA compute a CMAT percentile?
100 × the candidates of your session with a raw score at or below yours ÷ those who appeared, to 7 decimal places. With 41,872 appearing, 40,000 at or below you gives 95.5292319.
Key takeaways
- CMAT 2026 gave 100 questions 180 minutes, 108 seconds each, on one clock, so a first pass of every section can end with time left for a second look.
- Under +4 and −1, an answer right with chance p is worth 5p − 1 marks, so a doubtful answer pays above 20 %.
- The bulletin names Innovation & Entrepreneurship without describing it, so prepare it from Startup India, the Atal Innovation Mission and IP India and claim nothing about the paper.
- An NTA percentile is 100 × the candidates of your session at or below your raw score ÷ those who appeared, to 7 decimal places.
Terms in this lesson
- DPIIT recognition
- The status the Department for Promotion of Industry and Internal Trade gives an entity that meets the government's definition of a startup. The government's schemes for startups are open to recognised entities.
- percentile (test score)
- A score that says how a candidate did relative to others, as the percentage of test takers at or below (or below, depending on the test's own definition) the candidate's score.
- raw score
- The marks a candidate's answers earn before any conversion: in CMAT 2026, 4 for each right answer minus 1 for each wrong one, out of 400. An NTA percentile is computed from it.
References
- CMAT 2026 Information Bulletin (National Testing Agency (NTA))
- CMAT 2026: press release on the result (National Testing Agency (NTA))
- CMAT (National Testing Agency (NTA))
- About the Startup India initiative (Startup India, DPIIT, Ministry of Commerce and Industry)
- DPIIT startup recognition and tax exemption (Startup India, DPIIT, Ministry of Commerce and Industry)
- Startup India: eligibility and benefits of recognition (Startup India, DPIIT, Ministry of Commerce and Industry)
- Startup India: intellectual property rights (Startup India, DPIIT, Ministry of Commerce and Industry)
- Credit Guarantee Scheme for Startups (Startup India, DPIIT, Ministry of Commerce and Industry)
- Atal Innovation Mission: overview (Atal Innovation Mission, NITI Aayog)
- Atal Incubation Centres (Atal Innovation Mission, NITI Aayog)
- Atal Community Innovation Centres (Atal Innovation Mission, NITI Aayog)
- Atal New India Challenge (Atal Innovation Mission, NITI Aayog)
- Mentor India (Atal Innovation Mission, NITI Aayog)
- About IP India (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Basics of patents (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Basics of designs (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Basics of trademarks (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Geographical indications: an introduction (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Basics of copyright (Office of the Controller General of Patents, Designs & Trade Marks (IP India))
- Monetary policy: overview (Reserve Bank of India)
- Key to Budget Documents, Budget 2026-2027 (Ministry of Finance, Government of India)
Related tools
Report a problem with this lesson
Kept only in this browser. Your Learn progress