CAT 17: General thoughts (Attempts, Percentiles, etc)

Hope you all had a good go at CAT’17. In the immediate aftermath of the exam, it is inevitable that the mind will think about the potential score. Still, percentile guessing is a mugs game. Accuracy is a critically important but unavailable data point for doing this exercise. Nevertheless, we all like to pretend that we can forecast stuff with reasonable accuracy. But first, my take on the three sections.

DI-LR was delightfully tough

DI LR section was brilliantly set up. There were no freebies, no innocuous loosteners, no place to gather our breath and get a few easy ones under the belt. It was like facing West Indies Fast bowlers on a 1980’s WACA pitch. If you were not careful you could get physically bruised – at least I felt like that. But no question was outlandishly tough, there were nothing that was vague. For half the questions, I could not even figure out whether it was DI or LR, but the standard and quality of questions were breath-taking. The most prudent strategy would have been acknowledge the brilliance of the paper, set some modest targets, achieve these and then proceed to nail Quant.

VARC and Quants were of good quality but at far lower levels of difficulty

It was conceivable that most aspirants had to the time to see almost all VARC questions and more than 75% of Quant questions. Quite a few were in the attemptable zone, and although there were tricky ones here and there, these two sections were a breeze in comparison to the dear middle section.

In VARC, the passages the were engaging and pleasant to read; while the para-jumbles were a gamble. In Quants, the spread of topics was immense – there were questions from Percentages, Averages, Profit & Loss, Progressions, Logarithms, Number Systems, Modulus, Functions, Geometry, Co-geo, Ratios, Mixtures, Speed-Time, Mensuration and Quadratic Equations. (It is likely I am missing a few here)

plus ça change, plus c’est la même chose

The more things change, the more they stay the same. For all the observations that we have made, the basic character of the exam was retained. It is a classic exam that tests aptitude over a simple syllabus. As ever, one of the sections was tougher than the other two. As has been the case in the past two years, that tougher section turned out to be DILR. I think the DILR section was perhaps even tougher on a relative basis in 2017 than was the case in 2016 or 2015. But since it is a competitive exam and everyone sees the same paper, this does not particularly help or hinder any student.

Percentile Ranges

On a +3 for correct answer, -1 for incorrect MCQ answer and no negative for TITA basis, I would expect the percentiles to be in the following ranges. Please keep in mind that there is absolutely no scientific basis beyond ”vague gut feel” for these predictions.

VARC Ranges

99.5th percentile plus – 73+
99th percentile – 66
97th percentile – 60
94th percentile – 52
90th percentile – 44
80th percentile – 34

DILR Ranges

99.5 th percentile plus – 51+
99 th percentile – 41
97th percentile – 37
94th percentile – 31
90th percentile – 24
80th percentile – 18

Quant Ranges

99.5th percentile plus – 78+
99th percentile – 70
97th percentile – 60
94th percentile – 55
90th percentile – 48
80th percentile – 36