Possibly REVOLUTIONARY strategy...need help deciding!

This topic has expert replies
User avatar
Junior | Next Rank: 30 Posts
Posts: 13
Joined: Tue Feb 14, 2012 12:34 pm
Followed by:1 members
Alright, revolutionary may be too strong a word, but i wanted as many people's opinion on this as possible :) [in advance, apologies for the long post but do read on...may prove rewarding!]

Some context:
We keep hearing that initial questions are more important than the questions later on and after checking out various forums and articles I've come to believe that this is FALSE. Only those with this point of view need even read on as I base my trial strategy on this idea (i.e. All questions on GMAT be it Quant or Verbal are weighed equally, only differing based on their difficulty level, the way your score sways after a correct/incorrect answer notwithstanding).

The Strategy:
I find that in many practice tests, the accuracy with which I attempt and even comprehend Q's is significantly affected by the proportion of time i have remaining relative to my question position.
Say I'm on Q28 on Quant with 12 minutes to go (giving me 1min 12sec on avg. for the remaining 10 Q's) - Knowing that, would unnerve me as well as all but the most competent test takers and may eventually lead to either a series of guesses towards the end or worse, leaving a couple of Q's unanswered thus taking a severe toll on your score.
What i suggest then is THIS --> Start a section off with gusto, set extremely low time thresholds on the first 8-15 Q's (say shoot for 1min/Q in Quant and guessing and moving on if i hit 90secs at the most). What this does (at least for me) is it frees up more time per Q for the later Q's in the section and relieves the pressure which would otherwise creep in towards the end of a particularly challenging bout of Quants. To conclude this may not work for everyone, in fact it should ONLY work for those in similar circumstances as myself. But then, the rewards may outweigh the risk.

Personally, I've been hitting constant 700's (and one 690) on prep tests (both GMATprep and MGMAT) irrespective of the prep I've been doing over the last 7-8 weeks. I've tried this strategy on both the MGMAT and the GMATprep tests and seen my Quant score increase from a stagnant 47 to a 49 both times (once with and once without AWA). I'm yet to try it out properly for the Verbal section though.

The Concerns:
Since there is still a lot of smoke surrounding the GMAC scoring algorithm, I'm deeply concerned if such a tactic could backfire for some reason on the actual GMAT. Possible complications include the randomness of pretest/unscored Q's, whether creators of GMAT are in fact lying when they say all Q's are weighed equally (unlikely :)) and maybe for the simple fact that none of the practice tests help one decide if a strategy that worked on them are likely to work on the real deal.

Will be taking one more MGMAT before my scheduled GMAT appointment on Monday (26th Mar) to try out my strategy on both sections simultaneously. The results along with whatever anyone has to say here should help point me and possibly even future test takers in the right direction.

Cheers,
aK
Source: — GMAT Strategy |

Junior | Next Rank: 30 Posts
Posts: 26
Joined: Thu Jan 12, 2012 3:33 pm
Thanked: 10 times
GMAT Score:760

by yesman238 » Thu Mar 22, 2012 1:58 am
I'm not sure if this is really such a new idea, I'm sure at one point or another, it has crossed someone's mind. It definitely did cross mine.

I will personally advise quite strongly against this strategy.

By taking on this type of test taking, it was my experience that I ended up half guessing on many more questions.

My test taking philosophy is to hit 100% accuracy on every question. Of course, this sometimes leads to a time deficient. To gain back time, I would choose questions to 'drop' very strategically. These will be questions that I know will either take a very long time to solve, or that I do not have a firm enough grasp of the concepts to ace. When I would come across such a question, and if I'm in a time deficient, I would basically 'drop' the question, and take only 15-30 seconds to quickly eliminate 2-3 obvious wrong answers, and guess on the question with a 33-50% accuracy. This would deposit back 1.5 min into my time bank. Usually I don't have to drop too many questions, but here and there, I would, so that I stay in the positive in terms of time.

The method you're suggesting, which I have also tried a similar variation, will lead to this scenario: where you spend 15-20 minutes on the first 15 questions, but in the end, because of insufficient time on each individual question, guess many of those questions with a 50% accuracy (did enough work to limit the options down to 2, but not completely certain of which way to go - this will be especially true for DS questions), and then try to hit the other 22 questions with 100%. In the end, because of the frequent guessing, and at still only a 50% accuracy, will lead to many more wrong answers than if you strategically 'dropped' certain questions. In fact, many of those original questions, if you had put in a bit more time, you probably would have been able to answer with 100% accuracy. So why give those questions up? Give up the ones that you know with certainty that you can't beat.

Summary: If I'm very low on time (or sometimes to build a surplus), I would prefer to take a bigger leap on a smaller number of questions (but like I said, with familiarity of the test and some common sense guessing techniques, I can usually drop 2-3 obviously wrong options), thereby giving me a bigger boost in time. Trying to gain back time by reducing the time over several questions will lead to more guessing and more opportunities for error.

I hope this makes sense. Anyway, these are my thoughts. It's still up to you and what you feel comfortable with.

User avatar
Junior | Next Rank: 30 Posts
Posts: 13
Joined: Tue Feb 14, 2012 12:34 pm
Followed by:1 members

by ak69er » Thu Mar 22, 2012 2:37 am
So revolutionary is too strong then :) Only reason i thought this idea was sort of new was because i couldn't find it on any forum or on random google searches. It probably strikes many people im sure, as it did yourself.

I guess my strategy works for me as i usually get only 2-4 wrong in the first 15 odd questions, either way I tend to agree with your points on paper and they do make a lot of sense. My method doesn't entail any active decision making on whether you can tackle a question (general focus area) or will need to drop it. I've found on occasion that i go down the wrong path and later realize this and am forced to retrace my steps (wasting valuable time).

Your suggestion of actively dropping only certain Q's didn't ever cross my mind during my initial prep as i was usually reluctant to guess and move on.
I'm not sure if trying the method you suggested this late in my prep would yield good results. But I certainly will keep that in mind.

Master | Next Rank: 500 Posts
Posts: 381
Joined: Wed May 19, 2010 10:15 pm
Thanked: 41 times
Followed by:2 members

by AbhiJ » Thu Mar 22, 2012 3:34 am
There is no way to game the system PERIOD. Only way to do it is to develop competency and that comes through content knowledge,extensive practice and review. What you are trying to do artificially is what a high scorer would do naturally.

For A 750 scorer the first 8 questions will from 500 - 700 level which he will answer in 1 mins time/question. Once 700 - 800 level will come he will take 2 mins/question till the 37th question (making some mistakes in the process).

User avatar
Junior | Next Rank: 30 Posts
Posts: 13
Joined: Tue Feb 14, 2012 12:34 pm
Followed by:1 members

by ak69er » Thu Mar 22, 2012 3:55 am
Abhi,

I understand what you're saying. It was never my intention to say that this strategy was an alternative to building up competency. Also, I'd like to add that I'm not trying to do anything "artificially" or even "trick" the system...

The scenario was that i had done a significant amount of prep as far as Quant was concerned, but was seeing little to no increase in my scaled subscore (hit a ceiling at 47), since i consider math one of my strong points i absolutely had to get this score up (Holds true for Verbal to a certain extent too)...Considering my quandary, I had no choice but to keep altering my strategies to up my score.
What I've outlined in my initial post was simply something that worked for me during my experiments.

While I'm not a 750 scorer...i certainly do aspire to be! Hence the Q, What now?

Master | Next Rank: 500 Posts
Posts: 381
Joined: Wed May 19, 2010 10:15 pm
Thanked: 41 times
Followed by:2 members

by AbhiJ » Thu Mar 22, 2012 10:52 am
There could be three reasons why you have hit a plateau

1.) Certain areas that give you trouble and you have lower accuracy on those - for example topics like geometry, inequality, probability, statistics. If that's the case you are better off doing targeted practice on these topics.

2.) Speed issues across the board. You have no areas of weakness and you can solve all the Qs in 5 mins , however solving in 2 mins you are making mistake. Here you need to give many sectional tests.

3.)Silly mistakes.

There are at least 10 resources for practicing quant. Once you mention what's the issue can point you to the right resource.

User avatar
Junior | Next Rank: 30 Posts
Posts: 13
Joined: Tue Feb 14, 2012 12:34 pm
Followed by:1 members

by ak69er » Thu Mar 22, 2012 7:19 pm
Its probably a combo of points 1 & 2 u've mentioned. Although I'd like to think its more 2 than 1 :)
Silly mistakes are fewer and have decreased since when i first started prepping.
Any how i appreciate the gesture but i really don't have any time left to venture into new resources...
My date is in 3 days and I'm trying my best to remain calm, brush up my basics (reading flash cards etc.) and continue taking a couple of tests.

Fingers (and toes) crossed! :mrgreen:

Master | Next Rank: 500 Posts
Posts: 381
Joined: Wed May 19, 2010 10:15 pm
Thanked: 41 times
Followed by:2 members

by AbhiJ » Fri Mar 23, 2012 4:05 am
I would say the best way to maximize score would be to be careful in first 10 and last 7 questions. If you want to guess, guess in the middle, because there is where you have chance of experimental questions.

User avatar
Junior | Next Rank: 30 Posts
Posts: 13
Joined: Tue Feb 14, 2012 12:34 pm
Followed by:1 members

by ak69er » Sat Mar 24, 2012 6:34 am
I've tried the method you were suggesting again (on MGMAT #4) but i got my same old stale Quant score of 47. Did surprisingly well on verbal (42) and hence an overall of 720 :)
Hope i can bring up my best Quant score (49) and match it with my best Verbal score (42) on the real thing...There's no turning back now!
Thanks guys.

Master | Next Rank: 500 Posts
Posts: 381
Joined: Wed May 19, 2010 10:15 pm
Thanked: 41 times
Followed by:2 members

by AbhiJ » Sat Mar 24, 2012 11:23 am
MGMAT tests don't have experimental questions. BTW GPrep Quant is easier than MGMAT. BofL.