Getting "THAT" One tough Qn right and bumping up %ile

[ameya.ahr]

Hi MGMAT Team,

While analyzing my first MCAT I realized that after problem no. 6 I was at 77%ile. Up until that Qn almost every Qn was from 700-800 difficulty level and I had answered four correctly and two incorrectly.
I solved the 7th qn - got it right - in about 4+ mins (!!!) and the percentile shot up to 98 !Got the next one wrong and was back in the 70s. (@73).
I have read quite a few posts all over this forum and others advising against spending more than 2 mins (and when you know you're on the right track, then maybe upto 2:30) but if getting "That One Qn" right can bump up the percentile so high, is it worth taking a shot a couple of times - as a strategy ?? (of course, you run two major risks
1. of getting it wrong - maybe because you "think" you're on right track while you really aren't) (
2. it could be one of those "dud" questions which won't be evaluated after all
)

Goes against conventional wisdom, but do you think the reward for getting it right outweighs the risks?

Ameya

[StaceyKoprince]

No, 100% not. :)

You listed two risks, but you didn't list the most important one: you've just used up time that you need to spend on other questions.

If you go over on this one by 2m, and you do that on several others throughout, now you're behind by 6+ minutes. When you get towards the end of the section, you're going to start speeding up, and now you're going to start making a ton of errors (or not even having time to try to solve before you have to guess).

At that point, your score is going to drop, and where you end is what you get. Your score is not calculated afterwards as an average of everything. Your score is just where you're at when the test ends. So, boom, you've just killed your score.

Also, remember what your reward is when you spend a bunch of extra time to get a really hard question right: an even harder question. What are you going to do, spend 6 minutes on that one? See where this is going? :)

If it takes you 4 minutes to get that question right and bump your score up, then that score is not a sustainable score for you. Your score will drop before the test is done - and it will end up dropping below what you could have gotten if you'd maintained a more steady performance and time management plan throughout.

There are literally times when you do NOT want to (try to) get a question right, because of the implications for the rest of your test. Use the tennis analogy: that's like running so hard after a ball you're probably not going to get that you smash into the fence and twist your ankle. Not only might you lose this one point anyway, but you might injure yourself for the rest of the match. Not a good strategy!

[ameya.ahr]

You are a genius Stacey! (btw, going by the volume of your responses, you've got to be using a State of the art voice to text software, do share pls) in any case the reasoning you've put up is brilliant - and solid enough to remind that you really can't game this test after all :)
Thanks and may the force be with you!
Ameya

[StaceyKoprince]

You're welcome.

And no - I just type really, really fast. Like people-stare-at-me-in-airports / coffee shops fast. :)

[ameya.ahr]

Whoao! Still making my mind up on what's more awesome, the skill itself or its description! I always thought there's an army of people sitting in a room answering questions & giving you the credit - never took the variable of typing speed into account (make a mental note for CR & essay questions - check!)

Ok, back to business, tanked quant yet again in my CAT exam # 3. Time management to blame again. Got 2 700-800 Qs. wrong back to back & percentile went from 98 to 71. Had spent 3.5 mins on each thinking all the while that "I was justtttt there !! & answer was justttt around the corner" .. then spent 4 (!!!!) mins on one more 700-800 Qn and got it right (this didn't bump up the %ile though) . . But by this time, was behind the clock and randomly answered the next qn. just to save time to make up for lost time (this hurt the %ile again) & even after a series of 5-6 right questions of 700-800 level I was stuck in the %ile of 70s. (# will refer to this situation in next paragraph)..

Given the preamble that I am hardwired to choose accuracy over speed, what I don't understand is, if what's going to matter only towards the end is going to count (in terms of %ile), would it make sense to just randomly answer a few questions in between to make up for time and to concentrate a little better and harder on the questions towards the end? or, since this may or may not take me out of the 70s and put me in the 90s (as evident from the (#) above) look at the entire quant as a batch of 7-8 questions and attack accordingly? (like end every 8th question in 16mins no matter what) ..

Am confused !


Hope you're able to answer a not so articulate qn.

*edit: oh & also, there's a DS question - "Visualizing Triangles" - as big a fan as I am - of lateral thinking - there just doesn't seem to be any way on earth to solve this qn in under 2 mins. How do you recognize such irritants early on & get them out of way, especially in Geometry (esp. so, because its easy to skip them in word problems/no. theory - you either know or you don't - Geometry can always be tricky cz sometimes in just one trick, BAM!! you get the answer!)

[StaceyKoprince]

lol - yeah, if you were standing next to me talking at normal speed, I could pretty much transcribe what you were saying in real time. I've actually had people ask if I'm just randomly banging on the keyboard and not really typing. :)

This:

Got 2 700-800 Qs. wrong back to back & percentile went from 98 to 71.

is not a problem.

This:

Had spent 3.5 mins on each thinking all the while that "I was justtttt there !!

is a problem.

I don't care that you got them wrong - and neither should you. You were gonna get them wrong anyway, and you can get them wrong and still score 700+.

I do care, though, that you wasted time, because that cost you other questions that you COULD have gotten right if you hadn't wasted time on the too-hard questions.

Read this:
http://www.manhattangmat.com/blog/index ... -to-do-it/

Sound familiar? Do what it says.

Basically, if you can do a question within the normal timeframe (roughly 2 minutes, sometimes up to 2.5m), then do so. But if you're sitting there thinking "But I'm sure I can figure this out..." then you don't really know what you're doing and you should cut yourself off right now.

Don't think of it as "every 8 questions, I skip" or "I have to skip X questions early on." Make the decision based on what you see on the screen. But you DO have to learn to recognize when you're just being stubborn and spinning your wheels - and you have to let those go.

The test-writers don't really care if you know how to find the area of a circle - and neither do the b-school admissions committees. They do care, very much, whether you're able to set priorities and make hard decisions about how to use limited resources (time) - in short, they care whether you're exhibiting 'good business person' behaviors. A good business person doesn't spend 3 hours trying to do something that has a likely return of $1 while ignoring a task that could take 3 minutes and return $1,000, right?

That's what they're really testing you for - so show them that you aren't going to get distracted when they toss something that's too hard at you. You can still keep your eye on the big picture and you know when to say "That? That's not going to get done today. It's not worth my time."
http://www.manhattangmat.com/blog/index ... to-win-it/

How to know what to move on? Read this:
http://www.manhattangmat.com/blog/index ... nt-part-1/

(and start doing what it says!)