What is the best strategy for guessing on the GMAT?


As you are probably aware, the GMAT is an adaptive test.  It determines the next question to give you based on if you got the last question right or wrong.  Get the question right, your score goes up but you get a harder question next time.  Get the question wrong, your score goes down but you get an easier question next time.  There is both good news and bad news in this wacky world of algorithm scoring.

Bad News

The bad news is that the test will quickly figure out how good you are and give you questions designed to always be just at the edge of your ability.  If you have a good run through the test, you should feel like you’ve been hit by a truck.  If you thought the test was easy – look out – you might have had difficulty getting the test to trigger hard questions.  Remember, your score is NOT primarily driven by how many questions you get right and wrong.  Your score is primarily driven by how HARD the questions are that you get right and wrong.  Most test takers, regardless  of their score, will miss about 1/3 of the questions on the quant and about 1/3 of the questions on the verbal portions of the test.


Good news!

The good news is that you will miss a bunch of questions regardless of how you score. That can be used to your advantage.  One difference between a very good test taker a novice test taker is that the novice test taker mostly allows the GMAT to dictate which questions he or she gets right or wrong.

An experienced test taker takes more control – quickly guessing on questions he/she knows will take too much time or questions that target a weakens and therefore offers a lower chance of getting a right answer.

Guessing on questions and moving quickly along is a critical skill to develop.  It feels just awful at first.  We are taught from a young age that the way you do well on tests is to get as many questions right as you can.  Guessing damages your score.

On a conventional test that’s true, but the GMAT is not conventional.  As you practice guessing on practice tests you’ll see that your score will actually go UP as you get more savvy about guessing.   I know…but it’s true.

Remember, in the middle of the test getting one right/one wrong (a 50/50 chance) is good.  That pattern won’t really affect your score.  Your job isn’t to get all the answers right.  Your job is to avoid stringing lots of wrong answers together in a row (getting strings of answers wrong in a row is what kills your score).  Getting the answer down to 2 choices is your job in the middle of the test.

Two precious things

On the GMAT you have two things that are equally precious to you:  1) correct answers and 2) time.  You are always trading one off for the other.  The better you get at trading those two things off, the higher your score.

On the GMAT, selecting answers is much more like placing bets in Las Vegas than taking a calculus test in college.  Your job is to increase the odds of getting a right answer as much as you can in as little time as possible.  Sometimes, you won’t increase your odds to 100% certainty.  You won’t have enough time to do that.   Instead, get the odds as good as you can in about 2 minutes or so.

Evaluating your practice tests

As you evaluate your test strategy on your practice tests, look for four things.

First, how many of the first 8 questions did you get right?  To score very well you only need to get 4-6 correct.   If you got the first 4 wrong in a row you will probably have a very difficult time getting the test to trigger harder questions.

Second, how many strings of 3 or more wrong answers in a row do you have?  You will definitely get a bunch of questions wrong.  That’s a given.  It matters a lot WHEN those wrong answers occur.  We want the wrong answers spread out – sprinkled lightly though out a sea of correct answers.  Strings of wrong answers in a row kill your score.

Third, how many questions did you take 3:30+ minutes on?  Taking too long on any one question is a rookie mistake.  Usually no question is worth 3:30 or more.  If you take 3:30 or longer on a question there are only 2 outcomes and both are bad.  The first outcome is you take 3:30 and get the question wrong.  The second outcome is that you take 3:30 seconds and get the question right.  That triggers a harder question – can you image how hard that question will be and how much time that question will take you to do?  You are playing the test above your ability.  You need to bring the test back down.  Remember:  beat the test, DON’T beat the question.

Fourth, how much time did you have left for the last 10 questions?  It’s okay to be tight on time but if you go into the last 10 question with 5 minutes left you are just opening the door for the penalty monster!  Sometimes even the more experienced test taker can get caught in that trap.  If you have 7 questions left and only 6 minutes, don’t try and do all the questions.  Target every other question.  Remember, your job is to avoid getting 3 or more questions wrong in a row.

Finally, don’t spend too much energy trying to figure out if you are doing well or if the questions seem too hard or too easy.  First it’s hard to judge that. Second, a bunch of the question are experimental and the GMAT is just gathering data.  And you won’t know which ones are real and which ones are experimental.


But most importantly, try and keep your mind focus on what you can control and what you are doing.   The GMAT is going to give you whatever question it decides to give you.  You can’t control that so don’t worry about it.  Your job is to decide if you want to tackle the question and how much time you want to spend on it.  Focus on doing that question.  Then move on to the next one.

No Comments

Be the first to start a conversation