In this article, we’ll examine a guessing strategy to consider when tackling advanced Data Sufficiency questions on the GMAT.

To set this up, let’s assume that I’m working on a Data Sufficiency question, and I examine statement (1) first.

__Aside__: There’s no rule that requires people to begin with statement (1). So, whenever I work on a Data Sufficiency question, I begin with the statement that looks the easiest to handle.

Okay, so let’s say that I begin with statement (1), and I find that it is sufficient. Whenever this happens, I breathe a sigh of relief. Why?

There are two reasons.

First, if statement (1) is sufficient, then I know that, even if statement (2) stumps me, I can be certain that I’ve already narrowed down the correct answer to either A or D.

Second, if statement (1) is sufficient, then I know that I’m already halfway through the question. Conversely, if statement (1) is insufficient, I may not be halfway through the question, because if statement (2) is also insufficient, then I’m then forced to complete a time-consuming __third__ step and see what happens when I combine statements (1) and (2).

As you might imagine, I’m not fond of Data Sufficiency questions that have C or E as the answer. Do I dislike these questions because they’re more difficult? No. I dislike them because they often take more time to answer.

**My Theory**

Because C and E Data Sufficiency questions typically take more time to answer, many students come to believe that these questions are more difficult than other Data Sufficiency questions. This belief can have unintended consequences.

Let’s say that you encounter a very difficult Data Sufficiency question on the GMAT, and you don’t have a clue where to begin. You’re unable to make any progress with either statement, so you’re forced to guess. What’s your guess? Well, if you believe that C and E Data Sufficiency questions are especially difficult, you may be inclined to guess C or E.

Bad idea. There’s a big difference between *difficult* and *time-consuming*.

The truth of the matter is that the correct answer to an advanced Data Sufficiency question is more likely to be A, B or D (remember “BAD”) than C or E.

This, of course, doesn’t mean that the correct answer to every advanced Data Sufficiency question is A, B or D. However, these answer choices are more likely. We can see this in the Official Guide (13^{th} edition), where all 174 Data Sufficiency questions are arranged by level of difficulty. Among the __last__ (hardest) 30 questions, the breakdown of correct answers is:

A: 23% B: 23% C: 7% D: 37% E: 10%

Compare this to the breakdown of correct answers for the __first__ 144 questions:

A: 19% B: 19% C: 18% D: 26% E: 17%

Now, it’s important to note that we’re talking about **advanced** Data Sufficiency questions here. In other words, the “BAD” strategy is meant for questions in the 80^{th} percentile and beyond. So, if your quantitative score is average or below average, and you encounter a difficult Data Sufficiency question, this strategy may not apply. Having said that, if the practice questions in the Official Guide are representative of the official test, it still appears that A, B and D might be appropriate responses to any difficult DS question. That is, **if you are forced to guess**.

Naturally, the best approach to any Data Sufficiency question is to accurately determine whether each statement is sufficient and enter your response accordingly. However, when it comes to the super difficult Data Sufficiency questions that leave you scratching your head, it may be in your best interest to guess A, B or D.