jakeqs
I understand this, but I am trying to understand logically why it wouldn't give an exact answer all the time. If you have a symmetrical shape centered at a lattice point, how could the number of lattice points not equal the total area minus the lattice points on the shape?
Dear
jakeqs,
I'm happy to respond.

There are a few different way to answer your intriguing question.
I guess one way is to say: I don't know how deeply you understand the circle. I don't know, for example, if you are familiar with the ancient problem of trying to
square the circle, a problem that deeply frustrated the ancient geometers & Leonardo da Vinci (possible the smartest human being ever) and eventually was proven insoluble in the 19th century, because it turns out that pi is a
transcendental number, not an algebraic number. In other words, the nature of pi and the nature of the circle are 100% completely incommensurate with our linear way of thinking. The more one appreciates the deeper realms of mathematics, the more one sees that the circle is not at all to be underestimated. Now, if the beginner knows the basic formulas, C = 2(pi)r and A = (pi)r^2, that beginner might really feel that he knows almost everything there is to know and appreciate about the circle, and in a profound way, this is a kind of hubris. There are many aspects of the circle that are very difficult to understand, and some things (such as squaring the circle) that are mathematically impossible to accomplish. Of course, all that is well beyond anything the the GMAT would ever ask, but your question touched deeply into these issue. Those two formulas are a very large part of what the GMAT expects you to know about the circle, but they are less than 1% of what there is to understand about the circle in the broader world of higher mathematics.
A more straightforward way to answer is to say: imagine a circle in the coordinate plan slowly increasing in radius. Think about the lattice points that are included as it increases:
Attachment:
circle increasing in Cartesian plane.JPG [ 40.01 KiB | Viewed 38684 times ]
The number of lattice points would not increase smoothly or linearly. Instead, at certain radii, a bunch of lattice points could be outside the circle, and then when the radius nudges a little bigger, suddenly a whole bunch get included. The jumps in the number of lattice points are discrete and irregular, and does not reflect the smooth increase in area at all. At a very profound level, there is absolutely no reason that the area of the circle (which is almost always an irrational number) would be equal to the number of lattice points contained (which is always a positive integer). There, we would be talking not only about the incommensurate nature of the linear vs. the circular, but also the incommensurate nature of of integers vs. irrational numbers. The span in incongruent logical realms is truly mind-boggling. Expecting those two to be exactly equal would be like expecting an octopus to full in love with a cockroach.
Now, even though there is essentially no way that these two numbers could be exactly equal, the area of the circle is an excellent general approximation for the number of lattice points, especially if the circle contains dozens of lattice points, and especially if the answer choices are spaced out.
Even though you don't need to know all the advanced stuff about the circle, it's very good to appreciate what things in mathematics you never should underestimate. The circle is one of those topics.
Does all this make sense?
Mike