"How much can I trust the diagrams?" is a really interesting question. It almost seems intuitive at first, but the more you think about it, the more you realize that it's a complicated question. Here's the simplest answer I can give:
- "Scale" applies to things like sizes and angles. For instance, if two triangles aren't drawn to scale, you don't know which one is larger and which one is smaller. You don't know which side of each triangle is the longest/shortest. You don't know their angles, so you don't know whether they're acute, obtuse, or right (unless the problem tells you). You don't know whether the triangles are equilateral or isosceles (unless the problem tells you). You don't know whether they're similar to each other (unless the problem tells you).
- If a diagram in a PS problem says 'not drawn to scale', you can't assume anything about sizes or angles based on the diagram.
- If it
doesn't say 'not drawn to scale', the GMAT's rules say that it
is drawn to scale. Therefore, you can use visual estimation. However, the problem above is an example of the GMAT breaking its own rules. If I saw a problem like this on an official test, assuming that wording is still in the rules, I might complain to the GMAC. To be safe, I'd also redraw diagrams and double-check that the scale matches the information I'm given in the problem.
- There are also things that the concept of scale just doesn't apply to. For example, suppose that a triangle isn't drawn to scale. You don't know much about it. But you
do know that it's a triangle, and not, for example, a rectangle. They can't draw something that looks like it has three sides and tell you it actually has four sides, just because it's 'not to scale'.
Similarly, if there were four dots on a diagram, they couldn't say that there were actually five dots, just because it's 'not to scale'. The diagram has to be accurate to
how many things there are (dots, sides, lines, etc.)
If two lines are shown to intersect in a diagram, you know that they intersect, even if the diagram isn't drawn to scale.
If one shape is drawn inside of another shape (for example, a dot inside of a circle), you know that it's inside the circle and not outside of the circle, even if it's not drawn to scale. You don't know exactly
where inside of the circle it is, but you know that it's in there.
This bullet point also applies to Data Sufficiency. If there's a triangle drawn in a DS problem, you know that it's not necessarily to scale. You might be able to redraw it in multiple ways. However, you do know that it's a triangle and not a circle!
_________________