I have been studying for the math GRE for quite sometime now. I have been going through the princeton review and old GRE tests, and in fact without much very difficulty at all. As a way of getting more practice, I got myself a copy of "The Best Test Preparation for the GRE" produced by REA. Aside from the extraordinary number of typos and errors in solutions, I have noticed that some of the questions in REA are considerably more difficult than the ones in the Princeton Review and practice tests, involving some rather off the wall, obscure topics. Has anyone else noticed this when they compare REA to the practice tests; how about when comparing the REA book to the actual test? I ask because if the actual test is as difficult as the REA, I definitely need to do some more studying. I have read on several different places on the internet that this is the case (that many of the questions in the REA book are at a higher level, that they are obscure and/or unlikely to find their way on an actual GRE, etc.), but I am looking for further corroboration or disconfirmation of this.
The reason I ask is because I am quite worried about the GRE, so much so that I spend nearly the entire day every day studying for it, and I am not exaggerating--okay I occasionally take a Saturday off. This has taken away some time from studying higher maths to prepare me for graduate courses (e.g., going through Rudin's Real and Complex Analysis, a book on functional analysis, etc.), which I find very frustrating. So I am looking for advice: I don't want to waste time by over-studying for the GRE, but I also don't want to get a bad score.
In short, my first question is, how do the REA tests compare to the two other sources I have been using; and my second is, will doing well on the practice tests and Princeton review suffice to increase the probability of success on the GRE (i.e., is this enough)? If not, do you recommend going through further sources; are there any other practice test lurking out there? I have only been able to find four. I plan on working through Stewart, Friedberg, a few of Schaum's books, etc. Are there books superior to these?
EDIT:
I just want to thank everyone for their input. I definitely feel reassured and not as anxious over this test. I feel that I can reduce the intensity of my studying, and happily go back to studying graduate topics! I think I will still go through REA's book, but it certainly won't be my primary source anymore.
I go to a tier two graduate school in the US (rank 10-20 in math), and here is my experience.
From my year, the two people who scored $99\%$ and $98\%$ that I know were rejected by the admission... People who were admitted scored around $75\%$ to $95\%$ (though I do not have a very large data pool).
I would say for my school taking graduate classes and strong recommendation letters are more important than the GRE sub score.
You can go on thegradcafe, it is a website for people to post their application result; some include their GPA and math sub scores.
Also the math gre forum, they have one sticked thread every year about their application results with their detailed info.