There are some differences in our tests vs. the real tests, yes, and the most difficult thing to match is verbal. The natural "style" of the question-writers comes through more in verbal, so people who pick up on that will find more differences when the test-writers change (that is, when you're taking a company's practice test vs. the real test).
It's also the case that the real test has been emphasizing meaning on SCs more lately (as opposed to grammar) and so you may have noticed some differences with that as well. Ah, yes, okay - I just saw your comments about the meaning issue. Go look at some OG problems again but make yourself deal with all meaning issues FIRST and only then let yourself go to grammar. You can use problems you already did before, since you'll be looking for new issues!
On your quant comments, yes, it is important to know when to let go / guess (on both sections, actually), but you also have to make sure that you're not letting go so quickly that you give up on questions you could have gotten - so there's a balance to find there.
You mention that you seemed not to be able to recognize much. Did you study how they disguise topics and questions? At the highest levels (and that's what you're aiming for), they are still testing the same material (they don't add calc or trig or anything), but they do a much better job of disguising what they're testing, so that you also have to figure out what it is in the first place.
I discuss an example in this article:
http://www.beatthegmat.com/mba/2010/09/ ... t-problemsNumber properties topics, in particular, are easy to disguise in very tricky ways.
It doesn't surprise me that you didn't see much on probability or geometry - those topics aren't that commonly tested anyway (and aren't worth a ton of study time). Inequalities are more commonly tested and can also be disguised, so I'm wondering whether you might have seen some but didn't recognize them.
I plan to emphasize understanding my errors, reworking problems, and even looking for multiple solutions.
add:
- how they disguise things and how you can learn to decode them (use that article above as a starting point)
- looking for traps, understanding how / why the traps work (how they suck people in, how they get people to start down the wrong solution path on quant, how they get people to thing that a wrong verbal answer is actually better than the right answer, etc)
- you mention practicing and reviewing in an "appropriate ratio"; at the highest levels, that tilts even more over to review, picking these problems apart, thinking of endless variations, what if they did this to me instead, how would that change the problem, etc.
So are you going to take it again? Or are you still deciding? :)