I took both the September and December 2014 LSAT administrations, and I got a 171 each time. In both instances, I got a -5 on the logic games, which really killed me. I have been doing the foolproof method religiously for 6 months now, and in the weeks before my December LSAT, I was doing 30 games a day. I felt comfortable and really thought I would get a -1 or -0 on the section. Instead, I got -5. The story was very similar for September (although I was only doing about 15 games per day for that one).
In my preparation I did fully simulated practice tests for September 06 onward, and got -0 on games six different times, -1 or 2 five times, and -5 or more on three tests (June 10, Oct 12, and Dec 12). The games seem to have gotten harder for me in some recent tests.
I have done every single game from Preptest 1 onward, and I've done many of those 13+ times. All the games are familiar to me by now. I'm always -2 or better on each LR section, and -3 or better on RC. LG is really holding me back.
I was wondering if you guys had any ideas for how I can get better. I am willing to try anything. I think it's possible that I get time-pressured a bit, and coupled with the pressure of the actual exam, stop making deductions as clearly and resort more to "brute force". Also, there is an intimidation factor of seeing a brand new game, whereas now I'm practicing with games I know I've done before. I'm planning to take the LSAT for a third time in February (I have already submitted my apps), because I know I can do better than a 171 --- my average on 14 fully simulated Preptests was over 176, which included 2 consecutive 180's. One idea I had was to start doing four games at once in only 24 minutes in order to train with the pressure of reduced time. And also maybe do some fake new games I find online. But again, I'm just looking for any tips or insight. I'm willing to try ANYTHING.
Comments
The new LSAT LGs have trouble being THAT different than anything you've seen before, I know they throw in an oddball most tests but if you know all the previous ones inside out even the tests with radically different games you'll have less trouble with. When it happened in the test I took I panicked, in retrospect it wasn't that bad to set up once I stopped panicking. Focus on the layout you could use for that question, don't try and determine if it's perfect, jot it down without making the inferences and quickly see if the next couple questions would work well with this model. I find having a model already written with the questions really puts the game into perspective and identifies what model should be used if it's not the one you did take (and 90% of the time you do choose the right one). Spending 2 minutes deciding a model sucks, this lets me get to inferences quickly. The good thing about the oddball tests is that in those tests the curve is likely to be higher - the bad thing is that might mean nothing at 175 where people are likely still -0'n it.
I would highly recommend not attempting independent test makers questions. I did that for (especially RC/LR but LG as well) in the beginning and there are questions that just suck. I'll hand it to LSAC, they're tests are almost always immaculate. Almost no ambiguity when taken completely literally, rules that are almost always used at least once each, fairly consistent logical jumps at the high end questions. In my experience this has not been the case for independently made questions. I found them for the most part too easy to feasibly practice with (currently you're likely losing points on the last question of sets and on the hardest set of the section when you do -5 right? No point training yourself for easy tests) and often requiring stupid reasoning or interpreting somewhat open statements as if they had stated something rather than implying something open-endly. Maybe you can find another resource that's decent for LGs, but I wouldn't recommend it. I think some of the LSAC packages include previously released tests (international, February, etc). If you're really thirsting for new questions this might work, but do keep in mind stuff from 1990 probably isn't much more useful than a 2011 test you've already done.
And yeah if you're completing quickly, don't do that. I wish I could complete LG quickly, it's my slowest section and while I don't screw up questions very often, I also can't complete it consistently. Having 5 minutes at the end is nice, but it's not like RC or especially LR where it makes sense to skip a couple hard questions, or best guess them, and then come back to them at the end. Unless you're a machine (in which case I imagine LGs would be your best section) coming back to questions in the LG section is very difficult. You have to return to a question you had trouble with after familiarizing yourself with the problem and the rules, totally blank, and make insights that don't forget any of the restrictions. I would strongly advise you take as long as you need to answer the questions well, and make sure you adjust for the fact that you'll have passing familiarity with most of the ones you attempt at this point.
For test day questions I wouldn't even advise skipping within a set since LSAC often seems to set up a set of questions so that previous ones will force/help make inferences that later questions will need. That said if you have to skip a local question early on in the set, coming back to it after you have tackled your global problems can be a big help because global problems sometimes help point you towards an inference about the general game that often helps with the trickier questions.
Sorry if this is all old hat. You've been at it a lot longer than me and with better results (though I did butcher my LG section), so don't know what I can mention usefully.
Applicants with 175-180 are down 20% YTD.
Just reprint all LG's from old PT's and do them over.
I think maybe my situation is different though, it looks like Canadian law schools don't care about the LSAT nearly as much as American ones. I'm not sure their rankings even use it, they certainly don't like to post about it if they do. I hate how GPA matters so much. It makes sense for a lot of reasons, but they all fall apart if every undergraduate degree from every school is treated the same... which for feasibility reasons pretty much has to happen. Standardizing stuff makes it all so much simpler.