User Avatar
vandyzach452
Joined
Apr 2025
Subscription
Free
User Avatar
vandyzach452
Saturday, Aug 30 2014

Generally, I agree that the hardest passage is either the third or fourth, but there are exceptions. The Puerto Rican code switching passage was second and it was by far the hardest in that PT.

User Avatar
vandyzach452
Tuesday, Sep 30 2014

[edited by Student Services]

User Avatar
vandyzach452
Friday, Aug 29 2014

I'm with Christian- I don't think there are necessarily "trends". A large, general trend is that games are a little bit more open-ended than before as a general rule in that there are many possible worlds. It just seems to me (in my opinion) that this is the case.

Each test is its own beast. PT 62 has really hard games, while there are other PTs in the 60s that have easier games.

I really don't think there is a trend. If there is one, it would be that the test will likely include one game that is highly unusual. But you approach those like you do any other game, and read it multiple times if you need to.

Good luck!

User Avatar
vandyzach452
Tuesday, Aug 26 2014

My diagnostic was somewhere in the low 140s. I am now PTing in the low 160s, and I'm not stopping there. Take as long as you need.

User Avatar
vandyzach452
Monday, Aug 25 2014

LSAT Blog did multiple interviews with a former test writer. They were pretty informative and I enjoyed reading them.

User Avatar
vandyzach452
Monday, Aug 25 2014

Eh, whatever it was, he maxed out his LSAT score. He took every PT. He probably reviewed the right way. He might not have even taken one. Just maxing out your score is the important thing.

User Avatar
vandyzach452
Thursday, Jul 24 2014

Where are you stuck at? I ask because being stuck in the 140s is much different than being stuck in, say, the low 170s.

Also, yes, studying for this test w/ an 8 to 5 job will suck, but it can be done. I am doing it now. But you CAN do it. You will just have to make sacrifices.

User Avatar
vandyzach452
Sunday, Aug 24 2014

Thanks so much for the input. I am going to try xuianyi's method and see how it goes. It seems really intuitive.

User Avatar
vandyzach452
Tuesday, Sep 23 2014

To a large extent yea

User Avatar
vandyzach452
Tuesday, Jul 22 2014

You should see the assumption. If you can't analyze the argument and see how the premises don't NECESSARILY lead to the conclusion, then you need to blind review problems (starting with easy ones) until you are good at it and/or look back at the relevant lesson.

Now, seeing the assumption and predicting the correct answer choice are two different things. You need to see the assumption. But you likely aren't going to predict the answer word-for-word a whole lot, especially on more difficult questions.

A prephrase is something that is based on the assumption. For example, if the assumption in the argument is that correlation ----> causation, the correct answer can weaken this argument in multiple ways.

It can point out the effect without the cause, it can introduce an alternate cause for the effect, it can show that it is actually the effect that is causing the cause! All of these are weakeners.

I recommend blind reviewing problems, starting with a few easy ones with an eye for seeing the gap and also of course picking the right answer.

User Avatar

Tuesday, Aug 19 2014

vandyzach452

Memory method for comparative passages question

I have a question about the memory method.

I have arrived at PT 52 and now have comparative passages on my PTs.

I know the memory method suggests pausing after each paragraph to absorb and rethink about what you just read, and then run through each paragraph again once you finish reading the passages before you start the questions.

But what about the comparative passages? After I read passage A, should I run through passage A paragraph by paragraph before reading passage B? Or should I read passage A, then read passage B, and then remind myself of the paragraph contents of both passage A and passage B? Which does the memory method suggest I do?

User Avatar
vandyzach452
Friday, Jul 18 2014

While I haven't taken every PT yet, I can't remember where the LSAT tested the negation of "most" in a question.

I am guessing that you are talking about a necessary assumption question. The negation test can help with these, but I would encourage you to use it when you have the answer choices down to two or so, and don't sue it unless you find it helpful.

Anyhow, give us the PT, Section, and question number from the problem you're having trouble with. Like I said, I can't remember when the LSAT tested this, but it could very well be the case that I have not yet seen it, or that I forgot it.

User Avatar
vandyzach452
Monday, Aug 18 2014

I think PT 68 Game 4 is a great example of where logic games can play to your strengths or weaknesses.

There are a couple of baby inferences to make up-front, but other than those (and they don't help you hardly at all), there is a TON of plug and chug in this game. Many local questions kick no rules, at least that I could see without drawing them out.

Games like these test your mental aptitude, but in a different way than games where you can basically solve the whole game up-front. Games like these, in my opinion, test your ability to remain calm more so than games where you can solve the whole thing sup=front, because in the back of your mind, you're thinking "Holy crap- I'm doing so much brute forcing in this game, did I miss something?". Then there's another questions where nothing kicks. And another. And another. At this point, the doubt becomes even more prevalent.

Sorry I kind of rambled here, but I do think different types of logic games test different abilities. You definitely have the ability that this game tests, so good work! And yes, this is generally considered a very hard game!

User Avatar
vandyzach452
Thursday, Jul 17 2014

Hi Joe,

BR works well for either of those options. The closer you get to test day, the more PTs you'll want to take. But BR works really, really well for individual LR sections, RC passages, or logic games.

User Avatar
vandyzach452
Sunday, Jul 13 2014

I find that doing the easier questions first boosts my confidence, and confidence is really important (in anything in life, really).

Personally, I wouldn't like doing the end of the section first because it would likely throw my confidence off.

But do what works for you. It's worth a shot to experiment on a PT or two.

User Avatar
vandyzach452
Sunday, Jul 13 2014

I honestly don't think it's that big of a deal.

Sometimes I just find myself doing some underlining and bracketing automatically/naturally. I say do whatever works for you.

User Avatar
vandyzach452
Monday, Aug 11 2014

You're going to have to dig deeper. Digging deeper will allow the forum to help you and it will allow you to figure out the best thing to do going forward.

Are there certain sections that are lower? What's harder? On LR, is it harder to follow the argument (see what it's saying)? On RC, are the passages harder to follow? Are you missing inferences on games or are you not finishing in time?

User Avatar
vandyzach452
Sunday, Aug 10 2014

Miguel, you are on the right track, but your last post seems to suggest that you're translating English into conditional logic incorrectly.

Christian Wayne explains the contrapositive perfectly. Just to clean it up into a few lines:

"No candy is bitter".

Candy ----> NOT Bitter

The contrapositive is: Bitter ----> NOT candy.

So if a substance is bitter, you know that it cannot possibly be candy.

Now, for the part of your post that I'm really concerned about:

NO, you CANNOT have a candy that is bitter, because of the definition of necessity. If the sufficient condition is met, the necessary will ALWAYS follow. Conditional arrows don't mess around.

Do you see what I mean? Please follow up! You are right on the brink of understanding a huge concept that the LSAT tests.

User Avatar
vandyzach452
Sunday, Aug 10 2014

Above poster nailed it.

Just to put it in slightly different terms:

In both SA and PSA questions, there will be a gap between the premises and the conclusion. Correct answers to SA questions will completely fill the gap: they will FORCE the conclusion to follow from the premises. Correct answers to PSA questions will PARTIALLY fill the gap- they will make the conclusion MORE LIKELY to follow from the premises.

User Avatar
vandyzach452
Friday, Aug 08 2014

STOP TAKING PTs!!!!!

You do not understand the concepts the LSAT is testing. So do NOT take another PT for at least a month or two.

Go through each problem slowly. Go through the games slowly. Go through LR questions slowly.

You have plenty of low-hanging fruit to grab that will easily propel you above 150.

Consider a 7sage course to learn the basics and more. But start with the basics. and be prepared to work your ass off.

Good luck!

User Avatar
vandyzach452
Friday, Aug 08 2014

Sometimes on questions like these where the argument is super-complex, instead of trying to tell myself the flaw in my head, I just go into the questions with the ARGUMENT, rather than the FLAW, in mind.

So on this weaken question, we know this is not a valid argument. But looking at the argument, it seems like a pretty strong one, at least compared to most other LSAT arguments.

So I went into the answer choices thinking to myself, "Which of these answer choices makes it less likely that, because over the last 25 years, the ratio of price paid per car to individual income has increased, that individuals who buy new cars spend a higher proportion of their income on those cars?".

When I got to (E), I thought, "Oh, well this means that lots of big new cars were bought by businesses and the government, so this could be a reason that the average price paid per car relative to individual income has gone up, yet individuals do not spend a higher portion of their income on cars".

The other answers don't even come close to giving us that.

Since I couldn't really tell myself what the flaw exactly was, I just instead understood the argument and went into the answer choices with an open mind (which is really important for weaken questions in my opinion).

Another key that helped me get through this one with no problem was realizing that this was a pretty dang good argument! So a weakener could come out of left field- and it did: entities other than individuals are the ones buying the new cars!!!!

Hope that helps!

PrepTests ·
PT123.S3.Q19
User Avatar
vandyzach452
Monday, Jul 07 2014

This argument made me squint because it kind of jumped out at me.

Did anybody else think that the conclusion was going to say something like, "So politicians' words don't match what voters say they want", but the conclusion instead makes a prediction about the future.

Anybody else have to double take after reading the conclusion?

User Avatar

Wednesday, May 07 2014

vandyzach452

Is this dangerous?

Hello all,

Whenever you start applying rules to arrive at inferences, do you go methodically through your list of rules with each piece of info you have? For example, if you know that V is in, do you go through each of your [however many] rules AND THEN take another piece of info you get from what you found with V being in and go methodically through your rules with THAT piece of info?

Or, rather, do you see what happens when V is in and "let the inferences fly from your pencil", so to speak? I have noticed that I do it this way naturally, and while it is quite an amazing feeling when all those inferences come together and it's just like bam-bam-bam, I feel like I may get a piece of information from an inference and then forget to apply a rule to it.

For example, if, from an inference I made, I discover that Z must be in, and Z being in kicks two rules, I feel like if I do not go through the rules methodically with my new piece of information, I might miss one of the rules that Z being in kicks.

But the problem with going methodically through the rules is that it seems unintuitive. I feel that it is much more intuitive to just let the inferences fly off my pencil than to take each piece of info I get and methodically go through the rules, even though the "let the inferences fly" way seems more error-prone.

So do you think it is dangerous to just let the inferences fly off you pencil, instead of taking each piece of info you have and applying it to each rule in your rule list?

User Avatar
vandyzach452
Wednesday, Sep 03 2014

Once again, unless it is a particularly brutal game/question/passage, assuming that I have left enough time in between taking the section the first time and taking it a second time, I find that I don't remember the passage and questions hardly at all and they provide not only great review but also great experimental sections for PTs.

I have no idea how repeating sections leads you to skim. But props for knowing your limits.

User Avatar
vandyzach452
Tuesday, Sep 02 2014

Another way to avoid the problem of using up PTs for experimental sections is to use repeat sections for your experimental. Yea, you'll know which one is experimental, but it won't matter, because it's practice and the point is to build up stamina. You're not going to remember all the answers from a section you did three months ago, unless one particular question/ passage terrorized you and you reviewed it for like two hours. When this happens (which hopefully isn't too common), you still get the stamina training, and repeating questions certainly won't hurt you.

User Avatar
vandyzach452
Tuesday, Sep 02 2014

I couldn't disagree more (respectfully, of course). Why train for a 10 mile marathon by running 8 miles when you practice?

User Avatar
vandyzach452
Monday, Sep 01 2014

I type out explanations for myself like what Phillip Kraft does (he posts them on the LSAT page of reddit). It really helps me see where my mistakes are.

Confirm action

Are you sure?