This looks fantastic! I work in test preparation myself (though not for the LSAT) and this ticks all the boxes for the best approaches. I also really appreciate it being direct and opinionated without the obnoxious tendency of a lot of guides to denigrate alternatives.
rbarnes01 · 4h ago
> A history exam might ask a question like "When was the US Constitution ratified?" Suppose you pick answer "(a) 1776". Then you look at the answer key and you see that the correct answer was actually "(b) 1789".
Wasn't the constitution ratified in 1788?
toast0 · 4h ago
The Constitution was ratified by state's conventions in 1787, 1788, 1789, and 1790. The convention of the ninth state (New Hampshire) ratified on June 21, 1788.
However, the operation of the Constitution didn't commence until March 1789 [1] [2]. Prior to that commencement, government still operated under the Articles of Confederation. The question is poorly worded, if they're asking for when the federal government began operating under the Constitution; but it's also poorly worded if they're asking for when the ninth state ratified it.
That sums up roughly 95% of daily practice in US trials and appeals!
mdaniel · 3h ago
Heh, this is the thread I came to see on a submission about the LSAT on HN o/
ethan_smith · 2h ago
The Constitution was indeed ratified on June 21, 1788 when New Hampshire became the 9th state to ratify it, though it didn't go into effect until March 4, 1789.
rayiner · 4h ago
This is great. My tip: don’t spend money on LSAT prep (unless you have no impulse control and can’t learn from a book). Just take practice tests, and when you get an answer wrong, carefully articulate exactly why each of the wrong answers was wrong and the right answer was correct. At least when I took it (17 years ago now) there was always a clearly right answer and three clearly wrong answers. I’ve never seen a single LSAT where that wasn’t the case. During the test you should be able to not only identify the correct answer, but articulate in your head why each other answer is clearly wrong.
piker · 2h ago
Absolutely correct.
rsingel · 1h ago
Don't bother with the writing section. Leave it blank, turn in your test and leave early. The proctor might say you can't do that but you can. Nobody cares about it. No one ever has. It's only there to deflect criticism from LSAC.
almostgotcaught · 3h ago
the only secret to the LSAT is to grind practice tests. back in the day (~10 years ago) i torrented all the available tests (~70) and all the practice books. I did one of the books to learn the "tricks" and then just took all the tests. scored a 178 (and then never went to law school lol).
graeme · 45m ago
This works if you're already smart. Have seen plenty of people grind through all the tests while barely improving their score or ending well below their goals.
almostgotcaught · 38m ago
Anecdotes abound <shrug>. My only advantage going into it was 1) I was a good reader 2) I was well familiar with logical fallacies from a childhood of trolling/debating people online. But it's a standardized test that covers a very small set of skills. So just like any other standardized thing it's trainable.
giantg2 · 1h ago
That's a good score. I took a practice LSAT once. I think I got a 148 or a 152. I thought that was pretty good until I looked it up. Not terrible for not studying, but it was only the 40th percentile haha.
gowld · 2h ago
Did you track your progress across practice tests? Did you need 70 tests to reach 178 performance?
lyrrad · 1h ago
I think 70 tests is overkill. As long as you are able to analyze and learn from your mistakes, you should be able to improve doing a fraction of the tests.
My experience was from almost two decades ago and is atypical. I had decided to write it only just a month before the LSAT, and so I started prior tests only four weeks before the test date. In that time, I did 27 official practice tests. I also read a couple books and tried two unofficial tests.
I tracked my results in a spreadsheet and had a marginal improvement from 173.4 to 176 for the first five and last five tests. But, essentially all improvement happened after just ten tests.
In all practice tests, I did best in logic games, answering 98.1% correct, compared with 93.5% for logical reasoning and 91.4% for reading comprehension.
I was somewhat lucky and got 97/99 on the actual test, enough for a 180 that time (2 incorrect in logical reasoning). I had only got that twice in the practice tests with an expectation between 174 and 180 with a median of 176.
After each test I analyzed every incorrect answer as well as the ones I was less sure about. There were online forums where I could ask about, or see other people's analysis of the problem questions.
I think a strategy aimed understanding wrong answers in your weakest area can be a more efficient use of preparation time.
giantg2 · 1h ago
"I think a strategy aimed understanding wrong answers in your weakest area can be a more efficient use of preparation time."
This is how I study. I don't need to review my strong areas, I need to learn the weak ones. I can cruise through tech certs very quickly this way. People are always surprised when I can learn earn a cert after studying only 2-10 hours for it.
almostgotcaught · 1h ago
Not with intent - I remember that I was excited the first time I got a 180 (maybe half-way through?) but I didn't stop at that point. Each test only took about two hours (I didn't practice the essay section because it was well-known it didn't matter) so it wasn't such a huge investment to do them all. Including the practice book it took me about 2 months of 1-2 practice tests a day.
tzs · 4h ago
I was puzzled by the site only talking about two kinds of sections on the LSAT, so did some checking.
In 2024 they dropped the analytic reasoning section. Now it just tests logical reasoning and reading comprehension.
jmogly · 3h ago
Mixed on this — I personally struggled with logic games, being dyslexic, the diagrams would easily get mixed up. On the other hand, an extra logical reasoning section could really beat the crap out of you mentally. Wonder what people who have taken both versions think?
tzs · 3h ago
Funny thing about the logic games. I aced that section and enjoyed them.
A few weeks after the test I was looking at the magazine rack at the grocery store, and saw a magazine called "Dell Logic Puzzles". I took a look inside and it was full of the same kind of logic games as on the LSAT, with 5 different difficulty levels.
From the ads in the magazine it was clear that the target audience for this was old ladies.
I bought it figuring that since I had aced those puzzles on the LSAT (and on all my practice tests beforehand) I'd have no trouble with any of the puzzles in the magazine, but the harder ones might provide some mild amusement.
Man was I wrong. I could barely do the level 1 puzzles.
jmogly · 3h ago
Haha, I could imagine. The logic games in the LSAT aren’t particularly “hard” as much as they’re just uncommon. You’re kind of just applying procedural rules, like a human calculator. I suppose if you enjoy it as an activity, the LSAT logic games must be like child’s play.
retskrad · 3h ago
Speaking of these standardised tests, I think it's ridiculous that even for someone like Elon Musk, who read for up to 10 hours a day for most of his of pre college days and learned how to build rockets with the use of books och instruction manuals, only scored 730 on the Math section and 670 on the Verbal section for a total composite score of 1400. The fact that he couldn't get higher score means something is wrong with these tests. They are mirroring a very narrow frequency of aptitude.
Jtsummers · 3h ago
> The fact that he couldn't get higher score means something is wrong with these tests.
Does it? As others pointed out, those aren't bad scores. Not the highest scores, but not embarrassing for a better student.
It could be that his claimed reading time is just that, a claim, and not true. Or the material he was reading wasn't challenging in a way that would help with test prep to get him a higher score.
giantg2 · 1h ago
I was going to say, you can study 10 hours a day on topics that you like or find valuable and not cover all the material for the test. I took the SAT as a junior and didn't really study for it, so I hadn't covered some of the math types yet. I still did fine and got into all the colleges I applied to. It really only seems to matter if you want to go to a prestigious school.
tzs · 3h ago
Assuming those are SAT scores, at the time Musk would have taken it those scores would be around 90th percentile for math, 75th for verbal, and 92nd for the overall test. That's pretty good.
Aurornis · 3h ago
> only scored 730 on the Math section and 670 on the Verbal section for a total composite score of 1400.
Those numbers are well above average. Good results. I don't know the context but if he didn't prepare much or at all then those results are quite good.
If you expected him to get a perfect score, your expectations might be too high or there might be some hero worship going on.
ghaff · 3h ago
Been a very, very long time. But I expect that getting a top score involved being both "smart" (whatever that means exactly) and having done a ton of prep work.
Wasn't the constitution ratified in 1788?
However, the operation of the Constitution didn't commence until March 1789 [1] [2]. Prior to that commencement, government still operated under the Articles of Confederation. The question is poorly worded, if they're asking for when the federal government began operating under the Constitution; but it's also poorly worded if they're asking for when the ninth state ratified it.
[1] https://supreme.justia.com/cases/federal/us/18/420/ [2] https://tile.loc.gov/storage-services/service/ll/usrep/usrep...
That sums up roughly 95% of daily practice in US trials and appeals!
My experience was from almost two decades ago and is atypical. I had decided to write it only just a month before the LSAT, and so I started prior tests only four weeks before the test date. In that time, I did 27 official practice tests. I also read a couple books and tried two unofficial tests.
I tracked my results in a spreadsheet and had a marginal improvement from 173.4 to 176 for the first five and last five tests. But, essentially all improvement happened after just ten tests.
In all practice tests, I did best in logic games, answering 98.1% correct, compared with 93.5% for logical reasoning and 91.4% for reading comprehension.
I was somewhat lucky and got 97/99 on the actual test, enough for a 180 that time (2 incorrect in logical reasoning). I had only got that twice in the practice tests with an expectation between 174 and 180 with a median of 176.
After each test I analyzed every incorrect answer as well as the ones I was less sure about. There were online forums where I could ask about, or see other people's analysis of the problem questions.
I think a strategy aimed understanding wrong answers in your weakest area can be a more efficient use of preparation time.
This is how I study. I don't need to review my strong areas, I need to learn the weak ones. I can cruise through tech certs very quickly this way. People are always surprised when I can learn earn a cert after studying only 2-10 hours for it.
In 2024 they dropped the analytic reasoning section. Now it just tests logical reasoning and reading comprehension.
A few weeks after the test I was looking at the magazine rack at the grocery store, and saw a magazine called "Dell Logic Puzzles". I took a look inside and it was full of the same kind of logic games as on the LSAT, with 5 different difficulty levels.
From the ads in the magazine it was clear that the target audience for this was old ladies.
I bought it figuring that since I had aced those puzzles on the LSAT (and on all my practice tests beforehand) I'd have no trouble with any of the puzzles in the magazine, but the harder ones might provide some mild amusement.
Man was I wrong. I could barely do the level 1 puzzles.
Does it? As others pointed out, those aren't bad scores. Not the highest scores, but not embarrassing for a better student.
It could be that his claimed reading time is just that, a claim, and not true. Or the material he was reading wasn't challenging in a way that would help with test prep to get him a higher score.
Those numbers are well above average. Good results. I don't know the context but if he didn't prepare much or at all then those results are quite good.
If you expected him to get a perfect score, your expectations might be too high or there might be some hero worship going on.