Need LG speed improvement advice (possible monetary prize)

User avatar
sach1282
Posts: 330
Joined: Sun Oct 18, 2009 1:50 pm

Need LG speed improvement advice (possible monetary prize)

Postby sach1282 » Tue Oct 26, 2010 3:32 pm

Hey, I'm a 2012 applicant who is preparing for the LSAT. I already took it once, about a year ago, and got more questions wrong on the LG than all other sections combined. I'm preparing to retake, either Decemeber, February, or June, depending on when I feel ready, but my LG speed is just not up to snuff. I've used Pithypike's study strategy(http://www.top-law-schools.com/forums/viewtopic.php?f=6&t=41657) of photocopying games and doing repeats/new ones, but it only seems to improve me for that one game. Even though I don't consciously remember, I must at some level because novel material is almost always the same, 70% right, 50% too slow. I took a Kaplan course and bought all the new LSATs and own the LG bible, so I have access to every LSAT question ever released. If anyone has a new strategy that worked for them, please, please, please let me know. If it works I would even be willing to provide monetary compensation. I don't want speculation though, I only want advice from those who have done extremely well on the LG. I consistently get -1 or -0 on both RC and LR, so please, no discussion of those.

tourdeforcex
Posts: 428
Joined: Fri Jun 11, 2010 2:19 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby tourdeforcex » Tue Oct 26, 2010 4:52 pm

wait.

please clarify something: how many wrong do you get in LG?

because according to what you provided, your number wrong in LG is merely "more than other all sections combined." which as i understand could be 3 as you closed by explaining that you score -0 or -1 in LR and RC.

if you are getting about 3 wrong in LG, you're pretty set and i'm not sure if anyone can really actually help.

if you are getting more than that, it is a different story. please clarify.

User avatar
sach1282
Posts: 330
Joined: Sun Oct 18, 2009 1:50 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby sach1282 » Tue Oct 26, 2010 4:59 pm

Right now, on new content, I get about 6 or 7 wrong, while giving myself about 45:00 to do it. I made this post after a particularly frustrating LG practice session, apologies if it's confusing.

On repeat games, even though I don't consciously remember them, I'm doing fine and under time, but the novel material still presents issues.

User avatar
sach1282
Posts: 330
Joined: Sun Oct 18, 2009 1:50 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby sach1282 » Tue Oct 26, 2010 5:14 pm

Thanks a lot for your input, but unfortunately for me (or fortunately) I already do all those things. When I practice, I'm just LG sections, 1 game at a time. After I finish each game, I write at the top how long it took. After the whole section is done, I go through and make sure I completely understand every question I got wrong, and why I got it wrong, and what I would have to have done to get it right. I then jot down, under the time of each game, how many I got right out of the total.

Edit: I responded to a post that got deleted, but I'll leave my response here anyway.

User avatar
applepiecrust
Posts: 476
Joined: Sat Jun 12, 2010 11:38 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby applepiecrust » Tue Oct 26, 2010 5:21 pm

sach1282 wrote:Right now, on new content, I get about 6 or 7 wrong, while giving myself about 45:00 to do it. I made this post after a particularly frustrating LG practice session, apologies if it's confusing.

On repeat games, even though I don't consciously remember them, I'm doing fine and under time, but the novel material still presents issues.


45 minutes is FARRRRR too long for the games. I averaged 28 minutes per section, generally -0 or sometimes -1 (rarely -2). We'll see how I do on the real test but as for tips:
1. Buy the LGB. Work through it all the way through. Get familiar with the different question types.
2. Once you've done that, print out/copy all the games in all the PTs you're not planning to use as full-length tests. Make multiple copies -- up to 5, even.
3. Work through about half the games once without worrying about time.
4. For the remaining games, time yourself at 8 minutes each. You will/may go over time on some of the harder games, but that's ok, as some games can be done in under 5 minutes.

The key is to really have your diagramming down (and making the maximum possible inferences before you even start), and to not get thrown off by plug-and-chug questions where you have to try multiple options.

tourdeforcex
Posts: 428
Joined: Fri Jun 11, 2010 2:19 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby tourdeforcex » Tue Oct 26, 2010 5:30 pm

the simple answer is LG Bible.

don't focus on the advice. just do the games. for the time being, give yourself infinite time. just get the questions right. and record how long it takes. do all the games. then look at if there is a pattern? i was particularly bad a grouping games. i took longer. i would get at least 1 wrong even if i'd seen the game before.

then when you've identified weakness, do those games. all of them in LGB every day. then move onto your next weaker type. then do all the games again.

then do full 35 min sections. give yourself exactly 35 minutes. understand how quickly you need to go. find the game you will do best and do it first.

User avatar
typ3
Posts: 1362
Joined: Sun Feb 28, 2010 12:04 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby typ3 » Tue Oct 26, 2010 7:11 pm

I disagree with the LGB suggestion.

My suggestion is to get the Atlas Logic Games book (now manhattan Logic Games) and the 1-20 grouping books to practice on.


While not fundamentally different from Powerscore, Atlas does give you two different diagramming techniques for relative ordering and conditional games.

I used to be terrible at long conditional grouping games. PS's methods were unnecessarily lengthy and didn't consist of much more than writing out the conditional + the contrapositive, and then trying to connect them for more inference rules to write out. This ended up me writing a lengthy tome at the bottom of the game that didn't really improve my speed.

Atlas uses the tree diagram system like Steve Schwartz uses on the LSAT Blog for relative numbering game. And for grouping / conditional game it uses a diagram method called the "logic chain". It's not like powerscore's though, and I simply call it the Pinball diagram. I can slay any grouping/conditional game in under 6 consistently now. For those two diagrams alone I would recommend Atlas over powerscore. Pure sequencing games are the same under Atlas's methods as are open / closed distribution games.

The difference is Powerscore gives you 4 different types of diagrams. Atlas gives you five. 3 that are the same and 2 unique ones.


Like someone said earlier that finishes in 28 minutes. The best way to improve at logic games is to simply get faster.

There will be 1 game that will suck a lot of time and be killer, and two easy/medium games and one medium/hard game.

If you can get the two easy games done in 5-7 minutes each. Then you free up a lot of extra banked time to spend on difficult games.

EDIT: You might want to sign up for a games only course if they're giving you trouble. I knew a few prep companies offer them.

User avatar
applepiecrust
Posts: 476
Joined: Sat Jun 12, 2010 11:38 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby applepiecrust » Tue Oct 26, 2010 8:11 pm

I think you need to find a diagramming technique that works for you. I followed LGB to get a sense of the games but figured out diagramming on my own.

User avatar
Remnantofisrael
Posts: 335
Joined: Wed May 26, 2010 4:38 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Remnantofisrael » Wed Oct 27, 2010 8:00 am

applepiecrust wrote:I think you need to find a diagramming technique that works for you. I followed LGB to get a sense of the games but figured out diagramming on my own.


Hey- you got a 'tar! Congrats.

User avatar
Blumpbeef
Posts: 3814
Joined: Sat Sep 11, 2010 3:17 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Blumpbeef » Wed Oct 27, 2010 11:14 am

Drill baby drill. Go one question at a time, try to hit as low a time as possible, below 8:45. I was doing 10-15 minute games, then I finished early on both lg sections on the real thing. Just go through about 20 LG sections worth of prblems nd you will improve.

Find the questions that are the biggest sore spots and get a strategy for them, try different diagramming techniques. My drilling took me tot the mountaintop, but then lsatblogs in/out diagrams (hopefully) got me to the promised land.

Hedwig
Posts: 835
Joined: Sun Jun 27, 2010 1:56 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby Hedwig » Wed Oct 27, 2010 11:20 am

Could you try organizing your games by type? Doing a bunch of the same type and then another bunch of a another type might help you get used to a certain type of game.

Really focus on diagramming your game. This won't help with the time issue, but maybe try a couple games, just don't time anything at all, try to figure out what would be the best diagram and make a very nice diagram. Then time yourself as you go through the questions using this nice diagram.

This may be counter intuitive and just a personal thing, but I found I worked better by doing little hypotheticals by all the questions. I hypo my way through most of the games. I have my rules and inferences but I generally have to test them out. I think this helps on the newer tests because a lot of people were complaining they had to really hypo 61's LG, but I always hypo so it seemed normal to me and I was able to finish in time.

User avatar
Blumpbeef
Posts: 3814
Joined: Sat Sep 11, 2010 3:17 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Blumpbeef » Wed Oct 27, 2010 11:31 am

I used to do a lot of hypos because I would often miss inferences. It was good training :lol: , but ideally you should b getting the correct answers with as little per question diagramming as posible. You should automatically know where the constraints are and be able to answer at least a fraction o the questions in under 15 seconds.

One thing I foun useful was to do a game, review it, then just go for a walk and think of all the possible shortcuts you could have taken, about how answer choice d could have been eliminated instantly, about how a and b are functionally identical, and are therefore both automatically wrong, and so on. Pretend like you are teaching it to someone else.

User avatar
Adjudicator
Posts: 1108
Joined: Fri Jul 23, 2010 4:18 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby Adjudicator » Wed Oct 27, 2010 11:35 am

There's not much to add; find and learn a coherent system for setting up the games, and then practice them until you get faster.

User avatar
3|ink
Posts: 7331
Joined: Wed Dec 16, 2009 5:23 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby 3|ink » Wed Oct 27, 2010 12:40 pm

Keep your money.

Even though all games are 'different', there are certain things you're supposed to recognize to solve them efficiently. For instance, when doing a grouping game, you need to take those exclusion rules to the bank.

Hedwig
Posts: 835
Joined: Sun Jun 27, 2010 1:56 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby Hedwig » Wed Oct 27, 2010 2:02 pm

dhrizek wrote:I used to do a lot of hypos because I would often miss inferences. It was good training :lol: , but ideally you should b getting the correct answers with as little per question diagramming as posible. You should automatically know where the constraints are and be able to answer at least a fraction o the questions in under 15 seconds.

One thing I foun useful was to do a game, review it, then just go for a walk and think of all the possible shortcuts you could have taken, about how answer choice d could have been eliminated instantly, about how a and b are functionally identical, and are therefore both automatically wrong, and so on. Pretend like you are teaching it to someone else.


I don't think that ideally you should be using as few hypotheticals as possible. Using hypotheticals I usually score between -0 and -1 (and usually -0) on the tests. Of course, if I failed on the real thing, I will come back here and let you know if hypotheticals suck on the real thing. However. For practice. I liked em.

cw2010
Posts: 38
Joined: Wed Sep 01, 2010 3:26 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby cw2010 » Wed Oct 27, 2010 2:11 pm

Two Suggestions:

1) You need to actively analyze each game you do. Keep a Game Journal, in which you break down every single game you do. Ask yourself what you did well and not so well in terms of the setup, timing, efficiency, questions, etc. You must begin to become very precise in terms of what you are doing.

2) Go through the LGB if you haven't already. Study it, learn it, and then apply it. Keep it with you at all times during your games review.

3) A few weeks before the Oct 2010 exam, I completed the Advanced Games Course by Powerscore. This was a big help for me. It really tied everything together and helped me to analyze a game. I suggest it for anyone who wants to improve their games scores. But please note that the instructor uses the setup notation from the LGB, so you need to have a firm grasp on this before taking the course.

4) After doing a game that really gave you trouble, WRITE a new game using similair constraints (don't copy everything obviously). I believe this will really help you to see games in a different light.
Last edited by cw2010 on Wed Oct 27, 2010 2:13 pm, edited 2 times in total.

2011Law
Posts: 822
Joined: Wed Jun 30, 2010 3:40 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby 2011Law » Wed Oct 27, 2010 2:13 pm

sach1282 wrote:If it works I would even be willing to provide monetary compensation.

:? How would that work exactly?

jlhero
Posts: 49
Joined: Mon Feb 02, 2009 11:52 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby jlhero » Wed Oct 27, 2010 4:05 pm

6-7 errors in 45 minutes means that your foundation is weak.

suggestion?

buy as many LG books as you can, learn all the methods, pick the ones you are most comfortable with, and then start drilling with 35 minutes sections.

also, redo the hardest LGs at least 3 times.

User avatar
AverageTutoring
Posts: 298
Joined: Tue Jul 27, 2010 10:18 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby AverageTutoring » Wed Oct 27, 2010 4:16 pm

You are going to hate me for this but you actually have more time then you think. Ever notice how when you're in a rush things seem to go by ultra slow? It is the same concept with games. It pays to have a sense of urgency and to be altert but rushing actually causes you to go slower. In fact, I've been studying for my structural theory exam by completing practice problems under timed conditions, and the same principle holds true. When I take my time I am more accurate and more efficient. When I rush, I fuck up and take longer.

Outside of that, it's all practice man. Take your time!

Cheers

thechee
Posts: 174
Joined: Tue Oct 06, 2009 8:42 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby thechee » Wed Oct 27, 2010 4:32 pm

http://www.amazon.com/LSAT-Logic-Games- ... 073860111X

Generally, this book has terrible reviews, but it worked very well for me. Pay zero attention to the strategy advice in this book (use LGB for that). The great thing is that this book has 100 fake logic games, most of which are considerably harder than anything you will ever find on the LSAT. If you can do all the games in this book, you'll be set. Also, there are a number of typos and wrong answers, but that doesn't diminish the book's utility as a big collection of practice problems.

User avatar
Blumpbeef
Posts: 3814
Joined: Sat Sep 11, 2010 3:17 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Blumpbeef » Thu Oct 28, 2010 12:22 am

eit wrote:
dhrizek wrote:I used to do a lot of hypos because I would often miss inferences. It was good training :lol: , but ideally you should b getting the correct answers with as little per question diagramming as posible. You should automatically know where the constraints are and be able to answer at least a fraction o the questions in under 15 seconds.

One thing I foun useful was to do a game, review it, then just go for a walk and think of all the possible shortcuts you could have taken, about how answer choice d could have been eliminated instantly, about how a and b are functionally identical, and are therefore both automatically wrong, and so on. Pretend like you are teaching it to someone else.


I don't think that ideally you should be using as few hypotheticals as possible. Using hypotheticals I usually score between -0 and -1 (and usually -0) on the tests. Of course, if I failed on the real thing, I will come back here and let you know if hypotheticals suck on the real thing. However. For practice. I liked em.


At one point I was doing 4-6 hypotheticals per question in certain game types, and it was killing me. Learning how to reuse previous hypotheticals or intuitively find the correct answer really does help with time. Of course, I say all this with all of the caveats you mentioned, and then some. I was struggling with Logic Games until the very last days up until the real thing, but at one point it all started to make sense to me and I am relatively confident about my LG performance on the actual test. It is very difficult to be sure though because I didn't have many opportunities to practice after I started feeling like I actually understood it.

Where making fewer hypos hurts is in MBT questions. I noticed that I would often put down a CBT answer and move on. Just something to watch out for.

waketiger
Posts: 54
Joined: Mon Jul 12, 2010 1:27 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby waketiger » Thu Oct 28, 2010 12:42 am

Maybe just me but I did only LG straight for about 2 weeks and after that never had any problem with them again. Pretty confident I got -0 on the October LSAT and that's after starting at -15 on my diagnostic in July.

Hedwig
Posts: 835
Joined: Sun Jun 27, 2010 1:56 am

Re: Need LG speed improvement advice (possible monetary prize)

Postby Hedwig » Thu Oct 28, 2010 12:53 am

dhrizek wrote:
eit wrote:
dhrizek wrote:I used to do a lot of hypos because I would often miss inferences. It was good training :lol: , but ideally you should b getting the correct answers with as little per question diagramming as posible. You should automatically know where the constraints are and be able to answer at least a fraction o the questions in under 15 seconds.

One thing I foun useful was to do a game, review it, then just go for a walk and think of all the possible shortcuts you could have taken, about how answer choice d could have been eliminated instantly, about how a and b are functionally identical, and are therefore both automatically wrong, and so on. Pretend like you are teaching it to someone else.


I don't think that ideally you should be using as few hypotheticals as possible. Using hypotheticals I usually score between -0 and -1 (and usually -0) on the tests. Of course, if I failed on the real thing, I will come back here and let you know if hypotheticals suck on the real thing. However. For practice. I liked em.


At one point I was doing 4-6 hypotheticals per question in certain game types, and it was killing me. Learning how to reuse previous hypotheticals or intuitively find the correct answer really does help with time. Of course, I say all this with all of the caveats you mentioned, and then some. I was struggling with Logic Games until the very last days up until the real thing, but at one point it all started to make sense to me and I am relatively confident about my LG performance on the actual test. It is very difficult to be sure though because I didn't have many opportunities to practice after I started feeling like I actually understood it.

Where making fewer hypos hurts is in MBT questions. I noticed that I would often put down a CBT answer and move on. Just something to watch out for.


What you want to do is make a hypo that uses up several answer choices at once. I can't think of an example right now (been too long since I did a PT), but if it's like "which must be true" try to make a hypothetical that breaks as many of them as possible or if you just make a random one, see if it breaks more than one.

User avatar
Blumpbeef
Posts: 3814
Joined: Sat Sep 11, 2010 3:17 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Blumpbeef » Thu Oct 28, 2010 3:00 am

Yes, thats useful for Can't be Trues.

If d) says "K can't be third" and e) says "G can't be second", make a hypo with G in second and K in third.

User avatar
Gemini
Posts: 1943
Joined: Thu Mar 11, 2010 9:23 pm

Re: Need LG speed improvement advice (possible monetary prize)

Postby Gemini » Thu Oct 28, 2010 12:30 pm

Pithypike really is awesome. For it to truly work, though, you have to learn what you did wrong. It becomes instinctual after you keep doing it.
The key here, really, is to be able to pull all the important inferences out of the rules.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: dontsaywhatyoumean and 2 guests