Page 1 of 1
Tips for "grouping - undefined" games?
Posted: Sun May 20, 2012 4:51 am
by humbugger
Per subject.
For some reason, this type of game gives me fits. I'm generally pretty good at games, though I tend to make a few mistakes per section, probably due to rushing.
I recently did PT 58, though, and the last game I just sat and stared at for 15 minutes or so. It's a bizarre and uncomfortable feeling. I was just wondering if anyone has tips or alternate methods to diagram these games. In general I've been using the method of the PS LG bible for them.
Thanks.
Re: Tips for "grouping - undefined" games?
Posted: Sun May 20, 2012 2:33 pm
by 2014
Here is Manhattan LSAT's diagram and question explanations for it that might help on that game specifically.
http://www.manhattanlsat.com/forums/gam ... 965bffde3c
When I saw a game that was clearly a curveball or a time drain like some of the grouping ones you see one thing I often did was just immediately get into the questions and kind of brute force through them. It is absolutely always better to be able to get a nice diagram put together quickly, but not all questions lend themselves well to that approach. So in those situations if I read through a questions and see that its a type that throws me off, instead of like you said staring at the game for 15 minutes, I jump straight into the first question. You can then hopefully use the work you did forcing your way through the first question on subsequent ones and ultimately should be able to get through the game instead of leaving questions blank.
It's a good call to save those for last too

Re: Tips for "grouping - undefined" games?
Posted: Sun May 20, 2012 3:52 pm
by MLBrandow
humbugger,
I spent ~11-12 minutes on this game when I took this PT, although I think I probably could have worked a bit faster. I was at 14 minutes when I got to this game and went extremely carefully to ensure I wasn't making any mistakes (worked each question twice). Here's what I did:
H L M P S T W
H --> ~S and ~M
M --> ~P and ~T
W --> ~P and ~S
From there I just did in/out T-charts as hypotheticals. I've got about 7 on my page.
Essentially for these games, you want to maximize your time on the actual questions and minimize your time on the rules. I generally prefer exactly the opposite of the method of that Manhattan advocates (the logic chain) which I feel over-represents the rules in a way that just jumbles things up rather than making things clearer.
When I get a game like this, I'm into the questions by about 40 seconds, whereas with Manhattan's approach, I might spend upwards of 3-4 minutes on the initial diagram, taking extra care to ensure that I don't miss a single link (otherwise the entire chart is useless).
I really feel that if you have a strong conceptual understanding of conditional statements, doing the logic chain is just needlessly time consuming. If you can read a conditional statement backwards and understand its contrapositive, then that saves you all the extraneous diagramming time.
I'm not sure your approach on these games, but you might try this minimalist approach and see how it works for you. Undefined Grouping games were singly my achilles in games until I adopted this approach, and now they give me no trouble. (Further, it's worth noting that the logic chain approach is limited and inflexible, whereas my approach, which simply diagrams the rules as they are, is flexible for any potential hybridization.)
Hope this helps and good luck!
Re: Tips for "grouping - undefined" games?
Posted: Mon May 21, 2012 3:24 am
by humbugger
Thanks for the replies, I really appreciate it!
I looked the Manhattan diagram, and I have to say, I was very skeptical. What a mess.
Then I tried it out and I'm a little embarrassed to admit: the spaghetti actually helped me. I totally understand where you guys are coming from with the "just use the rules" approach. It makes perfect sense. I've gotten to the point where I don't write the contrapositive if I've got only one or two conditional statements in a game. It's just this one game type, where all the rules are conditional, that throws me off so badly. I think having a map helps me keep track of where I'm going with the rules and it definitely helps with where to stop. I realize the risks with this approach, but I'll be sorely tempted to use it. For me, concentrating for a few minutes to make a perfect diagram is easier than concentrating for 10+ min to not screw up any answer choice.
Thanks again.
Re: Tips for "grouping - undefined" games?
Posted: Mon May 21, 2012 11:16 pm
by Manhattan LSAT Noah
humbugger wrote:I looked the Manhattan diagram, and I have to say, I was very skeptical. What a mess.
Then I tried it out and I'm a little embarrassed to admit: the spaghetti actually helped me.
MLSAT diagram = Embarrassing medical aid device.
Love it.
Re: Tips for "grouping - undefined" games?
Posted: Tue May 22, 2012 12:45 am
by Paraflam
I have no idea how to make sense out of that MLSAT diagram. I just write the conditionals w/ contrapositives and use placeholders for games like that.
Re: Tips for "grouping - undefined" games?
Posted: Tue May 22, 2012 3:26 am
by humbugger
Delicious spaghetti...
--ImageRemoved--
Paraflam wrote:I have no idea how to make sense out of that MLSAT diagram. I just write the conditionals w/ contrapositives and use placeholders for games like that.
what is a placeholder?
Re: Tips for "grouping - undefined" games?
Posted: Tue May 22, 2012 9:25 am
by Paraflam
Kind of hard to explain on here. Basically, you write out the conditional and its contrapositive, and if both of the variables on the right side of each conditional are either both negated or both not negated, you can reserve a place for one of those variables in the in or out side of the diagram (in if they're both not negated, out if they are negated).
Like if we're given this:
X --> ~Y
Y --> ~X
So if X is in, Y is out, and vice versa. That means both variables can't possibly both be in, so we'll reserve a placeholder for X/Y in the out column. That helps you remember that at least one (possibly both) will always be out.
Re: Tips for "grouping - undefined" games?
Posted: Tue May 22, 2012 8:24 pm
by noobishned
MLBrandow wrote:humbugger,
I spent ~11-12 minutes on this game when I took this PT, although I think I probably could have worked a bit faster. I was at 14 minutes when I got to this game and went extremely carefully to ensure I wasn't making any mistakes (worked each question twice). Here's what I did:
H L M P S T W
H --> ~S and ~M
M --> ~P and ~T
W --> ~P and ~S
From there I just did in/out T-charts as hypotheticals. I've got about 7 on my page.
Essentially for these games, you want to maximize your time on the actual questions and minimize your time on the rules. I generally prefer exactly the opposite of the method of that Manhattan advocates (the logic chain) which I feel over-represents the rules in a way that just jumbles things up rather than making things clearer.
When I get a game like this, I'm into the questions by about 40 seconds, whereas with Manhattan's approach, I might spend upwards of 3-4 minutes on the initial diagram, taking extra care to ensure that I don't miss a single link (otherwise the entire chart is useless).
I really feel that if you have a strong conceptual understanding of conditional statements, doing the logic chain is just needlessly time consuming. If you can read a conditional statement backwards and understand its contrapositive, then that saves you all the extraneous diagramming time.
I'm not sure your approach on these games, but you might try this minimalist approach and see how it works for you. Undefined Grouping games were singly my achilles in games until I adopted this approach, and now they give me no trouble. (Further, it's worth noting that the logic chain approach is limited and inflexible, whereas my approach, which simply diagrams the rules as they are, is flexible for any potential hybridization.)
Hope this helps and good luck!
I'm going to back this method 100%. I was doing the Manhattan method until I talked to MLBrandow (THANK YOU!). It may help some people, but for me it just wasn't the way.