Thoughts on Manhattan approach to Binary Grouping (in/out)

User avatar
Easy-E
Posts: 5708
Joined: Fri Feb 18, 2011 1:46 pm

Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Easy-E » Tue Sep 13, 2011 7:35 pm

I just completed the Manhattan LG guide on Binary Grouping games (in and out), and I have to admit, it seems like the approach prescribed is a bit tedious. Assuming you can manage the contrapositives in your head and follow the chain both ways, would the Manhattan method be superior to a simpler chain, such as the one below which corresponds with the example game given at the beginning of the chapter.



L --> M --> ~H --> U
|
---->W --->G


It's a bit difficult to type it, but the line splits off from L to M and W, and the two separate chains. Again, assuming I have no trouble following the chain both ways, does the Manhattan method hold an advantage? I can't remember whether the chain method I did was from the LG Bible or the Nova book I read waaaay back in my prep

User avatar
Blessedassurance
Posts: 2081
Joined: Mon Jun 27, 2011 3:42 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Blessedassurance » Tue Sep 13, 2011 7:41 pm

emarxnj wrote:I just completed the Manhattan LG guide on Binary Grouping games (in and out), and I have to admit, it seems like the approach prescribed is a bit tedious. Assuming you can manage the contrapositives in your head and follow the chain both ways, would the Manhattan method be superior to a simpler chain, such as the one below which corresponds with the example game given at the beginning of the chapter.



L --> M --> ~H --> U
|
---->W --->G


It's a bit difficult to type it, but the line splits off from L to M and W, and the two separate chains. Again, assuming I have no trouble following the chain both ways, does the Manhattan method hold an advantage? I can't remember whether the chain method I did was from the LG Bible or the Nova book I read waaaay back in my prep


It depends. I found it more useful compared to others but some may disagree. If you can overcome timing issues and remember to account for rules that do not fit neatly into the diagram, it can be very helpful.

I can't follow your diagram though. From what I remember it was something like....

In Out
X X
Y Y
Z Z

With connecting arrows etc. Maybe we're talking about different things

broncos_fan_26
Posts: 255
Joined: Tue Aug 16, 2011 6:46 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby broncos_fan_26 » Tue Sep 13, 2011 7:48 pm

I found it got way too confusing when it came to Or statements. Trying to signify that on the plan just would make it all too much in the heat of the moment on test day. If you can get it, all the more power to you, but I felt that I can still work through binary grouping games pretty quickly by just listing the conditions out traditionally. Anyone agree?

User avatar
Easy-E
Posts: 5708
Joined: Fri Feb 18, 2011 1:46 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Easy-E » Tue Sep 13, 2011 8:00 pm

Blessedassurance wrote:
emarxnj wrote:I just completed the Manhattan LG guide on Binary Grouping games (in and out), and I have to admit, it seems like the approach prescribed is a bit tedious. Assuming you can manage the contrapositives in your head and follow the chain both ways, would the Manhattan method be superior to a simpler chain, such as the one below which corresponds with the example game given at the beginning of the chapter.



L --> M --> ~H --> U
|
---->W --->G


It's a bit difficult to type it, but the line splits off from L to M and W, and the two separate chains. Again, assuming I have no trouble following the chain both ways, does the Manhattan method hold an advantage? I can't remember whether the chain method I did was from the LG Bible or the Nova book I read waaaay back in my prep


It depends. I found it more useful compared to others but some may disagree. If you can overcome timing issues and remember to account for rules that do not fit neatly into the diagram, it can be very helpful.

I can't follow your diagram though. From what I remember it was something like....

In Out
X X
Y Y
Z Z

With connecting arrows etc. Maybe we're talking about different things



Yeah, the Manhattan method was like what you have, with a column of IN and OUT (or whatever the two groups might be) and then the conditionals diagrammed as arrows criss crossing the middle or along the outside to signify two items were in the same group. The method I prefer is to write out the conditionals, and link them together. What I'm saying is that I do NOT agree with the Manhattan method. It seems unnecessarily confusing.




User avatar
Blessedassurance
Posts: 2081
Joined: Mon Jun 27, 2011 3:42 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Blessedassurance » Tue Sep 13, 2011 8:49 pm

emarxnj wrote: Yeah, the Manhattan method was like what you have, with a column of IN and OUT (or whatever the two groups might be) and then the conditionals diagrammed as arrows criss crossing the middle or along the outside to signify two items were in the same group. The method I prefer is to write out the conditionals, and link them together. What I'm saying is that I do NOT agree with the Manhattan method. It seems unnecessarily confusing.


It's just a matter of personal preference, really. Some are more comfortable with the Bible method and others tend to like the Manhattan method. A combination of the two is also credited. Many-ways-to-kill-a-cat kinda thing.

Obelisk18
Posts: 39
Joined: Sat Aug 27, 2011 12:34 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Obelisk18 » Tue Sep 13, 2011 10:13 pm

emarxnj wrote:I just completed the Manhattan LG guide on Binary Grouping games (in and out), and I have to admit, it seems like the approach prescribed is a bit tedious. Assuming you can manage the contrapositives in your head and follow the chain both ways, would the Manhattan method be superior to a simpler chain, such as the one below which corresponds with the example game given at the beginning of the chapter.



L --> M --> ~H --> U
|
---->W --->G


It's a bit difficult to type it, but the line splits off from L to M and W, and the two separate chains. Again, assuming I have no trouble following the chain both ways, does the Manhattan method hold an advantage? I can't remember whether the chain method I did was from the LG Bible or the Nova book I read waaaay back in my prep


The Manhattan method doesn't really require the ability to "manage the contrapositives in your head". The advantage of the Manhattan method, IMO, is that it's incredibly easy to simply memorize how to draw the contrapositive arrows AND, once you've drawn the arrows properly, incredibly easy to follow the chain. Memorize isn't even the right word- there are only two patterns to remember. When a statement is in the form of "if A, then B" and "if If A, then not B". It's important to understand, in principle, why arrows go in certain directions but you're not given any points for making these connections on a test. From the moment I recognize an in/out game, I can draw almost any Manhattan chain with 75 seconds. This turns a typical in/out game into a 4-7 minute affair. Sure, there are a few oddball entry's that don't fit into the chain, or do fit in but only after a great deal of thinking (I'm thinking of one game in PrepTest B) but powerscore's method doesn't make these any more sensible. That said, it probably depends on how you feel about conditional logic. I dislike diagramming. I find that, in LR, for 90% of problems it's more time consuming and no more accurate. And for LG...well, I think that LG should be rote...I think if you're able to go through an LG section without having one original thought in your head, you're probably doing alright. And I think the Manhattan method makes in/out games completely rote. Provided, again, that you understand why, in principle, it works. You won't be rewarded for sussing out the reasons for each arrow on the test, but you're liable to be burned if you've simply memorized the method, without understanding conditional logic at all.

broncos_fan_26
Posts: 255
Joined: Tue Aug 16, 2011 6:46 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby broncos_fan_26 » Wed Sep 14, 2011 12:45 am

So how exactly do you deal with these oddball entries? The point of Manhattan's system is that you can just follow the chain easily, in a mechanical fashion. How do you deal with A --> B or C, and its contrapositive?

Obelisk18 wrote:
emarxnj wrote: Sure, there are a few oddball entry's that don't fit into the chain, or do fit in but only after a great deal of thinking (I'm thinking of one game in PrepTest B) but powerscore's method doesn't make these any more sensible.

User avatar
Blessedassurance
Posts: 2081
Joined: Mon Jun 27, 2011 3:42 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Blessedassurance » Wed Sep 14, 2011 1:02 am

broncos_fan_26 wrote:So how exactly do you deal with these oddball entries? The point of Manhattan's system is that you can just follow the chain easily, in a mechanical fashion. How do you deal with A --> B or C, and its contrapositive?

Obelisk18 wrote:
emarxnj wrote: Sure, there are a few oddball entry's that don't fit into the chain, or do fit in but only after a great deal of thinking (I'm thinking of one game in PrepTest B) but powerscore's method doesn't make these any more sensible.


You can draw diagrams from A to B and from A to C and circle both to indicate the special circumstance. You can use dashes as opposed to solid lines or you can note the rule next to the whole diagram. I think the book has insights on how to deal with those. It helps to keep the diagram neat.

User avatar
Easy-E
Posts: 5708
Joined: Fri Feb 18, 2011 1:46 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Easy-E » Wed Sep 14, 2011 9:11 am

broncos_fan_26 wrote:So how exactly do you deal with these oddball entries? The point of Manhattan's system is that you can just follow the chain easily, in a mechanical fashion. How do you deal with A --> B or C, and its contrapositive?

Obelisk18 wrote:
emarxnj wrote: Sure, there are a few oddball entry's that don't fit into the chain, or do fit in but only after a great deal of thinking (I'm thinking of one game in PrepTest B) but powerscore's method doesn't make these any more sensible.



I think I was accidentally quoted here :lol:


I'm curious as well as to what the best way to diagram an either/or conditions, such as "If A then B or C, but not both". Maybe you could have a single line coming from A, then branching it off to B and C, including some kind of simple notation where the line splits to signify that it is one or the other? I suppose you would need a different system for notating a condition that allows for one or both B and C.

I'll be honest, I practiced a bit with Manhattan's system and it is starting to click, especially with games with a high number of elements that all don't necessarily fall along one single logic string. I just want to make sure I can adapt Manhattan's system to any constraints thrown at me.

Manhattan LSAT Noah
Posts: 746
Joined: Tue Oct 12, 2010 8:43 am

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Manhattan LSAT Noah » Wed Sep 14, 2011 11:38 am

We know that the chain can take a bit of time to absorb, but in our experience, it's worth it. There are definitely some games where it doesn't destroy the game any faster than other approaches and instead simply serves as a visual model of conditional statements (i.e. there are few connecting links). And there are some games that are quite heavy in compound statements and we tell students to write those to the side instead of working to put those into the chain.

For compound statements, when I first started working with the chain I would do what an earlier poster did and connect elements with circles. Then i moved to writing elements as combos, for example for X + Y --> Z, I'd write "X + Y" as an element, but now I am more of a fan of simply putting an asterisk next to the involved elements (the triggers) and noting the rule to the side. It's good to have all those options at your disposal, scan the rules to see how involved the compound statements are, and then decide.

User avatar
Easy-E
Posts: 5708
Joined: Fri Feb 18, 2011 1:46 pm

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Easy-E » Wed Sep 14, 2011 12:02 pm

Manhattan LSAT Noah wrote:We know that the chain can take a bit of time to absorb, but in our experience, it's worth it. There are definitely some games where it doesn't destroy the game any faster than other approaches and instead simply serves as a visual model of conditional statements (i.e. there are few connecting links). And there are some games that are quite heavy in compound statements and we tell students to write those to the side instead of working to put those into the chain.

For compound statements, when I first started working with the chain I would do what an earlier poster did and connect elements with circles. Then i moved to writing elements as combos, for example for X + Y --> Z, I'd write "X + Y" as an element, but now I am more of a fan of simply putting an asterisk next to the involved elements (the triggers) and noting the rule to the side. It's good to have all those options at your disposal, scan the rules to see how involved the compound statements are, and then decide.



Thanks Noah, this seems pretty solid. I can see how the chain is great for open games, I just need to get the hang of using it for games with closed assignments.

Manhattan LSAT Noah
Posts: 746
Joined: Tue Oct 12, 2010 8:43 am

Re: Thoughts on Manhattan approach to Binary Grouping (in/out)

Postby Manhattan LSAT Noah » Wed Sep 14, 2011 3:51 pm

Yeah, for closed, often the chain is not going to "pin ball" - but it works nicely to see who is in what group and whatever conditional relationships are in there. The truth is, though, there are very few closed grouping games.




Return to “LSAT Prep and Discussion Forum”

Who is online

Users browsing this forum: Baidu [Spider], cherrygalore, dontsaywhatyoumean, HokieHi307, SunDevil14 and 4 guests