grouping games Forum

[elg21]

Grouping games seem to be the only game type giving me consistent problems. I have the most difficulty with them when I choose the wrong group as my "top group," ie placing abc into xyz(top) when it wouldve been much easier to place xyz in abc. Does anyone have a rule they use to help them choose the easiest one? I realize that both can work usually but it would make my life easier if I could get more consistent with picking the "right" one.

Thanks for your help

[magickware]

Fairly certain MLSAT and PS LGB both deal with this. I just don't remember it.

So if you have those books, look them up! =D

[elg21]

Thanks for the reply. I have gone through both, don't have them on hand at the moment. However, I believe the basic idea is to visualize which things you are placing to what, I could be wrong. This strategy never seems to work for me as I always seem to think that both entities have their merits.

[Reframe]

I'm sorry; can you give me an example of the sort of thing you mean? If I'm understanding you right, this is a fairly common problem. Generally when you pick which kinds of elements organize your diagram you are looking for the most stable and minimal ones, and for the ones that make reading the diagram most fluent. So, for example, in a hybrid game, ordering elements should dominate your diagramming before grouping elements to. But an example will help me flesh this all out.

[elg21]

Sure. For example, I just went through a game where food trucks, lets call them abcdef, serves one or more of three buildings, xyz. The way I set up the game was using xyz and my base, in other words placing trucks to buildings. This made the game very difficult and when I tried it the other way, buildings to trucks, the game was much easier. I chose building initially because it seemed to make the most sense to me logically, I imagine trucks moving to different buildings, no the other way around, however, my logic seems wrong.

[Balthy]

Deleted. wrong thread:P

[Reframe]

Sure. For example, I just went through a game where food trucks, lets call them abcdef, serves one or more of three buildings, xyz. The way I set up the game was using xyz and my base, in other words placing trucks to buildings. This made the game very difficult and when I tried it the other way, buildings to trucks, the game was much easier. I chose building initially because it seemed to make the most sense to me logically, I imagine trucks moving to different buildings, no the other way around, however, my logic seems wrong.

In that game, the structure really doesn't matter! That's a matching game - at least I think that's what it's normally classed as. The central feature of this game is that you can have more than one truck per building and more than one building per truck. This means the optimal diagram is going to be a hashchart kind of thing, you know, like

Code: Select all

._A_B_C_D_E_F_
x| | | | | | |
--------------
y| | | | | | |
--------------
z| | | | | | |
--------------

And for a chart like that, the information won't change depending on how you organize it; readability is your only concern.

[Daily_Double]

.

[elg21]

Hi everyone,

I believe I misspoke when I said my issue was grouping, I think it has more to do with matching games (I confuse the two often)

Hope this clarifies things and any help would be much appreciated.

[Clearly]

Sure. For example, I just went through a game where food trucks, lets call them abcdef, serves one or more of three buildings, xyz. The way I set up the game was using xyz and my base, in other words placing trucks to buildings. This made the game very difficult and when I tried it the other way, buildings to trucks, the game was much easier. I chose building initially because it seemed to make the most sense to me logically, I imagine trucks moving to different buildings, no the other way around, however, my logic seems wrong.

In that game, the structure really doesn't matter! That's a matching game - at least I think that's what it's normally classed as. The central feature of this game is that you can have more than one truck per building and more than one building per truck. This means the optimal diagram is going to be a hashchart kind of thing, you know, like

Code: Select all

._A_B_C_D_E_F_
x| | | | | | |
--------------
y| | | | | | |
--------------
z| | | | | | |
--------------

And for a chart like that, the information won't change depending on how you organize it; readability is your only concern.

Wouldn't do this game like this. In fact the only game I've ever used a yes no graph on is with 3 people voting yes or no on three bills (tax, recreation, something)

[Reframe]

Wouldn't do this game like this. In fact the only game I've ever used a yes no graph on is with 3 people voting yes or no on three bills (tax, recreation, something)

Then you're doing them very inefficiently! This is the only diagram that can really capture everything that's going on, since you have groups going in both directions. Give it a shot; it works for my students, with, as far as I can remember, no exceptions.

[Clearly]

Wouldn't do this game like this. In fact the only game I've ever used a yes no graph on is with 3 people voting yes or no on three bills (tax, recreation, something)

Then you're doing them very inefficiently! This is the only diagram that can really capture everything that's going on, since you have groups going in both directions. Give it a shot; it works for my students, with, as far as I can remember, no exceptions.

I can assure you that a yes no grid is not the most efficient way to do 99% of games.

[y2zipper]

I'd do the buildings as a base there and stack the trucks that serve the buildings vertically, too. It's easier for me to visualize it that way. That diagram above makes sense, too. Stumbled on that by accident doing the game with the cars and options yesterday.

[Reframe]

Wouldn't do this game like this. In fact the only game I've ever used a yes no graph on is with 3 people voting yes or no on three bills (tax, recreation, something)

Then you're doing them very inefficiently! This is the only diagram that can really capture everything that's going on, since you have groups going in both directions. Give it a shot; it works for my students, with, as far as I can remember, no exceptions.

I can assure you that a yes no grid is not the most efficient way to do 99% of games.

This is a specific game type for which this grid is particularly efficient. Did you think I was suggesting it for ordering or conditional games? Look, I've been learning or teaching this test for five years, and I specialize in LG. I can point you to other games of this type if you're interested, and I can explain how to spot them when they come up.

[Reframe]

I'd do the buildings as a base there and stack the trucks that serve the buildings vertically, too. It's easier for me to visualize it that way. That diagram above makes sense, too. Stumbled on that by accident doing the game with the cars and options yesterday.

Yup! That's another game of the same type.

[Clearly]

Wouldn't do this game like this. In fact the only game I've ever used a yes no graph on is with 3 people voting yes or no on three bills (tax, recreation, something)

Then you're doing them very inefficiently! This is the only diagram that can really capture everything that's going on, since you have groups going in both directions. Give it a shot; it works for my students, with, as far as I can remember, no exceptions.

I can assure you that a yes no grid is not the most efficient way to do 99% of games.

This is a specific game type for which this grid is particularly efficient. Did you think I was suggesting it for ordering or conditional games? Look, I've been learning or teaching this test for five years, and I specialize in LG. I can point you to other games of this type if you're interested, and I can explain how to spot them when they come up.

I teach as well, so I don't need you to explain anything, and I don't see how on earth you can say using a grid is more efficient than simply plugging people into the places. You are without fail using more pen strokes to redraw a yes/no grid every time than you would be to build one diagram and just punch people into it vertically, as in not having to redraw anything at all. To call this "very inefficient" is ridiculous. Agree to disagree on technique, but to call how I attack games very inefficient isn't right.

[Reframe]

I teach as well, so I don't need you to explain anything, and I don't see how on earth you can say using a grid is more efficient than simply plugging people into the places. You are without fail using more pen strokes to redraw a yes/no grid every time than you would be to build one diagram and just punch people into it vertically, as in not having to redraw anything at all. To call this "very inefficient" is ridiculous. Agree to disagree on technique, but to call how I attack games very inefficient isn't right.

Sorry, but it is. Maybe you're just not clear on what game type we're dealing with here. Your diagram is going to fail to capture features going in the other direction and you will need to either write them down in a different manner or simply try and remember them. That just will lead to inefficiencies, especially when it comes to making inferences.

Perhaps you're able to complete these games without fully diagramming them. If so, that's a testament to your intelligence, or something. But I have had student after student come to me wondering how to treat them, just as the OP here was, and things almost always fall into place when we began using this sort of diagram. It's a judgment call as to whether that's worth the five extra seconds it takes to draw a couple lines.

[Clearly]

I teach as well, so I don't need you to explain anything, and I don't see how on earth you can say using a grid is more efficient than simply plugging people into the places. You are without fail using more pen strokes to redraw a yes/no grid every time than you would be to build one diagram and just punch people into it vertically, as in not having to redraw anything at all. To call this "very inefficient" is ridiculous. Agree to disagree on technique, but to call how I attack games very inefficient isn't right.

Sorry, but it is. Maybe you're just not clear on what game type we're dealing with here. Your diagram is going to fail to capture features going in the other direction and you will need to either write them down in a different manner or simply try and remember them. That just will lead to inefficiencies, especially when it comes to making inferences.

Perhaps you're able to complete these games without fully diagramming them. If so, that's a testament to your intelligence, or something. But I have had student after student come to me wondering how to treat them, just as the OP here was, and things almost always fall into place when we began using this sort of diagram. It's a judgment call as to whether that's worth the five extra seconds it takes to draw a couple lines.

There are ways to include all of the information that don't require redrawing your entire diagram every question. All I'm saying. Also, if you concede its a judgement call, how can you turn around and tell me I'm wrong?

[Reframe]

There are ways to include all of the information that don't require redrawing your entire diagram every question. All I'm saying. Also, if you concede its a judgement call, how can you turn around and tell me I'm wrong?

There are very few diagrams that should be redrawn every question. This particular diagram should probably be drawn only once a game. There are ways of representing "temporary" yeses and nos - for example, Y and N as opposed to checks and Xs, or smaller symbols, or something. Most of these are fairly intuitive.

When I said a judgment call, I meant on the student's part. The way the time spent on inferences versus the time spent on diagramming shakes out will be different for almost every person. My experience - and I've worked on this with dozens of people - has been that, for almost everybody, they more than make up the extra time diagramming in how quickly the inferences appear.

[Clearly]

There are ways to include all of the information that don't require redrawing your entire diagram every question. All I'm saying. Also, if you concede its a judgement call, how can you turn around and tell me I'm wrong?

There are very few diagrams that should be redrawn every question. This particular diagram should probably be drawn only once a game. There are ways of representing "temporary" yeses and nos - for example, Y and N as opposed to checks and Xs, or smaller symbols, or something. Most of these are fairly intuitive.

When I said a judgment call, I meant on the student's part. The way the time spent on inferences versus the time spent on diagramming shakes out will be different for almost every person. My experience - and I've worked on this with dozens of people - has been that, for almost everybody, they more than make up the extra time diagramming in how quickly the inferences appear.

But you don't think there's a way to draw all the information into a diagram that is faster to reproduce? There is. I'm just gonna let this die, I'm too tired for TLS beef right now, but don't assume the way you do things is by default the most efficient. I'm constantly discovering faster ways of doing things, you should too.

[Reframe]

But you don't think there's a way to draw all the information into a diagram that is faster to reproduce? There is. I'm just gonna let this die, I'm too tired for TLS beef right now, but don't assume the way you do things is by default the most efficient. I'm constantly discovering faster ways of doing things, you should too.

I'm not "assuming" it by "default". :/ As I said, I've worked with many, many students on this.

[bp shinners]

Sure. For example, I just went through a game where food trucks, lets call them abcdef, serves one or more of three buildings, xyz. The way I set up the game was using xyz and my base, in other words placing trucks to buildings. This made the game very difficult and when I tried it the other way, buildings to trucks, the game was much easier. I chose building initially because it seemed to make the most sense to me logically, I imagine trucks moving to different buildings, no the other way around, however, my logic seems wrong.

At Blueprint, we call this a Profiling game because nothing is stable - food trucks can go to more than one building, and the buildings can have any number of food trucks. So it's hard to know what to use as the groups.

For these, there are two rules of thumb:
1) Whichever set has the most rules associated with it as far as limiting sizes goes, make that the base. If you can make a setup based on this principle, it's going to be the most helpful.
-For this game, I believe that you get several rules limiting the number of buildings that some of the trucks can go to. Since that would place a limit if you made the trucks the group, that's what you should do.

2) Whichever set you are assigning things to in the most rules, make that the base. Go with this if the first principle doesn't help much.
-I forget which way this pushes you in this game, however
a) If most of the rules were along the lines of "Both buildings X and Y are served by the falafel truck", I'd use the trucks as my base
b) If most of the rules were along the lines of "Building X is served by both the falafel truck and the ice cream truck", I'd use the buildings as my base.

These games should lead to a ton of deductions that solve a majority of the slots before you head down to the questions (I think this is the Profiling game with the fewest deductions, and there are still a ton). The deductions come in one of two varieties: figuring out the size of the groups, and figuring out what's in the groups. The former is what you should pay attention to because they're the trickier deductions to spot, and they usually come from the rules that are along the lines of, "The ice cream truck and the salad truck serve exactly one building in common."