Deciding when to create templates

[amkaras10]

I was just doing logic game #7 in the lesson 5 homework without a time limit, and despite a 5/6, when I reviewed the answers I realized I probably could have gotten a 6/6 5 minutes faster if I had created the three templates. It seemed so obvious when reviewing, but in earlier games I sometimes created templates when the book said you should do without, and vice-versa.

So my question is how can I quickly and easily identify when I should and should not template? I know that if I have a variable or variable set that can only go in one of 2 places, that can be a great hint, but it seems like that's not the only determining factor. I would greatly appreciate some advice! Thanks!

[ivanperez]

Hope this helps:



WHEN TO DEPLOY THE IDENTIFY THE TEMPLATES TECHNIQUE:

1) UNWIELDY BLOCKS WITH LIMITED SLOTS



XY

slots available:

_ _ _ _ _



This is an advantageous time when to generate and identify all templates.

REMEMBER: It will take some time to decipher each template, but you will get through the questions in a breeze. You will get highly paid for your investment.

If you think about it, you only have 4 possible templates that can be generated from such scenario. In the first scenario, place the XY block in slots 1 and 2 (respectively).

Continue to shift the XY block into the succeeding slots until you reach a "max", where the Y touches the last slot.

WARNING: DO NOT EMPLOY THIS TECHNIQUE OF IDENTIFY TEMPLATES WHEN YOU HAVE MORE THAN 4-6 POSSIBILITIES BECAUSE IT WILL BE COUNTERPRODUCTIVE IN TERMS OF TIME.

Also, when you have a split block such as: XY_Z and a limited number of slots, employ the technique of templates as well. Any block in a linear game gravitates towards this technique because there are usually no more than 8 slots/variables per possibility.

**OTHER TIMES WHEN TO DEPLOY IDENTIFY THE TEMPLATES:

2) SPLIT-DUAL OPTIONS

I've also noticed that in games where you have something as the following is an advantageous time when to implement the Template technique:

X/Y _ _ _ _ Y/Z

In this scenario, you only have two possibilities:


1) X _ _ _ _ Y

2) Y _ _ _ _ X

X or Y MUST be First or Last.



You can then start filling in each of the empty slots according to the conditions of each of the two templates. The dynamics of each will be different, since you have to accommodate the remaining variables based on the dynamics of the original rules.


Sometimes, you will realize that EACH of the two templates will often break into 2 or 3 more templates so you might end up having 4-6 templates total. The more you slots you fill in for each template the quicker you will get through the Local Questions because all you have to do is locate a template compatible with the local conditions given and there you have your holy grail.

Also note: You will come to realize often times that when you start splitting templates from your original templates, SOME OF THEM WILL INHERENTLY NOT WORK; MEANING THAT WILL BE IN VIOLATION OF ONE OR MORE OF THE CONDITIONS GIVEN IN THE GAME. JUST REMEMBER TO CROSS EACH UNWORKABLE TEMPLATE AS YOU GENERATE TEMPLATES FROM OTHER TEMPLATES.

Also Remember that you might generate inferences for all slots in some templates but not in others. The Rules will dictate how many inferences you can make based on the conditions of each template. It is very likely that the local questions reference partially-full templates since those are the ones that are usually of interest to test makers (since they want you make deeper inferences).

I will highly recommend you generate each template parallel to each other so you can discern what MUST BE TRUE (meaning what variable occupies a particular slot in ALL TEMPLATE POSSIBILITIES.) You can then circle the variable to let yourself know that IT MUST BE TRUE IN ALL CASES; you will most likely be tested in one of the global questions what must be true.




REMEMBER THAT THERE IS A CLEAR CUT LINE THAT DIVIDES WHEN TO USE THIS TECHNIQUE AND WHEN NOT TO.

OBVIOUSLY, THE CLEAR CUT LINE IS SOMEWHERE AROUND 8 POSSIBLE TEMPLATES. IF IT IS MORE THAN THIS, THE POSSIBILITIES WILL BE TOO MANY TO POSSIBLY GENERATE. REMEMBER THAT SOME GAMES HAVE MORE THAN 50 POSSIBILITIES, MAKING IT UNPRACTICAL TO USE THIS TECHNIQUE. MORE OFTEN THAN NOT, IT WILL BECOME EXPLICITLY OBVIOUS WHEN IT IS UNPRACTICAL TO DO SO.

*USE YOUR INTUITION WHEN DIAGRAMMING THE MASTER DIAGRAM AND THE RULES AS TO WHETHER OR NOT IT WILL BE WISE TO USE THIS TECHNIQUE.

REMEMBER THAT THIS TECHNIQUE IS A DOUBLE EDGE SWORD. IF YOU START DIAGRAMMING ALL THE TEMPLATES AND REALIZE THAT THERE ARE ACTUALLY MORE THAN 8 POSSIBILITIES, YOU ARE DOOMED IN TERMS OF TIME.

IAP
RICE UNIVERSITY

[Jon Denning]

Hey amkaras and ivan,

Thanks so much for the two posts, and welcome to the Forum! The idea of creating templates or showing limited solution sets for games is a really interesting topic, and I always enjoy the conversation! So I'm happy to see it being discussed here.

As a PowerScore instructor on a PowerScore Forum (and speaking to a PowerScore student), I want to chime in briefly to add some thoughts of my own and give you my take.

First, I should point out that as an attendee of a PS class at lesson 5 you're only just beginning your exploration of these concepts. There's an entire lesson of games later in your course that focuses exclusively on templates/limited sets, so if there's some confusion presently about when and how to use them that's totally understandable. A very comprehensive analysis is coming, and will undoubtedly provide more depth and insight than I could give here. So be patient

Second, and again, I love hearing non-PS people's, or new-to-PS people's, take on templates (and games in general, really). A variety of perspectives, provided they're not intrinsically/diametrically opposed, can often be a valuable resource! And Ivan certainly offers some great advice here, particularly in showing some examples where templates might indeed be a worthwhile approach and the elements therein that jumpstart that thought process! So good stuff!

That said, I'd caution anyone to avoid too many absolutes when it comes to their LSAT approach...the test makers are just too good at creating exceptions for a laundry list of strategies employing "always do X" or "never do Y" to be entirely reliable. People looking for universal formulas often find themselves struggling with the adaptability and logical dexterity this test demands.

With templates in mind, then, I don't know that I'd ever go so far as to say there's always a time or scenario when you'll use them--"if you see a certain construction you 'must' [or even 'should'] employ templates!" is just stronger than I'm comfortable with for a test like this. Similarly, suggesting that more than X number of templates makes it not worth it is probably true if that X is clearly going to be quite large (like a dozen+, say, or at least 7-8 extremely involved diagrams...), but I'd happily sketch out 8-10 quick sequences if that represents every possible order in a game and I can recognize that's going to be the case. The point: careful with absolutes as a general rule, but particularly when it comes to this topic!

Instead, let's talk about what tends to allow and disallow templates, or at least creates high likelihoods of use or avoidance.

Before I ever begin to attempt the use of templates, or showing the various hypotheticals that could occur, I look for two criteria to be satisfied:

1. There needs to be some fundamental dividing line(s) in the game that would send the possibilities (general templates or specific solutions) down a limited number of paths. So a block that has only a handful of placement options, or a numerical distribution that can only occur in a few ways, or split options for certain variables or spaces, or a select few groupings that must be selected...each of these, and many others, serve to channel the game in a powerfully restricted type of way. Recognizing the existence of this limiting factor is step 1, and interestingly enough the majority of games you'll see tend to have it! But it's not enough...

2. You also need to realize what's going to happen as you begin exploring (listing out) each/all of those pathways. And what needs to happen, is that as you journey along each further restrictions and inferences MUST come into play! For instance, let's say I have a grouping game where of the six people selected, exactly 2 out of A, B, and C must always be selected. Based on this we've satisfied our first criteria above: there are now just three main options for this game, AB, AC, and BC as your 2/3 groups. Each of those three represents a possible template, but they may not be sufficient to warrant templates. That decision depends on what happens as you consider each: does the group AB significantly affect the other variables/spaces based on the rules, or does your "journey" end at AB? If they're not influential templates may not be worth your time; if they are highly connected to the remaining variables and begin to fill the four open spots, templates are much more likely to pay off.

Which is fine and good, but here's the truth: I think most people have a fairly consistent standard by which "now's the time for templates!" becomes an obvious attack plan to consider (the two points above, with practice, aren't hard to measure). What gets particularly intriguing is where different people opt out. That is, where people make the decision that templates, even if possible, are NOT worth it. I bring this up mostly in reference to something Ivan mentions, that "templates are an investment." How, or whether, that investment is going to pay off is often a very subjective decision: 7-8 minutes setting up templates might be well worth it, or it might be disastrous, depending on the power of the setups and your facility with them. Showing 8 different templates might only take a few minutes, or it might be the only way you happen to be able to get through the game and succeed in the questions...or it might take so long and provide such a poor return on that investment that it ruins the whole section! So you have to be self-aware at every moment and make informed decisions about how you're going to potentially attempt to show templates, how much time that's likely to take, and how beneficial they are likely to be once created. A lot of moving parts, moving differently for pretty much everyone.

I know that's not a "hard and fast" type formula, but that's only because there really isn't one. It varies with the student, the game, and even the section as a whole (last game with 4 minutes left, or last game with 14?, for instance).

Finally, my colleague Ron wrote a fantastic blog article on these exact ideas, and I think you'll find it to be of great value: http://blog.powerscore.com/lsat/bid/317 ... -Set-Games

I hope that helps and please weigh in with additional thoughts or questions!

Jon