Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

From my experience: The Four Pillars of LG Consistency

BinghamtonDaveBinghamtonDave Alum Member 🍌🍌
in General 8684 karma

I started studying for the LSAT barely able to answer a single question on games. With no training in logic, I essentially had to start from the ground up. I credit 7Sage with allowing me giving me the tools become competent in games. I have been tutoring for about 8 months now and have been working on a system that I try to impart onto my students. Any parallels between the following steps and the advice that others have dispensed is purely coincidental. None of this is reinventing the wheel here, but rather some thoughts I have on LG consistency.

I have titled this thread "The Four Pillars of LG Consistency" for a reason. Notice that I didn't title this thread: "The Four Pillars of LG success." This is because 1.the definition of "success" will vary slightly from person to person, 2. Success has more elements to it than consistency. Consistency is a part of success. This post is better able to answer the question: "how can I keep my sections from being riddled with errors?" rather than "how can I get a -0?" The "pillars" are what emerged for me after doing over 2,000 games and teaching LG for several months. The general thread that runs through each aspect of the pillars is try to eliminate any "sloppy" errors and 2. knowing the exam well enough to be confident in our approach. Zooming out, what we are trying to do is build competency and eliminate any "mystery" the exam might present. Lets get started:

1.Develop your system whatever your "system" is, develop it and keep it rigorously consistent. Do you write your game pieces in the upper left hand corner each time? Then keep doing that. Do you write your gameboard to the right and your rules to the left? Great, keep doing that. But go deeper, make sure you are making "tick marks" next to your rules each time you write one down. Write your game pieces twice separated by a line: one set you never touch and the other you can scribble out to see what piece is left or what piece is the "floater." Have constant checks that you have properly translated rules. After you have translated all of the rules start from the last rule and read up to the first rule, each time checking that you have translated the rule correct. Read and extract precisely what each rule is telling you. The slogan here with check is: pay now or pay double later. If we made a mistake in translating a rule, we want to be in a position to catch it with our 20 second check rather than midway through the game (we all know that feeling :( ) when you have burnt 2 minutes. Have built in fail-safes. What also goes along with this first pillar is: be neat in your writing, what you are doing is worth money in scholarships.
I will tell a quick story here. When I first started studying for game I spent 6 months and did 1,400 games. I thought I knew what I was doing, I thought I would have that coveted -0. When I did my first timed section, I missed 7. Partially because I drew a "V" that looked like a "U," thereby messing up a rule and ruining my score. All of this stuff has to be solved item by item. For games we are looking to extract three basic things: the rules, our pieces and our gameboard. This information will often be partially contained within the "stimulus" (the block of text to begin a game before our rules.) I cannot understate how important this step it. I have calculated that for us to be consistent on an LG section between 16 and 19 things have to go well. We have to make sure we are translating the rules correctly, we have to make sure we have a neat and coherent system etc. But for things to go bad, just one or two of those things have to fail.

  1. Finding inferences, "the other side of the coin" and asking the right questions. so what is an "inference" on LG? An inference in its most appealing form is essentially a new rule. I say this because many times we will find "mini inferences" that aren't full fledged new rules but are restrictions that are helpful. An inference is a combination of the inherent constrictions of the gameboard plus a rule plus another rule in combination to create a new rule/restriction. An inference will allow us the ability to play the game essentially with a cheat code. An extra rule in our approach that we can use to our advantage. For my Survivor fans out there an inference on an LG game is essentially a hidden advantage! An inference can turn a 10 minute game into an 8 minute game. So now that we have generally defined an inference, how do we find them?

There are several ways to find the inference of a game. The most obvious is to focus on rules in which a piece is mentioned more than once. So if rules 1 and 3 both mention piece "A" we should be asking questions about what these rules mean in combination with each other along with the inherent constirctions of the gameboard (things like we are choosing 5 out of the 8 available in an in-and-out game etc.) Asking questions is a very helpful way to help us find the inference. So if a rule for a grouping game says something like: "piece A must be with exactly one other piece," we should be asking the question: who could that piece be? If in a five space sequencing game we have three pieces that are "followers" we should be asking the question: who can be first?" I believe that essential to finding an inference is asking the right questions.

What can aid us in asking the right questions? That is where the concept of the other side of the coin comes in. A coin has two different sides that express the same thing, we could be looking at the "heads" or the "tails" of a penny and those of us familiar with the penny will know that it is a penny and it has this particular value. Rules on LG are often presented to us in "the positive" or "the negative." By trying to find the other side of the coin we can help ourselves possibly find the inference. So for instance if we have a three group grouping game and piece A cannot go in the tennis group, by telling us that piece A cannot go into the tennis group the other side of the coin for that rule is: A must go in the volleyball or soccer group. It is the same thing stated in a different way, except now we have "the positive" of the rule or "the negative." You probably already do this, but often times the other side of the coin with rules is valuable information. It becomes valuable because questions will often be predicated on us knowing the other side of the coin. Take the aforementioned example. Piece A cannot be in the tennis group, which means piece A must be in the volleyball or soccer group. Another rule might tell us that a piece that piece A must stay away from piece B and a question might tell us that piece B is in the soccer group, which means, since we know the other side of the coin, that piece A must be in the volleyball group.

  1. Knowing The Questions The questions is divided into two steps, the first is to know automatically what the questions are asking. What is a CBT EXCEPT? What is a MBT? what are the wrong answers to those questions? How do the wrong answers show up? So for instance a MBT question the wrong answers are either MBF options or CBT options. One must be efficient with their knowledge of what the questions are asking. This piece has to be in place.

The other aspect of this step is building on an analogy I use quite a bit with my students. That is where do we start our analysis of the question? The analogy I like to use here is that if the car won't start before we replace the transmission, let's make sure that the car is not just out of gas. This is a variation on Occam's Razor. We start with the least complicated solution-a $5-$10 quick solution- rather than a $700-$900 complicated and long solution. In LG we can often eliminate an answer choice or two from a question by a mere application of our rules. Some choices are never going to work no matter what "If" condition they give us in the question stem is. By starting our analysis here we can possibly eliminate all the wrong answer choices quickly or efficiently or we can eliminate enough o make our "testing" of the available options quicker and more efficient. Take for instance PT 11 game 1 question 2. If we start our analysis of this question by first eliminating what could never be true with a mere straightforward application of our rules, we know that answer choices (A) and (B) violate rule 5 and answer choices (C) and (D) violate rule 4. We essentially do not have to do much work here because we have worked simple to more complex and we eliminated all of the incorrect answer choices on the basis of "checking the gas" i.e. a mere application of our rules. Always try to run answer choices against the rules or your inference or mini inference unless you know the answer from a sub-game board or split etc. We maximize our chances of getting the question correct in many instances.

Knowing the questions also comes down to knowing what to look for when they ask you a straightforward MBT: this is more than likely (but not 100% of the time according that my knowledge) a point that they are rewarding you for finding the inference.
Knowing the questions also comes down to counting the "steps" away something like an "If____, then what MBT?" question form takes. +90% of the time the answer to these questions will be between 1 and 4 steps away from the condition they gave us. So for instance if the if condition they gave us tells us to place something somewhere, we ask ourselves what does that trigger (by consulting our rules that are neatly placed next to our gameboard) that should trigger something, that is 1 "step" away. That thing should trigger something else according to our rules and pieces: that thing is 2 "steps" away. one can effectively predict the answer to these questions by knowing the amount of "steps away" something is from what they have asked us. The LSAT will rarely ask us for 1 step away and will rarely ask us for 4 steps away. Instead, they will be looking for 2-3 steps away most of the time.
LG has stayed remarkably consistent for the better part of 30 years.

4.Knowing what you know and executing that knowledge The last pillar here is the actual application of your knowledge to new games and timed exams. It is one thing to know how to dig deep into the games, it is another to be so comfortable with being able to do that that you can apply that knowledge to new games in a concise and coherent fashion. This is where a focused process of drilling comes into play. When you drill, look at all of the above pillars, consciously apply your system to the game, do each game several times, you will start to see the patterns upon which the games are built.

Feel free to reach out with any questions.


  • Beast ModeBeast Mode Monthly Member
    837 karma

    Thank you so much for this! It's very helpful.

  • BumblebeeBumblebee Member
    640 karma

    Wow. This is gold! Thank you so much for your amazing advice.

  • aussie_zacaussie_zac Alum Member
    90 karma

    This is amazing. You are a legend.

  • BlindReviewerBlindReviewer Alum Member
    855 karma

    @BinghamtonDave thanks for this -- super helpful!

    I had a question about your point on MBTs usually rewarding you for making key inferences. So I've usually been following JY's method of doing the "If" questions first and then coming back to the naked ones. However, I've wondered if it might have been more useful to spend that time up front knocking out the MBT question (if I didn't have the inference figured out already) because that might help me with the other questions. What are your thoughts on this?

  • Hopeful9812Hopeful9812 Member
    872 karma

    Thanks so much for this advice! Very helpful!!

  • BinghamtonDaveBinghamtonDave Alum Member 🍌🍌
    edited June 2019 8684 karma

    @BlindReviewer I also aim to prioritize the "if" questions, but could probably be more strict with this. My point about the straightforward MBT questions (that don't contain an "If" in the answer choices like PT 29 game 1 question 6) is that, let's think about how a straightforward MBT could possibly appear: it is either 1. something we have discovered MBT across all possible worlds meaning we have every single possible iteration in front of us and there is one piece that stays the same throughout all possibilities or 2. they are asking us for an inference we made previously, hopefully in our set up, because an inference can take the form of a new rule. I try not to speak in absolutes, but I am unaware of a straight forward MBT question asked on game history without an "If" in the answer choices that does not reward us for making the inference or some variation on the inference. which can be found in those two ways. this is the nature of a straightforward MBT.

    so now that we know how the answer to the question they ask normally shows up: from PT 7 game 3 through PT 72 game 4 (where if we find the inference, essentially every question asks us about the inference, 2 questions almost straightforwardly: what is the inference? and Please apply the inference.) The LSAT has (in my experience) stayed remarkably consistent in this throughout preptest history. Now that we have this knowledge, the question becomes: how do we find the inference? I believe at least a partial answer is contained above.

    As far as using the "if" questions to find a possible inference, that would work as a fail-safe for sure. As long as we can hone in on that piece that appears to be an inference and then try to falsify it and when we fail to, we would call that the inference.

    Edit: additional info

Sign In or Register to comment.