scenarios vs brute force approach to questions

dubnation

does anyone have any guidance for determining when to draw out scenarios vs brute force the answer choices? sometimes it's pretty clear that a new local rule in a question creates two scenarios and in those cases it's obvious that you should solve for each board before attacking the questions.

however there are also cases where it's less clear whether it's more efficient to break board into scenarios or find the correct answer by brute forcing answer choices. Here's an example: http://7sage.com/lsat_explanations/lsat-49-section-1-game-1/

On question 3, JY goes directly to the answer choices after deducing that there must be two IN blocks. He could instead have attempted to place those blocks and would have realized there are only two placements _ IN IN_ or _ IN _ IN . Then, after deducing those two scenarios, he could have gone to the AC's. I find that JY has a tendency to go to the AC's in these cases.

7sagelsatstudent180

Quick scenarios is the best way to understand a game. Do a few quick examples of how rules work to get an idea. This approach has savede a lot of time in the back end and largely the reason why I have improved at logic games

Pacifico

There are very few questions where brute forcing is required... If you feel like you need to do this a lot then you are probably not making enough inferences up front... For LG there is really no substitute for practice to teach you the intuitive logic/reasoning skills that you need to master... Outside of the weird games there are relatively few possible game types and the more you practice them the more you will see similarities across several different games and can start to pick out patterns... As an example if you have an empty master game board but then encounter a plain MBT question you should know you missed an inference up front... Another thing you can develop a feel for is the importance of floaters in determining outcomes in games... It's always best to try to make as many inferences as possible up front, but whether you translate that into multiple game boards is up to you... I used to go heavy on splitting and have backed off a bit to try and find the best middle ground for myself... You need to get a good 100 unique LGs under your belt to really get a good idea of the approach you need to take... But as I said before, if you are brute forcing more than a couple times per PT it is very likely that you're missing key inferences...

dubnation

Thanks for sharing your thoughts. I don't feel that I'm doing that a lot, nor do I feel that I miss many key inferences near the beginning. I've done every single logic game and at this point I'm trying to figure out tricks to enhance my speed.

The benefit of JY's approach is that if you have good intuition you can quickly scan the AC's and selectively test the ones you feel might be the right AC, which can save you time. However, you leave yourself open to being totally wrong, which obviously results in more time than doing the scenarios.

I was simply wondering if there were any guiding principles or patterns you guys have picked up for determining when it might be more efficient to create a couple mini scenarios based on the new local rule vs going directly to the AC's.

Pacifico

There are no real tricks, and if you've done every LG prior to PT 36 then you should have developed an intuitive feel for this kind of thing by now... There are basically two types of games: empty or almost empty game board games where you can make few or no inferences (question driven games) and partially filled or filled game board games (rule/inference driven games where almost all the work happens up front). The first kind rely on your ability to understand that there are many possible worlds and a quick look at the questions will show you if it will be a question driven game if you see several (if X is...) question stems. In these types of games, just move on to the questions, sketch the game board quickly but carefully and make inferences as needed. In almost all cases you should not have to sketch multiple scenarios for "if X is" questions. These types of games can require brute forcing to expose second/third order inferences of a more obscure nature than simply "X must be in" or "Y must be second" such as in the ruby/sapphire/topaz game. For rule/inference driven games, your ability to split game boards effectively or work intuitively with one master with many possible worlds is what will make or break you. Since most of the work should have been done up front in these games I would say that brute forcing should be a much more rare occurrence in these games. In general, if the question stem asks about a certain situation then brute forcing should always be the absolute last resort and more commonly it should not be necessary since they are telling you what world or worlds you could be in.

7sagelsatstudent180

One thing that hurts score in PTs and real life LSAT logic games is panicking. It has cost me and many other before and is the reason I canceled my last LSAT. But the reality is there are no new LSAT logic game types. Everything is a variation of another game. Once you get that mindset and understand it in reality, you will be fine