Stage I: By Type Practice

Download a preview of

LR Perfection Free Preview: Chapter 22 Part II

Now it’s time to put everything to practice. Attaining LR perfection can be achieved in three stages: 

Stage I: By Type Practice

As we have seen time and time again, different question types require different things from us. How we analyze the stimulus and what kind of answer choice we prefer are drastically different for different question types. 

So it’s absolutely crucial that we have a clear and discrete understanding of what is asked of us for each of the seventeen question types. For example, for SA Questions, the correct answer can often be anticipated, and we have a preference for a strongly worded answer choice. For NA Questions, on the other hand, while we approach the stimulus in the same way, the correct answer is harder to anticipate, and we have a preference for weaker answer choices. To test the validity of a SA answer, we plug it back into the stimulus, and see if this answer choice, when combined with the premises, can prove the validity of the conclusion. While to test a NA answer choice, we negate it and see if it can weaken the validity of the conclusion. 

If you are having trouble recalling the differences between each question type, it helps to make a one-page outline of what our job is for different question types. This was how I initially practiced. I would have this outline next to me as I practiced, constantly reminding myself of what I needed to do. When I have committed everything to memory, I threw away the outline. 

In other words, just seeing the question in an LR question should trigger a reaction in us. When I see a Weaken Question, for example, my immediate reaction is this: 

“Ok, Weaken Question. I will read for the argument in the stimulus, make sure I differentiate between the premises and the conclusion. Weaken answers can come in multiple forms. It’s best that I find an answer choice that attacks the gap. Second best if the answer choice uses outside information to independently question the truthfulness of the conclusion. If all else fails I can accept a premise attacker as well.” 

“Harder Weaken Questions can also have multiple ‘correct’ answers, so I need to rank the answers too. Make sure I find the answer choice that is most on point and most directly attacks the author’s reasoning.” 

Similarly, when I see a Flaw Question, even before I read the stimulus, my subconscious reaction is this: 

“Hmmmm, Flaw Question. Again, read the stimulus for argument, does the author commit one of those classic fallacies? I will also need to look for causal or conditional reasoning, does it exist and is it flawed? Flaw Questions are much easier if you can figure out what the flaw is before going on to the answer choices, so let me be extra careful reading the stimulus.” 

When I get to the answer choices, my first thought is this: 

“Flaw answers can be abstract, I gotta use the keywords here as clues as to what the heck they are saying. Also be careful of ‘assumes without warrant’ or ‘overlook’ type answer choices!”

In other words, I’m using the questions to mentally prepare myself for how to approach the stimulus and subsequent answer choices. 


Detailed Progression of By-Type Practice:

Nearly all the high scoring students will have practiced LR questions by type. But it’s not as simple as spending a few days on each question type and then moving on. 

When I was tutoring students whose first language was not English, so many additional unknown variables were involved.  I had to devise a more detailed, executable plan to get them to 170+. I had carefully recorded my by-type speed and accuracy rate by the time I was consistently scoring in the 175+ range. This was how I was performing on LR questions prior to my exam: 


Time Goal

Accuracy goal before moving to the next difficulty 

Level 1: Easiest



Level 2: Easier



Level 3: Normal



Level 4: Harder



Level 5: Hardest

2:00 – 3:00


LR Questions can be divided into five levels of ascending difficulty. Some commercial LSAT prep websites already do this for you. I would have my students practice different question types in segments of 10, starting with the easiest level. 

For example, if we are practicing Weaken Questions, I would have my students start with the easiest Weaken Questions. I would have them do 10 of these in a row. If they can get all of them correct, then we move to harder questions, so on and so forth. Once you can achieve a winning streak of 10 questions, it’s time to move to the next difficulty!

This way of practicing will determine where your plateau is. For example, if you are unable to consistently get 10 level 3 questions right in a row, then that’s where we stop and focus our efforts. The end goal is to get level 1-3 questions perfect, make one mistake every 20 questions or so in level 4 questions, and one mistake every ten questions in the hardest questions. 

Of course, you should always aim for perfection, but this is a quantifiable and measurable goal that you should aim for if your goal is 170+ or even 175+. 

By Type Timing

The most important thing to remember is that accuracy comes before timing. 

I’ll say it again, 


Only worry about speed when you are consistently getting questions right. 

Many people have said that you cannot learn the LSAT, which is blatantly false. But the LSAT is not a content-based test. The actual material that you will see on the exam will contain information which you have never seen before. So our job, when studying for the LSAT, is not to memorize and regurgitate information. 

This is where many students make a fatal mistake. I’ve had students come to me with hundreds of pages of notes. This doesn’t help at all. Think about it. On the actual exam you have on average a minute or two per question, you are stressed and extremely nervous. Your brain will be foggy and you will automatically revert to whatever habits you developed during PTs. How much of those hundreds of pages of information are you going to remember?

Our job, instead, is to train our way of thinking by developing good habits. Regardless of who you are, you will fall back on old habits on the actual exam. So the only thing we can do is to make sure that we have in our repertoire only those habits that can help us get the question right. 

For example, if we are consistently extracting abstract keywords from Role Question answer choices and asking ourselves what they are referring to during practice, then we will do the same thing on the exam. If we come up with the contrapositive of a conditional as soon as we find a conditional relationship in the stimulus, then we will do the same thing on the day of the test. The information contained in the stimulus and answer choices may be different, but our job is the same. 

Developing good habits involve breaking old habits and brainwashing ourselves into doing things differently. This takes time and repetition. The mind is a stubborn animal. If we tried to speed things up without fully developing these good habits, we are just setting ourselves up for failure down the road. This is the single most important reason why students plateau in the 150s, 160s, or even low 170s. 

So often we need to slow down in order to speed up. 

But when do we know that we are ready to focus on the timing aspect? We let data do the talking. For each level of difficulty, if you are consistently getting 100% correct, or 10 out of 10, that’s enough proof that you have what it takes to handle questions of this caliber. 

In the table above, I have also listed my average timing for each level of difficulty when I was scoring in the 175+ range. We know that a LSAT section is 35 minutes long, and will usually have 25 to 26 questions. So that’s roughly 1:45 per question. 

But not all LR questions are created equal. If you are spending 1:45 on the easiest question as well as the hardest question, something is seriously wrong. 

Instead, we should devote the majority of our time and attention to the harder and hardest questions. There were sections where I spent 25 minutes on the first 20 questions, but there were one or two questions which took over three minutes each. Remember, we are in pursuit of perfection here. 

Of course, if you attain this and there’s still time left before your exam, you should keep on going. After finishing writing this book and having gone through each individual question half a dozen times on average in the past two years, either tutoring students or just analyzing the questions on my own, I’m probably averaging a minute each for levels 1-4 questions (100% accuracy rate), and 2:00 for level 5 questions (95% accuracy rate). Yes, the returns are diminishing and I am still making mistakes and learning.

Differentiate Between Easier and Harder Questions

We have mentioned that commercial test prep websites will categorize LR questions in terms of difficulty for you, that makes our job a whole lot easier. But if for some reasons, you choose to go at it on your own, here is how to determine how to categorize a question in terms of its difficulty: 

There are six ways in which the test makers can make an LR question harder, five of which we have already seen: 

The five habits we have emphasized in this book, per our acronym SLAKR, are also five of the six areas of a LR question where the test makers can lay out traps to increase a question’s difficulty. Structurally, the test makers can deliberately word the stimulus in a way so that the main conclusion is hard to differentiate from the intermediate conclusion; logically, the stimulus may contain complex conditionals or even hybrid causal/conditional reasoning. The argument may have a hard to spot gap that needs to be filled by an assumption that we must provide; answer choices may contain subtle term shifts or vague terms that must be solved by keyword analysis; and there may be multiple attractive answer choices that we must compare and contrast. 

These are just some of the examples of how a difficult question may differ from an easier one. We have covered nearly all of the tricks and traps commonly seen in harder LR questions throughout the book. 

One final mark of harder questions is the discussion of esoteric topics and use of abstract language in the stimulus. Stimuli discussing scientific or philosophical ideas can be especially challenging. 

In my experience, level 4 questions usually have one or two of these traits, while level 5 (the hardest) questions usually have two or three of these. For instance, a level 4 Role Question may have a structural trap (conclusion placed in the middle of the stimulus, while the intermediate conclusion comes at the very end), as well as a correct answer choice that is stated in vague terms (2 out of 6). A level 5 Strengthen Question may have a really abstract stimulus topic, two answer choices that both strengthen the argument, and hidden causal logic. (3 out of 6)

Our goal is to be able to recognize a difficult question when it presents itself, and to treat it with the respect and attention it deserves.