At some point, every LSAT student stops improving. You did the lessons, drilled the question types, took practice tests—and your score climbed. Then it stopped. You keep studying, but the number will not move. That is a plateau, and it is the most frustrating phase of LSAT prep.
Here is what is actually happening: the strategies that got you your initial gains were general. You learned what a flaw question is. You learned to identify conclusions. You learned basic conditional logic. Those skills lifted you from your diagnostic score to your current score. But general skills have a ceiling. To break through, you need question-family-specific methods—deliberate techniques targeted at the exact places you are losing points.
This is the exact inflection point where many of my students come in for tutoring — not because they lack knowledge, but because they need someone to look at their specific error patterns and prescribe the right fix. The breakdown below gives you the same diagnostic framework I use in sessions, organized by question family.
Below is a breakdown of every Logical Reasoning question family, the primary method for improving in each, and specific advice you can apply immediately.
Part I: Argument Structure Family
These questions test whether you can identify the building blocks of an argument.
Main Conclusion
Primary Method: Distinguish conclusions from subsidiary conclusions and premises.
Most Main Conclusion errors come from selecting a subsidiary conclusion—a statement that is supported by evidence but also supports a further claim. The main conclusion is the statement that everything else in the argument is ultimately trying to prove. When you miss these questions, go back and ask: "Does this statement support something else in the stimulus?" If yes, it is not the main conclusion. Practice by writing out argument maps with arrows showing which statements support which.
Argument Part (Role of a Statement)
Primary Method: Label every sentence in the stimulus before reading the answers.
Before looking at answer choices, go through the stimulus and label each sentence: premise, subsidiary conclusion, main conclusion, background, counterpoint, or concession. If you cannot label a sentence, that is a sign you do not fully understand the argument's structure—and that is exactly where the test will target you. The correct answer describes the role of the cited statement within that structure. If your labels are correct, the answer is obvious.
Part II: Argument Analysis Family
These questions test whether you can describe or match the reasoning technique an argument uses.
Method of Reasoning
Primary Method: Name the method before reading the answers.
Students plateau on Method of Reasoning because they try to evaluate all five answers against the stimulus simultaneously, which leads to confusion between similar-sounding descriptions. Instead, read the stimulus, name the technique in plain language ("the argument uses an analogy" or "the argument gives a counterexample"), and then find the answer that matches your description. If you cannot name the method, you need to study the common methods until you can recognize them on sight.
Parallel Reasoning
Primary Method: Abstract the argument's structure into variables.
Do not try to match content—match structure. Translate the stimulus into a formula: "All A are B. X is an A. Therefore, X is B." Then find the answer with the same formula. Common errors: matching the conclusion's tone (positive/negative) without checking the full structure, or matching the topic instead of the logic. Always check that the number of premises, the type of logic (conditional, causal, comparative), and the conclusion type all match.
Flaw
Primary Method: Identify the specific gap before reading the answers.
General flaw knowledge ("this is a bad argument") gets you to a mid-range score. To break through, you need to name the exact flaw—not just "bad reasoning" but "the argument confuses a sufficient condition for a necessary one" or "the argument treats a correlation as evidence of causation." Build a flaw vocabulary. When you review missed Flaw questions, write down the specific flaw name and add it to a running list. Over time, your pattern recognition will sharpen and you will pre-phrase the flaw before seeing the answers.
Parallel Flaw
Primary Method: Name the flaw first, then match structure.
Parallel Flaw combines the skills of Flaw questions and Parallel Reasoning questions. First, identify the specific flaw in the stimulus. Then find the answer choice that commits the same flaw with the same logical structure. Do not be distracted by surface similarities in topic or conclusion direction—focus on whether the flaw mechanism is identical.
Part III: "Change the Argument" Family
These questions ask you to make the argument stronger, weaker, or complete.
Strengthen and Weaken
Primary Method: Identify the argument's core assumption, then find what supports or attacks it.
If you have plateaued on Strengthen/Weaken, the problem is almost certainly that you are not precisely identifying the gap between the evidence and the conclusion. Vague gap identification leads to vague pre-phrases, which leads to choosing attractive-but-wrong answers. Force yourself to articulate the assumption in one sentence before reading the choices. For Strengthen: the correct answer makes that assumption more plausible. For Weaken: the correct answer makes that assumption less plausible.
Evaluate
Primary Method: Apply the variance test to each answer.
The correct answer on an Evaluate question is one where different answers to the question would change your assessment of the argument. Test each answer choice by asking: "If the answer to this question is yes, does it affect the argument? If no, does it affect the argument?" If both a yes and a no answer change the argument's strength, that is your answer. If only one direction matters (or neither), eliminate it.
Sufficient Assumption
Primary Method: Find what, if added to the premises, guarantees the conclusion.
Sufficient Assumption questions are more formal than Necessary Assumption questions. The correct answer, combined with the existing premises, makes the conclusion logically certain—not just plausible, but unavoidable. These often involve conditional logic. Diagram the premises, diagram the conclusion, and look for the missing link in the chain. The correct answer is the conditional statement that completes the chain from premises to conclusion.
Necessary Assumption
Primary Method: Pre-phrase + Negation Test.
Pre-phrase the assumption by identifying the gap (use the five common types: Linker, Defender, Feasibility, Representative Sample, No Reverse Causality). Then verify your answer with the Negation Test: negate the answer choice and check whether the argument falls apart. If the negation destroys the argument, you have found a necessary assumption. If the argument survives, that answer is wrong—even if it sounds helpful.
Part IV: Principle Family
These questions test whether you can match abstract rules to concrete cases (or vice versa).
Apply the Principle
Primary Method: Translate the principle into a conditional, then match it to the facts.
Read the principle and convert it into if-then form. Then check each answer to see which set of facts triggers the principle's conditions and leads to the principle's conclusion. Students plateau here because they try to match on "vibes" instead of structure. A principle that says "Anyone who knowingly benefits from harm should compensate the harmed party" becomes: "If knowingly benefits from harm → should compensate." Now find the answer where someone knowingly benefits from harm.
Generalize the Principle
Primary Method: Abstract the specific facts into a general rule, then match.
You are given a specific scenario and asked which general principle it best illustrates. The trap is selecting a principle that is true but does not match the scenario's logic. Extract the structural relationship from the stimulus—not the topic—and find the principle that captures that same relationship at a higher level of abstraction.
Part V: Inference and Conflict Family
These questions test whether you can determine what follows from the given information.
Most Strongly Supported (MSS) and Must Be True (MBT)
Primary Method: Stay close to the text and combine no more than two premises.
The most common plateau error on inference questions is going too far. The correct answer is usually a modest claim that follows directly from one or two statements in the stimulus. If you find yourself thinking "well, that could be true if you also assume...," you have gone too far. Stick to what the stimulus actually says. Also watch for scope traps: the correct answer will not introduce new concepts or make claims stronger than the evidence supports.
Must Be False (MBF)
Primary Method: Find the answer that directly contradicts a statement or combination of statements in the stimulus.
MBF questions are the inverse of MBT. The correct answer is the one that cannot be true given the information provided. Look for logical contradictions: if the stimulus says "all X are Y," the correct MBF answer might say "some X are not Y." Diagram conditional statements and look for direct conflicts.
Paradox (Resolve/Explain)
Primary Method: Categorize the type of tension, then predict the resolution category.
When you hit a plateau on Paradox questions, it is usually because you are reading all five answers and picking the one that "sounds right" instead of pre-phrasing. Read the two conflicting facts, name the type of tension (causal, numerical, behavioral, statistical, economic), and predict what kind of information would make both facts true simultaneously. Then find the answer that matches your prediction.
Agree/Disagree (Point at Issue)
Primary Method: Apply the Agree/Disagree test to each answer.
For each answer choice, ask: "Would Speaker 1 agree or disagree with this statement? Would Speaker 2 agree or disagree?" The correct answer is the one where the two speakers clearly take opposite positions. If either speaker's position on the statement is unclear or not addressed, eliminate that answer. Students plateau here by picking answers where only one speaker has a clear position—both must have a clear and opposing stance.
How to Integrate This Into Your Practice
Do not try to fix everything at once. Review your last three practice tests. Identify the two question types where you are losing the most points. Read the corresponding section above and implement the primary method for those two types over your next 100 practice questions. Track your accuracy. Once those types improve, move to the next two.
A plateau is not a wall. It is a signal that your general strategies have taken you as far as they can, and it is time for precision. The students who break through are the ones who stop doing more of the same and start doing something different — something targeted at the exact questions they are missing.
If you've applied these methods for two or three practice tests and your score still hasn't moved, the bottleneck might not be strategy — it might be diagnosis. You may be targeting the wrong question types, or misidentifying the actual error pattern. That's the kind of thing that's hard to see in your own work but obvious to an outside observer. A free diagnostic consultation can tell you exactly where you're bleeding points and what to fix first.