LSAT Logical Reasoning Tips: Strategy Guide

The best LSAT logical reasoning tips start with structure: read every argument for its conclusion and the support beneath it, learn each question type cold, then review every miss until you can explain why each wrong answer fails. Logical Reasoning is now two of three scored sections, so this work compounds.

By Ali Unar, Georgetown Law, Founder of Lovare Institut. Ali Unar is a Georgetown Law student and the founder of Lovare Institut, an admissions and LSAT-prep practice built on primary-source law school data and more than 5,000 real application materials.

Verified August 2026.

How much of the LSAT is logical reasoning?

Logical Reasoning is a lot of the LSAT. According to LSAC, since August 2024 the test has three scored sections, two of them Logical Reasoning and one Reading Comprehension. That makes LSAT logical reasoning roughly two thirds of your scored sections, the single largest driver of your score.

The practical consequence is a study priority. If Logical Reasoning is two of the three scored sections, then any improvement to your logical reasoning accuracy is effectively counted twice on test day, while an equal improvement in reading comprehension is counted once. When you plan your weeks, give Logical Reasoning the majority of your hours. It is also the most learnable of the scored material, because the question types and the flawed reasoning patterns repeat in predictable ways from test to test, so deliberate practice pays off reliably.

According to LSAC, since August 2024 the LSAT has four 35 minute multiple choice sections, three scored and one unscored, and Analytical Reasoning, the old Logic Games, was removed, with the last Logic Games test administered in June 2024. A separate unscored Argumentative Writing sample is completed on its own. Here is the current structure of the LSAT.

SectionCountScoredLength eachLogical ReasoningTwoYes35 minutesReading ComprehensionOneYes35 minutesVariable (pretest)OneNo35 minutesArgumentative WritingSeparateNoCompleted separately

Because one of the four multiple choice sections is an unscored variable section that pilots future questions, you will not know during the test which Logical Reasoning section counts and which might be experimental. Treat every section as if it is scored. That mindset keeps your effort even and removes a distraction you cannot resolve in the moment.

How is the LSAT scored, and why does logical reasoning weigh so much?

According to LSAC, the LSAT is scored from 120 to 180. Your raw score is simply the number of questions you answer correctly, with no penalty for wrong answers, then converted to the 120 to 180 scale through statistical equating so a given score means the same across different test forms.

Two facts from that scoring method should change how you take the LSAT. First, because there is no penalty for a wrong answer, you should never leave a Logical Reasoning question blank. Fill in a guess on every question, including the ones you flag and never return to. Second, because scoring counts correct answers across the scored sections, and Logical Reasoning supplies two of the three scored sections, your logical reasoning accuracy has an outsized effect. According to LSAC, the raw to scaled conversion is set by equating and LSAC does not publish a single fixed table, since the conversion varies by test form. That is why you cannot memorize a raw to scaled chart, and why consistent accuracy, not a magic number of questions, is the goal.

How do you read a logical reasoning argument?

To read an LSAT logical reasoning argument, find the conclusion first, then identify the premises that support it, then name the gap between them. Most stimuli are short arguments. Once you can state the conclusion in your own words and see the assumption bridging the support, the questions get much easier.

The conclusion is the claim the author wants you to accept. It is not always the last sentence, and it is often signaled by words such as therefore, thus, so, or hence, or hidden in the middle of the stimulus. The premises are the reasons offered. The assumption is the unstated idea the author needs in order to move from those premises to that conclusion. Train yourself to pause after reading a stimulus and articulate three things before you look at the answers: what the author concludes, what support is given, and where the reasoning could break. That three part read is the single most transferable Logical Reasoning habit, because nearly every question type is asking you to do something with that gap.

Avoid reading passively. Many test takers read the stimulus like a paragraph to absorb, then read all five choices hoping one feels right. That approach is slow and error prone on Logical Reasoning. Read actively, form your own prediction of what the answer should do, and then match it to a choice. A prediction protects you from the well written wrong answers that are designed to sound reasonable.

What are the LR question types?

LR question types fall into three families. The assumption family asks you to find, strengthen, weaken, or attack a gap: assumption, strengthen, weaken, flaw, and evaluate. The inference family asks what must follow. The structure family asks how an argument works: main point, method, role, parallel reasoning, principle, and paradox.

Reading the question stem first tells you what task you are being asked to perform, which shapes how you read the stimulus. The table below lays out the recurring LR question types, what each stem is really asking, and the core approach that works for it. Learning to recognize the type from the stem is worth real points, because the same stimulus can support very different correct answers depending on the task.

Question typeWhat the stem asksCore approachNecessary assumptionAn assumption the argument depends onNegate the choice; if the argument collapses, it is requiredSufficient assumptionA premise strong enough to guarantee the conclusionBuild the missing link that forces the conclusion using formal logicStrengthenSupport that makes the conclusion more likelyProtect the gap or affirm the author's assumptionWeakenSupport that makes the conclusion less likelyAttack the gap between premise and conclusionFlawThe error in the reasoningName the flaw in the abstract before reading choicesEvaluateA question whose answer would test the argumentFind the relevant point that swings the conclusion either wayMust be true (inference)What follows from the statementsCombine the statements; reject anything not supportedMost strongly supportedWhat the statements best suggestPick the safest, most supported claimMain pointThe author's overall conclusionState the conclusion in your own words firstMethod of reasoningHow the argument is builtDescribe the structure abstractly, not the topicRole or functionThe job a statement doesAsk whether it is a premise, conclusion, or counterpointParallel reasoningThe choice matching the argument's structureMatch logical form, conclusion type, and validityParallel flawThe choice repeating the same errorName the flaw, then find the same flawPrincipleA rule that fits the case, or a case that fits the ruleMatch the general rule to the specific situationPoint at issueWhat two speakers disagree aboutFind a claim one affirms and the other deniesParadox or resolveA fact that explains an apparent conflictFind the choice that lets both facts be true

Do not try to master all of these at once. Pick one type, do a focused set of that type across many stimuli, and study the pattern until you can predict the answer before reading choices. Then move to the next. Studying by type builds the pattern recognition that raw, mixed practice alone develops far more slowly.

What are the hardest LR question types?

The hardest LR question types are usually parallel reasoning, parallel flaw, sufficient assumption, and abstract method or principle questions. Parallel questions force you to match an argument's structure while ignoring its topic. Sufficient assumption questions reward formal logic. Each rewards deliberate, structural practice rather than intuition, so drill them by type.

Parallel reasoning and parallel flaw questions are slow because you must abstract one argument into its logical skeleton, then test five other arguments against that skeleton. The trick is to characterize the original argument in general terms first: what kind of conclusion does it reach, is the reasoning valid or flawed, and what is the logical shape. Match those features, and eliminate choices that differ in conclusion type or validity even when the subject matter feels similar. Sufficient assumption questions are hard because they demand formal precision. You are looking for a statement that, added to the premises, makes the conclusion follow with certainty, so translating the argument into conditional statements and finding the missing link is usually faster than intuition. Principle questions that operate at a high level of abstraction reward the same discipline: state the rule plainly, then apply it.

How do you spot common LR flaws and wrong answer traps?

Spotting LR flaws means recognizing recurring patterns: confusing correlation with causation, mistaking a necessary condition for a sufficient one, attacking the person instead of the argument, and generalizing from a biased sample. Wrong answers repeat too: out of scope, too extreme, the reverse of what you need, or half right.

The classic flaws show up again and again on Logical Reasoning, so learning them by name accelerates everything. Common ones include causal flaws that leap from correlation to cause, conditional flaws that confuse sufficient and necessary conditions, sampling flaws that generalize from an unrepresentative group, equivocation that shifts the meaning of a term mid argument, circular reasoning that assumes the conclusion, ad hominem attacks on the source rather than the claim, and false dilemmas that pretend only two options exist. When you can name the flaw, flaw questions become fast, and strengthen and weaken questions become easier because you know exactly which gap to defend or attack.

Wrong answer types repeat just as reliably. A choice can be out of scope by introducing an idea the argument never addresses, too extreme by using absolute language the argument cannot support, a reversal that says the opposite of what the task requires, or half right by starting well and then adding a clause that ruins it. In review, label not only why the credited answer is right but also why each wrong answer is wrong, sorting it into one of these trap categories. Naming traps trains you to spot them at speed.

How do you improve LR timing?

To improve LR timing, make your process automatic so reading and answering are one motion, then work in two passes: answer the questions you see fast, flag the slow ones, and return with leftover time. Do not reread whole stimuli. Time full sections often so 35 minutes feels normal.

LSAC does not publish an exact count of scored questions, so there is no official per section number to pace against. A common non-official estimate puts the whole test at roughly 75 to 76 scored questions across the three scored sections, which would place a couple dozen questions in each 35 minute Logical Reasoning section. Treat that only as a planning benchmark, not an official figure. The point is that you cannot afford to stall. If a question is not resolving, mark it, guess, and move on, because a hard question and an easy question are worth the same single point. Most timing problems are not reading speed problems, they are decision problems: students reread because they did not commit to a prediction the first time. Build the prediction habit and the clock takes care of itself.

Practice with a real clock from early on, not just at the end of your preparation. Doing untimed drills forever teaches you to solve questions in a way that does not survive test conditions. Alternate: learn a type untimed, then immediately do a timed set of that type so the method holds up under pressure.

How do you review logical reasoning to actually improve?

The review process matters more than raw volume. After a timed set, do a blind review: redo every question you were unsure of without a clock, commit to an answer, then check. For each miss, name the question type and the trap, and log it. Patterns in your log become your drills.

Blind review is the highest leverage habit in LSAT preparation, and it applies directly to Logical Reasoning. During a timed section, mark every question you are not fully certain about. Before you look at the answer key, go back through those marked questions with no time limit, rethink each one, and settle on a final answer you can defend. Only then check the key. The gap between your timed answers and your untimed answers tells you how much of your miss rate is careless speed error versus genuine gaps in understanding. Careless errors get fixed with process and pacing. Real gaps get fixed with targeted study of that question type.

Keep an error log with a row for every miss: the question type, the flaw or trap involved, why you eliminated the right answer, and why the wrong answer tempted you. Review the log weekly and let it choose your next drills. Over a few weeks, your log turns from a scattered list into a short set of specific weaknesses you can attack directly. That feedback loop, not endless fresh questions, is what moves a Logical Reasoning score.

How do conditional logic and diagramming help on LR?

Conditional logic helps on a real share of LR questions. Learn to translate if then statements into arrows, take the contrapositive, and handle quantifiers like some, most, and all. You do not need to diagram everything. Reserve notation for sufficient assumption, principle, and formal questions where the logic is dense.

A conditional statement says that if one thing is true, another must be. The contrapositive reverses and negates it and is always equally true, which is why it is such a reliable tool: if A then B means if not B then not A. Sufficient and necessary conditions trip up many test takers, and a large share of Logical Reasoning flaws come from confusing the two, so getting comfortable with arrows and contrapositives pays off across strengthen, weaken, flaw, and assumption questions, not just formal ones. Quantifiers add another layer. Knowing how some, most, and all combine, and which inferences are valid when two quantified statements share a term, resolves a category of inference questions that otherwise feel like guesswork.

That said, diagramming is a tool, not a religion. On many Logical Reasoning questions, a clear mental grasp of the conclusion and the gap is faster than drawing symbols. Use notation when the logic is genuinely dense, and skip it when it would slow you down. The goal is accuracy at speed, and the right test taker uses the lightest tool that gets the question right.

What else should you know about LSAT logical reasoning?

A few quick answers cover the questions students ask most about LSAT logical reasoning: how many sections there are, whether wrong answers hurt you, how to weaken an argument, and whether logic games still appear. The short answers below distill the strategy above into fast, test day ready guidance.

How do you get better at LSAT logical reasoning?

You get better at LSAT logical reasoning by making argument structure automatic, learning every question type, and reviewing relentlessly. Read each stimulus for its conclusion and support, drill one question type at a time until the pattern is obvious, and keep an error log so your weakest types become your practice. Logical Reasoning rewards this steady, structural work far more than cramming.

How many logical reasoning sections are on the LSAT?

According to LSAC, since August 2024 the LSAT has two scored Logical Reasoning sections out of three scored sections, alongside one Reading Comprehension section and one unscored variable section. That means Logical Reasoning appears twice on every scored test, which is why it carries so much weight and deserves the majority of your study time.

Is logical reasoning harder than reading comprehension?

Whether LSAT logical reasoning is harder than reading comprehension depends on you, but logical reasoning matters more because it is two of three scored sections. Many students find logical reasoning more learnable, since its question types and flaws repeat predictably, while reading comprehension depends heavily on stamina and speed. Invest first in logical reasoning, then shore up reading comprehension.

How long is each logical reasoning section?

According to LSAC, each section of the current LSAT, including both Logical Reasoning sections, runs 35 minutes. With two scored Logical Reasoning sections plus one Reading Comprehension section and one unscored variable section, you face four 35 minute sections on test day. Practicing full timed 35 minute sections is the only way to build reliable Logical Reasoning pacing.

How many logical reasoning questions are on the LSAT?

LSAC does not publish an exact count of scored questions, so no official per section number exists for Logical Reasoning. A common non-official estimate puts the whole test around 75 to 76 scored questions across the three scored sections, which would place roughly a couple dozen in each Logical Reasoning section. Treat that as a planning estimate, not an official figure.

Do wrong answers hurt your LSAT score?

No. According to LSAC, your raw score is the number of questions you answer correctly, with no penalty for wrong answers, before it is converted to the 120 to 180 scale. Because guessing costs nothing, you should never leave a Logical Reasoning question blank. Fill in every answer, even the ones you flag and run out of time on.

Should you diagram every logical reasoning question?

No, you should not diagram every logical reasoning question. Reserve formal notation, arrows, and contrapositives for questions where conditional logic is dense, such as sufficient assumption, principle, and some inference questions. For most strengthen, weaken, and flaw questions, a clear grasp of the conclusion and the gap is faster than symbols. Diagram only when it saves time.

What is the difference between a necessary and a sufficient assumption?

On the LSAT, a necessary assumption is something the argument must rely on, so if it were false the argument would fall apart. A sufficient assumption is strong enough to guarantee the conclusion on its own. Necessary assumption questions reward the negation test, while sufficient assumption questions reward building the missing link that forces the conclusion.

How do you weaken an argument on the LSAT?

To weaken an LSAT logical reasoning argument, attack the gap between the premises and the conclusion rather than the premises themselves. Find the author's assumption and show it may not hold, or introduce a plausible alternative explanation. The strongest weaken answers make the conclusion less likely without needing to disprove it outright. Extreme or off topic choices are traps.

What is a blind review, and does it work for logical reasoning?

A blind review is redoing questions you were unsure of without a timer before you check answers, so you separate careless errors from real gaps. It works well for logical reasoning because it forces you to commit to reasoning, then compare it against the credited answer. The gap between your timed and untimed scores shows exactly what pacing is costing you.

Can you improve your LSAT logical reasoning in a month?

You can improve LSAT logical reasoning in a month if you study deliberately, though large jumps usually take longer. In a focused month, you can learn the question types, fix recurring flaws, and sharpen timing through daily timed sets and blind review. Because Logical Reasoning is two of three scored sections, even steady gains there move your overall score meaningfully.

Are logic games still on the LSAT?

No. According to LSAC, Analytical Reasoning, the section commonly called Logic Games, was removed starting with the August 2024 LSAT, and the last test to include Logic Games was June 2024. The current LSAT has two scored Logical Reasoning sections and one Reading Comprehension section, so your preparation should ignore Logic Games entirely.

Sources

LSAC, LSAT scoring: https://www.lsac.org/lsat/lsat-scoring

LSAC, what to expect starting with the August 2024 LSAT: https://www.lsac.org/blog/what-to-expect-starting-with-august-2024-lsat

Ready to see exactly where your Logical Reasoning stands and what it projects to on the 120 to 180 scale? Take a free diagnostic LSAT and get your projected score.