LSAT Flashcards: What Belongs on a Card

LSAT flashcards work for exactly three things — flaw types, conditional logic translations, and question stems — and they fail at everything else. Those three categories share one property: each is a small, fixed, recallable unit with a single correct answer. Everything else people try to put on an LSAT card, especially whole arguments and reading passages, is a skill rather than a fact, and skills are built by practicing them, not by flipping cards.

That distinction is the entire article. Below is what goes on the card, what the card should look like, what to leave off, and how to arrange the deck so it supports actual practice rather than replacing it.

The rule: card the vocabulary, practice the skill

A flashcard is a retrieval tool for discrete units of knowledge. The LSAT is largely a skills test with a small vocabulary underneath it. Card the vocabulary; drill the skill on real questions.

Think of it the way a musician does. Scales and chord shapes get memorized cold, because they are fixed and finite. Playing a piece is a different activity that uses them. Nobody memorizes a performance one measure at a time on index cards.

The LSAT's "scales" are small and surprisingly learnable:

That is a few hundred cards at most, and it is genuinely worth memorizing to the point of automaticity, because recognizing a pattern in two seconds instead of twenty is a real advantage on a timed test.

What it is not is a substitute for working through real questions under timed conditions. Anyone who tells you a deck alone will prepare you is selling something.

Confirm the current format with LSAC before you build

The test's structure has changed more than once, so check the current format directly with LSAC — the organization that administers the LSAT — rather than trusting a prep article's description of it.

This matters for deck planning. If you build a deck around a question type or a section that is no longer part of the current test, you have spent weeks memorizing something you will never be asked. Nothing in this article makes any claim about section timings, how many questions appear, how the test is scored, or which sections are currently included, because those are exactly the details that shift and that you should verify at the source.

The reasoning concepts below are stable in a way that format is not. A circular argument has been a circular argument for two thousand years. Conditional logic does not get revised between administrations. Those are safe to memorize deeply.

Carding logical reasoning flaw types

A flaw card should have the flaw's name on the front and, on the back, a one-sentence definition plus an eight-word abstract example — nothing longer.

The eight-word example is the part people skip, and it is the part that makes the card work. A definition alone stays abstract; a tiny concrete instance gives your memory something to grab.

Here is the shape:

Front: Necessary/sufficient confusion Back: Treats a required condition as though it guarantees the outcome. "You need water to live, so water keeps you alive."

Compare that to the card most people write, which puts a full 60-word argument on the front and a paragraph of explanation on the back. That card cannot be graded honestly, takes forty seconds per review, and never becomes automatic.

Flaw nameOne-line definition
Correlation to causationTwo things occur together, so one is claimed to cause the other
Necessary/sufficient confusionTreats a requirement as a guarantee, or the reverse
Unrepresentative sampleGeneralizes from a group that doesn't reflect the whole
Ad hominemAttacks the arguer instead of the argument
Circular reasoningAssumes the conclusion inside the premises
EquivocationA key word shifts meaning between premise and conclusion
False dichotomyPresents two options as though no others exist
Part-to-wholeAssumes what's true of a part is true of the whole
Whole-to-partAssumes what's true of the whole is true of each part
Percentage versus numberConfuses a rate with an absolute quantity
Appeal to authorityCites an authority outside their actual expertise
Appeal to ignoranceTreats absence of evidence as evidence of absence
Straw manRefutes a weaker version of the opposing claim
Ignoring an alternative causeOverlooks another explanation for the same evidence
Self-contradictionThe argument's own claims cannot all be true
Survivorship problemLooks only at cases that made it through a filter
Conflating opinion with factTreats what people believe as what is true
Temporal fallacyAssumes a past trend must continue unchanged

Run this deck in both directions once you know it. Front to back is "define this flaw." Back to front — reading the definition and naming the flaw — is closer to what the test actually asks you to do, and it is noticeably harder.

Carding conditional logic translations

Conditional logic is the highest-return deck you can build, because the translations are mechanical, finite, and constantly reused.

There are three card types here and they do different jobs.

Indicator word cards

Front: the word. Back: whether it introduces the sufficient condition or the necessary one, plus a three-word example.

IndicatorIntroducesExample
IfSufficientIf A, then B
When / wheneverSufficientWhenever A, B
All / every / anySufficientAll A are B
In order toNecessaryNeed B to A
RequiresNecessaryA requires B
OnlyNecessaryOnly B are A
Only ifNecessaryA only if B
Must / essentialNecessaryA must have B
Unless / withoutNegate the sufficientA unless B → not B, then A
No / noneNegate the necessaryNo A are B → A, then not B

The two rows at the bottom are the ones that cost people points, and they deserve extra cards. Unless is the classic trap: "I won't go unless you come" translates to "if you don't come, I won't go." Card that specific sentence, not just the rule.

Transformation cards

Front: a statement. Back: the one valid transformation and the two invalid ones, named.

Front: A → B. What follows? Back: Contrapositive: not B → not A. Invalid: B → A (converse). Invalid: not A → not B (inverse).

Two more transformation cards earn their keep:

Front: A → B and C Back: Contrapositive: not B or not C → not A. (The "and" flips to "or.")

Front: A or B → C Back: Contrapositive: not C → not A and not B. (The "or" flips to "and.")

That and/or flip is the single most commonly botched piece of formal logic on the test. It deserves three or four cards on its own.

Quantifier cards

Front: two statements. Back: what validly follows, or "nothing."

Front: All A are B. All B are C. Back: All A are C.

Front: Most A are B. Most A are C. Back: At least some B are C. (Both majorities must overlap.)

Front: Some A are B. All B are C. Back: Some A are C.

Front: Most A are B. Most B are C. Back: Nothing follows.

That last card — a valid-looking chain that produces nothing — is worth more than the three above it combined, because it is the mistake the test is built to catch.

Carding question stems

Front: the stem, paraphrased. Back: what the correct answer must do, in one line. Not strategy, not a checklist. One line.

Stem familyWhat the right answer must do
Most weakens the argumentAttack the link between premises and conclusion, not the premises
Most strengthens the argumentSupport that link, or close a gap in it
An assumption required by the argumentState something the argument cannot survive without
An assumption sufficient to justifySupply enough to make the conclusion follow completely
Must be true / properly inferredFollow from the stated information with no additions
Most strongly supportedBe the best supported claim, though not airtight
Principle that justifiesGive a general rule the specific case fits under
Parallel reasoningMatch the logical structure, not the subject matter
Reasoning is flawed becauseName the specific error, not merely disagree
Method of reasoningDescribe what the argument does, not whether it's right
Point at issueIdentify what two speakers actually disagree about
Role of the statementSay what job that sentence does in the argument
Resolve the discrepancyExplain how both facts can be true at once

The distinction between the two assumption rows and the two inference rows is where most of the value sits. Those four pairs are close enough to blur under time pressure, and cards are unusually good at keeping them separate.

What does not belong on LSAT flashcards

Anything you cannot answer in under ten seconds with a single correct response is not a flashcard. That rules out most of what people try.

Whole logical reasoning stimuli. A full argument with five answer choices is a practice question, not a card. Cards should hold the concepts; questions should hold the practice. Putting a stimulus on a card produces the worst of both — you memorize that particular answer and learn nothing transferable.

Reading comprehension passages. These are entirely a skill. Nothing about a passage generalizes to the next passage. There is no version of this that works.

Long analytical setups. If your prep involves diagrammed game-style setups, the diagrams themselves are practice artifacts. The rules for building them — how a particular constraint gets notated — can be carded. The completed board cannot.

Multi-part strategy checklists. "Five steps for attacking a weaken question" is a study note, not a card. It has no single correct answer, so it cannot be graded, and an ungradable card is a card you will lie to yourself about for weeks.

Vocabulary from passages. Tempting, and almost never worth it. The test does not reward knowing obscure words; it rewards tracking structure. Time spent on a vocabulary deck is time not spent on conditional logic.

If a card feels too big, split it. The general principle — one card, one fact, one unambiguous answer — is covered in more depth in How to Memorize Flashcards So They Actually Stick, along with what to do about a card you keep failing.

A workable deck plan

Build three small decks rather than one large one, and keep each under a hundred cards.

DeckRoughlyBuild it when
Conditional logic40–60 cardsFirst. It's the foundation and it's fastest to learn
Flaw types30–40 cardsSecond, once you've seen enough real flaws to recognize them
Question stems25–35 cardsThird, as a maintenance deck you review briefly and often

A deck this size can be built in one sitting from a spreadsheet. The free flashcard maker at flashcardmakeronline.com takes a CSV import, so a two-column list of indicators and translations becomes a working deck in a couple of minutes, and the shareable link is handy if you're prepping with a study partner and want the same deck on both sides.

Two habits make the difference between a deck that helps and one that just sits there:

Add cards from your own mistakes. After a practice set, look at what you got wrong. If the error was conceptual — you misread an "only if," you missed a percentage-versus-number flaw — write the card immediately, in your own words. Cards born from your own errors are the highest-value cards in any deck.

Review before practice, not instead of it. Five minutes of card review as a warmup, then the real work. Flipping cards is comfortable and practice questions are not, which is exactly why the deck can quietly become an avoidance mechanism.

If plain review gets stale, the retrieval formats in Games Using Flashcards: 12 That Build Real Recall transfer surprisingly well to a study group — the three-pile sort and Odd One Out in particular work well on flaw types. And for a look at the opposite end of the flashcard design problem, where the requirement is exact wording rather than abstract patterns, Bible Flashcards for Verses, Books and People covers how word-for-word decks are built.

Frequently asked questions

Are LSAT flashcards worth making at all, or should I just do practice questions? Both, in that order of time. The deck is small — maybe 150 cards total — and it buys you speed on the things that should be automatic. Practice questions build the skill the test actually measures. A reasonable split is a short daily card warmup and the large majority of your time on timed practice.

Should I buy a pre-made deck or make my own? Make your own, at least for the flaw types and anything drawn from your errors. Writing the definition in your own words is itself a learning step, and a purchased deck usually contains a hundred cards you don't need alongside twenty you do. A pre-made conditional logic deck is the most defensible thing to buy, since those translations are standardized.

How do I card a flaw I keep missing on practice sets? Write it as a pair. One card with the flaw's name on the front and your own definition on the back; a second with a compressed version of the argument that fooled you on the front, and the flaw's name on the back. The second card is the one that fixes the problem, because it rehearses recognition rather than definition.

How many cards should I review a day? Enough to keep the deck warm, not enough to displace practice. For a 150-card deck, fifteen to twenty-five minutes of review a day is generous once the deck is built, and it drops considerably after a few weeks as the easy cards space out.

Can I use flashcards for reading comprehension? Not usefully for passage content. There is one narrow exception worth trying: cards for the structural moves passages make — "author introduces a theory, then complicates it," "two scholars disagree about causes" — with the question of what typically gets asked about each. That is a pattern deck rather than a content deck, and it stays small.

Where do I confirm what's actually on the current test? Directly with LSAC, which administers the exam. Format details have changed over time, and any secondhand description — including a prep book printed last year — may be out of date. Check the source before you plan a deck around a question type.