Flaw Questions: How to Name the Error in Under 60 Seconds

Flaw questions are the most common type on the Logical Reasoning section, and they are also the most mechanically learnable. Every flaw question has the same three parts: a conclusion, evidence, and a gap between them. Once you can name the gap in a phrase before reading the choices, the answers stop being a search and become a confirmation.

The reason students lose points here is procedural. They read the stimulus, then start reading five answer choices trying to spot which one sounds like a criticism. That inverts the order. The error must be named first, from the stimulus alone, without reference to the choices.

The 60-second procedure

1. Second 1–10: locate the conclusion

   Look for the sentence the argument is trying to establish. Signals: "therefore," "thus," "this shows," "the key is," or a claim the rest of the passage supports. If the stimulus has no explicit transition word, the conclusion is usually the claim that other sentences are offered in support of.

2. Second 10–20: state the evidence in one sentence

   If you cannot summarize the premises in a single sentence, you do not yet understand the argument, and reading the choices will not help. This is the step students skip.

3. Second 20–40: find the gap

   Ask: what would have to be true for the evidence to establish the conclusion, and is that stated or unstated? The gap is almost always an unstated assumption, an overreach, or a reversal.

4. Second 40–60: name it in a phrase

   "Treats correlation as causation," "reverses the conditional," "assumes the absence of contrary evidence means absence of the thing," "compares groups that differ on a relevant dimension." That phrase is what you match against the choices.

The error families

Flaw answers cluster into a small number of families. Recognizing the family from your one-sentence description narrows five choices to one or two.

Assumption and overreach account for the largest share; the rest appear less often but are each worth recognizing instantly.

The two distinctions that cost the most points

Two confusions account for most of the lost Flaw questions. Both are about a choice that sounds right but describes a different problem than the one the stimulus has.

1. An alternative explanation versus a flaw

   "The correlation might be coincidental" is a valid weakener. But if the argument already acknowledges coincidence and rules it out, repeating it is not the flaw. Always check whether the stimulus has already addressed the problem you are about to name — many questions include the obvious defense specifically to rule out the obvious answer.

2. Too strong versus logically invalid

   If the conclusion goes further than the evidence strictly supports, the error is overreach, and the correct answer describes the missing limit. If the conclusion does not follow at all, the error is invalid reasoning, and the correct answer describes the invalid step. These require different answers and the distinction is visible in how the choice is phrased.

A related trap: the stimulus may contain a flaw that is not the one the question asks about. If a passage has both an assumption gap and a causal error, and the question asks "most vulnerable to criticism," both are candidates — the correct one is the error that, if repaired, would most change the argument's strength. Read the question stem for whether it asks for the weakest link or merely some weakness.

Wrong answers that look right

Certain choice types appear on almost every Flaw question and are almost never correct. Recognizing them speeds up elimination considerably.

• True but irrelevant. Factually accurate about the world, but describes something other than the argument's gap. The most common trap.
• Repairs a problem the argument does not have. Strengthening a premise that was never weak.
• Objects to tone. "The argument is oversimplified" or "does not consider every case" — vagueness is not a logical error.
• Contradicts the stimulus. If a choice says the argument's evidence is false when the stimulus stated it as fact, it is wrong.
• Repeats a hedge already present. Stimuli often include qualifiers such as "may" or "in some cases." Claiming the argument ignored a caveat it stated is not an error.

How to drill this efficiently

1. Do Flaw questions untimed first, and write your one-sentence error description before looking at choices. The writing is the point — a silent hypothesis does not get tested.
2. Then compare your description to the correct answer. If yours was vaguer or differently framed, that gap is what to fix.
3. Track which family you miss. A pattern across ten questions tells you whether your problem is locating the conclusion, identifying the gap, or matching vocabulary.
4. Add the clock only once your untimed accuracy is high. Flaw questions should take well under 90 seconds — if yours take longer, the procedure is not yet automatic.

Flaw is worth disproportionate attention because it is both the most frequent type and the most consistent. Our guide to the 14 question types by frequency has the distribution, and the causal family gets its own treatment in six ways LSAT arguments break causally.

Questions we get asked

LSAT® is a registered trademark of the Law School Admission Council, which does not review or endorse this site. This article is educational and is not legal or admissions advice. Format and policy figures reflect LSAC publications available as of October 2026 — confirm against lsat.lsac.org before you register.