The rule: pick the answer that assumes the least
When you are down to two LSAT answer choices and both feel defensible, choose the one that requires you to assume the least. The correct answer almost never depends on an extra unstated leap; the trap answer usually does.
This is the single most useful tiebreaker in Logical Reasoning. It works because the LSAT is a test of supported reasoning. The right answer is the one the stimulus actually gives you grounds for — not the one that is merely possible if you grant it some help.
So when you are torn, stop asking "which one sounds better?" and start asking "which one am I adding the least to?" Whichever answer survives with the fewest invented assumptions is your pick.
Why being stuck between two is so common
Getting down to two answers is by design. The testmaker writes four wrong answers on purpose, and at least one of them is built to be attractive — it shares vocabulary with the stimulus, matches your gut prediction, or states something true in the real world.
That means the final two are rarely a coin flip. One of them is the credited answer and the other is an engineered near-miss. Treating the choice as 50/50 luck is exactly the mistake the question is testing for.
Your job at this stage is not to find the answer you like. It is to find the specific reason one of the two has to be wrong.
Eliminate with a logical reason, not a feeling
Eliminate an answer only when you can name what is wrong with it. "I don't like how it's worded" or "it feels off" is not a reason — it is a feeling, and feelings are what trap answers are engineered to produce.
Complicated, clunky wording is often deliberate. A correct answer can be phrased in an ugly, abstract way precisely to make you flinch and grab the smoother trap instead. The wording being uncomfortable is not evidence that the answer is wrong.
Before you cross anything out, finish this sentence: "This answer is wrong because ___." If you can't, you don't have grounds to eliminate it yet. Keep it and look harder at the other one.
The 4-wrong-answers method
Strong scorers don't hunt for the right answer; they disqualify the four wrong ones. Reframe every Logical Reasoning question as: find a concrete reason each of the four wrong answers fails, and a concrete reason the remaining one holds.
This flips the question from a subjective "which is best?" into an objective "which four can I rule out?" — and ruling out is far more reliable than ranking. It also makes your final two faster, because you arrive there having already attached a reason to each elimination.
Concretely, the common disqualifiers are: the answer is out of scope, it's too strong for what the evidence supports, it reverses a relationship, it's only half-right, or it states the opposite of what you need. Name which one applies.
A worked example
Stimulus: "The city's new bike lanes opened in March. Since then, reported car-bike collisions downtown have fallen by a third. The bike lanes have therefore made downtown cycling safer."
Suppose you're down to two answers on a weaken question. Answer A: "Many cyclists now avoid downtown entirely because they find the new lane layout confusing." Answer B: "Bike lanes in other cities have sometimes increased collisions at intersections."
Test each by what it forces you to assume. Answer B requires you to assume this city behaves like those other cities — an extra leap the stimulus never supports. Answer A needs no such leap: if fewer cyclists are downtown, fewer collisions could simply reflect fewer riders, not greater safety. A assumes less and attacks the actual argument, so A is the stronger weakener.
Notice you didn't pick A because it "sounded right." You picked it because B smuggled in an assumption and A didn't.
The common mistake: grabbing the answer that matches your prediction
The most frequent error in the final two is choosing the answer closest to the prediction you made before reading the choices. Predictions are useful, but the LSAT does not owe you your prediction — and testmakers can use a predictable prephrase to build a trap that's worded to feel like your guess.
If one of your two answers matches your prediction and the other doesn't, that is not a reason to pick it. Re-evaluate both on the same standard: which assumes less, and which one can you give a logical reason to eliminate?
Hold your prediction loosely. It is a tool for orientation, not a verdict. The text in front of you outranks the answer you hoped to see.
Frequently asked questions
How do I choose between two LSAT answers that both seem right?
Pick the one that requires the fewest unstated assumptions. The correct answer is supported by the stimulus as written; the trap usually needs an extra leap. If you can't decide, try to state a concrete reason one of them must be wrong — out of scope, too strong, reversed, half-right, or opposite.
Is it bad that I keep getting down to two answers?
No — that's normal and by design. The testmaker writes attractive wrong answers on purpose. Getting to two means your reading is mostly working; the skill to build now is the final cut, which comes from eliminating with reasons rather than feelings.
Should I just go with my gut when I run out of time?
If you must guess, the least-assumption rule still helps: favor the more modest, better-supported answer over the bolder one. But when you have time, replace gut with a reason — name what's wrong with the answer you reject before you cross it out.
Does the least-assumption rule work on every question type?
It's most powerful on weaken, strengthen, assumption, and inference questions, where 'support' is the whole game. On sufficient assumption questions the logic flips — there you may want the stronger answer that fully closes the gap — so always read the stem first.
Related Verbloom guides
Want LSAT logic to feel visual?
Verbloom turns argument structure into short visual lessons, drills, and explanations built for actual score movement.