How to Improve Your LSAT Score: A Realistic Guide to +10 Points

Ten points is a lot. On a scale whose midpoint sits around 154, moving from a 155 to a 165 takes you from the middle of the pack to roughly the top fifteen percent, and it changes which schools are realistic and what they offer you. It is a common gain and an achievable one. It is not a quick one, and anyone promising it in three weeks is selling something.

What follows is the loop that actually produces it, and the specific places where most study plans go wrong.

Why most study plans stall

The standard approach is volume: take a practice test, review the answers you missed, take another practice test. Repeat for three months. Scores usually rise for a few weeks, then flatten, and the flattening is the confusing part: the work feels the same but the number stops moving.

There are two reasons for that, and they're worth separating.

The first is that a practice test is a measurement, not a workout. It tells you where you are. It doesn't teach you anything you didn't already know, and taking one costs two and a half hours you could have spent on the thing the last one told you to fix.

The second is that a score is far too coarse to act on. "I got a 158" gives you one number describing seventy-odd questions that tested the nineteen Logical Reasoning question types and eleven Reading Comprehension question types we track. Two people with a 158 can need completely opposite study plans. Without knowing which questions you're missing and why, you'll practice what's comfortable, which is usually what you're already good at.

The fix is a loop with four parts, run continuously rather than in phases.

1. Diagnose

Start with a baseline, but treat the score as the least interesting output. What you want is the breakdown: your accuracy on Necessary Assumption versus Flaw, on Parallel Reasoning versus Main Point, on structural reading versus perspective tracking.

You need enough attempts per type for the numbers to mean something. A single missed Sufficient Assumption question tells you nothing, ten attempts at 40 percent tells you plenty. This is the main argument for drilling by type early on rather than doing full sections: full sections give you two or three questions of each type and no usable signal.

Sort what you find into three buckets:

• Types you're solid on (85 percent and up). Maintain, don't grind.
• Types you're inconsistent on (roughly 50 to 80 percent). This is where the ten points live. You understand these; something specific is going wrong.
• Types you don't have (below 50 percent). These need instruction, not repetition. Drilling a type you fundamentally don't understand just rehearses the misunderstanding.

2. Drill

Drill the middle bucket, hard, and drill it untimed to start.

Untimed practice is the part people skip and the part that matters most. The purpose isn't to be comfortable; it's to give yourself enough room to execute the full method, analyze, predict, eliminate, on every single question, until it stops requiring effort. Practicing under time before the method is automatic just rehearses the panicked version of what you already do.

Two things make a drill work:

Difficulty at the edge of your ability. Questions you get right easily teach you nothing; questions far beyond you teach you nothing either, because you can't reconstruct the path to the answer even after seeing it. You want questions you can get right with real effort, and get wrong when you're sloppy.

Mixed types, eventually. Blocks of one type are right at the beginning, when you're learning what a type asks. But blocked practice quietly does half the work for you, since you know it's a Weaken question before you read the stem. Once a type is landing consistently, mix it back in, because recognizing the task is itself a skill the test is measuring.

3. Review

This is the highest-leverage hour in LSAT prep, and it's the one most people do in five minutes.

Reading the explanation for a question you missed and thinking "ah, I see" is not review. It produces a feeling of understanding without changing anything, because the explanation makes sense in hindsight to almost everyone. The question is why it didn't make sense in the moment.

Review before you check

The most useful review habit costs nothing. After you finish a set but before you look at the answers, go back to every question you weren't confident about and work it again with unlimited time. Write down your answer the second time.

Now check both. The gap between your under-time score and your untimed score is the single most diagnostic number in your prep:

• Big gap — you know how to do these questions and you're not getting the chance to. That's a timing and process problem, and the fix is fluency, not more content.
• Small gap — more time doesn't help, so the method is what's failing. Slow down and fix the reasoning; speeding up would only produce wrong answers faster.

Most people assume they have the first problem. Many have the second.

Categorize every miss by cause

For each question you got wrong, name the failure rather than the topic:

• Misidentified the conclusion.
• Found the conclusion but not the gap.
• Made a good prediction, then talked myself out of it.
• Never made a prediction and shopped the choices.
• Misread the stem and answered a different question.
• Knew it, rushed, picked carelessly.

After twenty or thirty questions, a pattern emerges, and that pattern is worth more than your practice test score. "I talk myself out of correct predictions on Strengthen questions" is a fixable, specific problem. "I need to get better at Logical Reasoning" is not.

Review the questions you got right but were unsure about, too. A lucky guess is a question you'll miss next time, and it looks identical to mastery in your accuracy stats.

4. Time

Add the clock only once the method is automatic. When you do, work on it directly rather than hoping it resolves itself.

The most valuable timed skill is the skip decision. Every question is worth the same point, so a question that's eating three minutes is actively costing you the two easier questions sitting at the end of the section. Set a personal ceiling. Many people use ninety seconds. When you hit it without a clear direction, mark it, guess, and move. Come back if there's time.

This is a genuinely hard habit, because abandoning a question you've invested two minutes in feels like waste. It isn't; the two minutes are already gone either way. The alternative is spending a third minute to protect them.

Practice tests belong here too. Roughly one a week is plenty. They're for calibration: stamina, pacing, whether the drilling is transferring. Each one should generate a week of targeted work, not just a number in a spreadsheet.

What a plateau actually is

Nearly everyone hits one. It usually looks like this: rapid early gains as you learn the question types, then a wall somewhere in the high 150s or low 160s that lasts for weeks.

A plateau is almost always the point where the easy points have been collected and what remains are the questions that require the method you haven't fully built. The early gains came from learning what the test asks. The next ones come from executing consistently under pressure, which is slower and much less satisfying, because progress shows up as fewer careless errors rather than as new knowledge.

The wrong response is to change materials or start a new course. The right response is usually to go back to untimed work on the specific types that are still inconsistent, and to review harder, because a plateau frequently means the same three failure modes recurring, uncorrected, for a month.

A realistic timeline

Most people who gain ten points do it over two to four months at ten to fifteen hours a week. That's a rough shape, not a promise: starting point matters, the gain from 145 to 155 is not the same work as 165 to 175, and nobody can guarantee a score.

A sane distribution of that time:

• Weeks 1–2: learn the question types and the method. Untimed, low volume, high attention.
• Weeks 3–8: drill weak types untimed, review hard, take a practice test every week or two to check transfer.
• Weeks 9–12: add timing pressure, mixed sets, full sections, and work the skip decision deliberately.
• Final two weeks: full practice tests under real conditions, light review, no new material. Taper rather than cram.

And give Logical Reasoning the majority of your hours. It's two of the three scored sections, it improves faster than Reading Comprehension, and the argument skills carry over into the reading passages anyway.

What to do this week

1. Take one timed section and record every answer, including how confident you were.
2. Before checking, redo everything you were unsure about, untimed. Compare the two scores.
3. Categorize every miss by cause, not by topic.
4. Pick the two question types with the worst accuracy and drill only those, untimed, for a week.
5. Do not take another full practice test until you've done that.

That's the whole loop, and it's more effective than another full-length test telling you a number you already know.