It's Sunday night. You just finished PrepTest 78, the third one this month. Same score, within a point of the last two. You spend an hour reading the explanations for the questions you missed, and every one of them makes sense. You go to bed convinced tomorrow's session will be different. It won't. This is the plateau, and it doesn't break because you take more tests or read more explanations. It breaks when you stop treating the LSAT as one big skill and start attacking it as a stack of narrow, learnable ones, the way the people who actually score 170+ do. This page is that plan.
The single reason most students plateau is that they treat the LSAT as one big skill instead of a stack of narrow ones. The test has roughly fifteen distinct question patterns across Logical Reasoning and Reading Comprehension. You are not equally bad at all of them. You have three or four that eat your score. The whole game is finding those and killing them.
Start with a diagnostic, not a full PrepTest, which will just leave you demoralized. A short, targeted diagnostic tells you which question types you actually miss, and why. Our free 4-question diagnostic takes five minutes and returns a written plan built around your specific misses.
The LSAT you're studying for is two scored Logical Reasoning sections, one scored Reading Comprehension section, and one unscored experimental section that looks like one of the other two. Each section is 35 minutes. Scoring runs 120, 180.
Logic Games were removed from the test in August 2024. If any prep resource you're using still teaches diagramming for grouping games and sequencing games, throw it out. That content is a tax on your study time and will not appear on your test.
For the full format breakdown, read What is the LSAT?
Every effective LSAT study plan moves through the same four phases in the same order. Skipping ahead is the second most common reason students plateau, after skipping diagnosis.
Learn the reasoning vocabulary before you drill anything under time. That means conditional logic (sufficient vs. necessary, contrapositives), argument anatomy (premise, conclusion, assumption), and the recurring flaw patterns. Everything else on the test is a variation on these.
Deliverable: you can read any LR stimulus, in your own words, and say "the conclusion is X, the support is Y, and the gap is Z", without looking at the answers.
Now attack the LR question types. Do not mix them yet. Spend a few days per type, read the guide, do 20, 40 problems untimed, review every miss (and every guess that landed right), and write down the pattern you keep falling for. Move to the next type only when your accuracy on the current one is above 85% untimed.
Work in this order:
RC rewards the same reasoning as LR, applied to longer text. Two habits separate 170s from everyone else: reading for structure, not content (what is the author doing in each paragraph), and pre-phrasing an answer before looking at the choices. Drill one passage a day, blind-reviewed, before you touch timed RC sections.
Only now do you sit for full, timed PrepTests. The rule is one-to-one: for every hour of timed practice, spend at least an hour reviewing. Blind review means before you check the answer key, you go back to every question you weren't 100% sure about and re-do it untimed until you can defend your answer. The gap between your timed score and your blind-review score tells you exactly what to work on next.
Reading the explanation is not review. Real review looks like this:
Score improvement is not mysterious. It comes from two things: closing the accuracy gap on the two or three question types that cost you the most points, and converting the "sure" questions into faster time so you can spend real time on the hard ones. Every minute you save on the first 15 LR questions is a minute you get back for the last 10, where the points hide.
If your score has been stuck for three weeks or more, stop taking PrepTests. Go back to Phase 2 on the type you miss most, and don't resume timed sections until that type is above 85% untimed.
Most students need three to six months at 10, 20 focused hours a week. Raw hours matter less than what you do with them: untimed diagnosis first, drill your weakest question type second, timed sections only after accuracy is in place.
Yes. Many 170+ scorers self-study. What they don't skip is a diagnostic and a written plan. Grinding practice tests without knowing which question types cost you points is the most common way to plateau.
Logical Reasoning first, it's roughly two-thirds of your scored questions and the reasoning transfers to Reading Comprehension. RC after. Writing gets its own short block near test day. Logic Games were removed from the test in August 2024; skip that content entirely.
Quality over quantity. A serious plan is 8, 15 fully timed, fully reviewed PrepTests. Every miss gets a blind-review pass, a category tag, and a drill queue. Ten reviewed tests beats twenty rushed ones every time.
If you've plateaued for three-plus weeks, if you're on a deadline, or if you can't diagnose why you miss what you miss, yes. Otherwise, a diagnostic-first self-study plan works for most people. See our comparison of self-study vs tutoring.
1-month, 3-month, and 6-month plans, honest about tradeoffs.
What separates 165 from 172, and how to close the last five points.
When to keep grinding solo and when a tutor pays for itself.
Four questions. Five minutes. No email wall. A written plan built around your actual misses.
Diagnose my weak spots →Last updated July 10, 2026 · by the theLSATguru team