.png)
How to improve LSAT logical reasoning when you have plateaued: stop reviewing answers and start reviewing reasoning. Write three lines on every miss before you read the explanation, covering what you thought the argument did, why your choice pulled you, and the word that would have saved you.
Almost always because you are reviewing answers rather than reasoning. The distinction sounds pedantic and it is the entire difference between a plateau and a climb.
Reviewing answers means reading the explanation, understanding why C was right, feeling the click of comprehension, and moving on. It produces a strong sense of learning and almost no transfer, because the thing you learned was a fact about one question.
Reviewing reasoning means reconstructing the decision you made: what you thought the argument was doing, which choice you nearly picked, and what specific word would have told you otherwise. That produces a rule you can carry to the next question.
A wrong-answer journal, three lines per question, written before you look at any explanation. The before matters more than the format.
LineWhat you writeWhy it works1. What I thought the argument was doingYour reading of the structure at the timeSurfaces whether the error was in comprehension or in evaluation2. Why my choice was attractiveThe specific feature that pulled youWrong answers repeat their tactics; naming one immunizes you against the family3. The word that should have told meOne phrase from the stimulus or the choiceThis is the transferable line, and the one people skip
Because once you have read the explanation you can no longer reconstruct what you thought. The corrected version overwrites the original, and you will sincerely believe you nearly had it.
This is a documented failure of self-assessment rather than a personal weakness. In Roediger and Karpicke (2006), Psychological Science (https://colinallen.dnsalias.org/Readings/2006_Roediger_Karpicke_PsychSci.pdf), students who studied material four times predicted the highest retention, at 4.8 out of 7, and performed worst a week later, at 40 percent against 61 percent for students who had been tested. Confidence and learning came apart completely.
So write first. Three lines take ninety seconds and they are the only record of what you actually did.
Diagnose before you prescribe. Most people are strong in one of these and weak in exactly one other, and treating the wrong one is why months disappear.
Failure modeWhat it looks like in your dataWhat fixes itMisreading the conclusionErrors cluster in main conclusion, method and flaw questionsRebuild every argument as premise-therefore-conclusion before evaluatingPrediction skippingYou are consistently down to two choices and pick the wrong oneForce a written prediction before reading choices, for two weeksStandard driftErrors cluster in necessary assumption and must be trueLearn the negation test and the certainty standard as mechanical checksTiming collapseAccuracy is fine in the first 15 questions and falls off a cliffThis is a pacing problem, not a reasoning problem, and drilling harder questions makes it worse
Blocked drilling teaches the pattern and hides whether you can identify it, because you knew the type before you read the stimulus. On the real test, identifying the type is half the skill.
The measured cost of blocking is large. Interleaved practice produced 61 percent correct against 38 percent blocked at a 33-day delay, an effect of d = 0.83 across 787 students and 54 classrooms, per Rohrer, Dedrick and Hartwig (2019), Journal of Educational Psychology (https://gwern.net/doc/psychology/spaced-repetition/2019-rohrer.pdf).
Blocked practice also feels better, which is why people stay in it. Kornell and Bjork (2008), Psychological Science (https://gwern.net/doc/psychology/spaced-repetition/2008-kornell.pdf) found 78 percent of students judged massing as good as or better than spacing while 78 percent performed better spaced.
Stop taking timed sections as your main activity. Practice tests alone produced g = .19 at p = .233, not statistically significant, against g = .40 for workbook-style structured practice, per Hao and colleagues (2025), Review of Educational Research (https://journals.sagepub.com/doi/10.3102/00346543251360775).
Stop adding hours. Neither contact hours at p = .489 nor program duration at p = .314 predicted outcomes in that same Hao analysis. If more hours were the answer you would already have the score.
And stop reviewing on the same day. The testing effect rises to g = 0.69 at a delay of a day or more against g = 0.41 under a day, per Rowland (2014), Psychological Bulletin (https://pubmed.ncbi.nlm.nih.gov/25150680/). Same-day review is the most common habit and among the least productive.
Every LSAC figure on this page was taken from LSAC's own pages and last checked on July 26, 2026. Prices, dates and policies move without notice, so the source link sits next to each number rather than in a bibliography at the bottom.
Where LSAC has not published something, this page says so in those words rather than filling the hole with a plausible number. That is the difference between a gap and a guess, and only one of them is safe to plan around.
Usually because you are reviewing answers rather than reasoning. Reading why C was right produces comprehension without transfer; reconstructing the decision you made produces a rule you can reuse.
Three lines per wrong question, written before you read any explanation: what you thought the argument was doing, why your choice attracted you, and the exact word that should have told you otherwise.
Fewer than you think, reviewed harder. Volume shows no dose response in the research, while structured review of the reasoning is where the measured effect sits.
Only until you can identify the type from the stem. After that, mixed sets, because interleaved practice beat blocked practice 61 percent to 38 percent at a 33-day delay, per Rohrer and colleagues (2019).
At least a day later. The testing effect is g = 0.69 at a delay of a day or more against g = 0.41 under a day, per Rowland (2014), so same-day review gives up most of the benefit.
Written by Ali, Georgetown Law, founder of Lovare Institut.
August 5, 2026
August 5, 2026