The whole method on one page
Here’s the protocol, with the question I ask out loud next to each step. Reviewing alone, run the on-paper column. With a partner or a coach, the out-loud column on the right is what to say.
| Step | Time | On paper | Out loud |
|---|---|---|---|
| 1. The deposition | 2 min | My answer, the correct answer, one honest sentence about what I was thinking when I picked it. Before the key. | “What were you thinking when you picked it?” |
| 2. The rebuild | 3 min | Three lines: evidence, conclusion, and the words from the text that support each choice I kept. In my words, badly. | “What’s the conclusion? What’s the evidence?” Then: “Can you point to the lines supporting that?” |
| 3. Where the point died | 2 min | Circle one: stimulus, stem, choices. | “Stimulus, stem, or choices?” |
| 4. The why-chain | 8 min | Why did that feel safe? Why? Why? Three minimum. Stop at the stop test. | “Does that need to be the case?” Then: “Why did it seem like it did?” |
| 5. The rule and the forecast | 3 min | One sentence, my words, future tense. One line on where this skeleton shows up next. | “What would you do differently on the next one of these?” And: “Where will this show up again?” |
| 6. The rematch | 2 min | A date, one week out. The skeleton logged. | “When do you try it cold?” |
| 0. The Check-In, test day only | 3 sec, every question | Nothing. It runs in your head. | “What did I read? What’s the conclusion? What’s the method of reasoning?” |
The six steps add up to twenty minutes. Each question in the right-hand column is the shortest version, and the chapters give the longer ones. The bottom row is what the six turn into under a clock, three questions you ask yourself before you commit to an answer, and chapter 8 shows where it came from.
What makes it Socratic
The object of the review is you. Every review in chapter 1 ends in a sentence about the question. This one ends in a sentence about the person who missed it, a belief they hold about how arguments work, in their own words. Every exchange in this book lands there.
The order is fixed, and there’s a reason for it. The deposition comes first because the instant you see the key, your memory of your own reasoning changes, and the only honest account you’ll ever get is the one you write before. The rule comes after the why-chain because you can’t write the rule that would have saved the point until you know what cost it. Each step stands on the one before it. When one gets stuck, it’s usually because an earlier one got skipped.
The why-chain ends at a test.
The stop test. You’ve hit root cause when two things are true: the answer is a belief about how arguments work, not a fact about this question, and it would explain other misses too.
You keep asking why until a sentence passes. “I rushed” fails both halves. “I treat what would help an argument as what it requires” passes both.
The questions are the method. If I could keep only one part of the card, I’d keep the right-hand column. Anyone can learn the steps in a minute. What makes them work in a room is asking those questions in that order and explaining nothing until they’ve been answered. Alone, you ask them of yourself, on paper.
A review without a rematch is a theory, and the rematch is the experiment. You wrote a rule. A week later you meet the question cold and find out whether the rule is real. If you miss it again, the review didn’t fail. The root cause just wasn’t root yet, and now you know which review to run again.
Put those five together and you get the sentence the whole method is built on.
Drilling measures you. Review changes you.
Before you start: what has to be true
When the setup is wrong, the method fails in predictable ways, and the student usually decides the method was the problem. Five things have to be true before step 1.
You write before you look. Your answer and your reason go on the page before the key, the explanation, or a friend’s opinion. If you’ve already seen the answer, say so in the deposition, because a reason written with the answer in your pocket is a different reason.
You review troubled questions only. That means misses, guesses that landed, right answers that took too long, and right answers you weren’t sure of. A question you got right quickly, for the right reason, doesn’t need twenty minutes. A guess that landed does, because it’s a miss that hasn’t happened yet. For any right answer you’re unsure about, run the confidence audit. Write one sentence on why it wins and one on why the trap is wrong. If the second sentence is fuzzy, the question gets the full twenty minutes.
It goes somewhere you’ll see it again. A notebook, a spreadsheet, a voice memo. In my experience the students who try to do this in their heads get much less out of it than the ones who write it down.
There’s a date. Every review ends with a day you’ll try the question cold, written next to it. Alone, that’s a week out.
The rule is specific to you. “Read more carefully” won’t change anything on the next question, because it was already true on this one. If your rule would fit every question on the test, it isn’t about yours yet. Keep asking why.
Run it once, on yourself
Before you watch anyone else, you’re going to be the witness. The argument is an everyday one I wrote for this page, no LSAT text, with one right answer and one trap built for a particular belief. Cover everything below the choices. Pick an answer for real, because the rest only works if you’ve committed.
In March, my gym added a 6 a.m. spin class. That same month it ran a half-price deal for new members. Morning attendance is up by a third since then. So the spin class is responsible for at least part of the jump.
Which of these does the argument need to be true?
(A) Spin classes have gotten more popular at other gyms in town.
(B) The half-price deal had no effect on morning attendance.
(C) The jump in morning attendance wasn’t caused entirely by the half-price deal.
(D) The gym plans to add a second spin class at 7 a.m.
(E) Most of the gym’s members live within a mile of it.
Committed?
The answer is (C). (B) is the trap, and it’s built for a reader who mistakes what would make an argument airtight for what the argument needs.
Your deposition may not read like the one below. If it doesn’t, go with yours, since that’s the reasoning that actually happened.
If you picked (B), here’s the whole card, compressed.
The deposition: “The deal was the obvious other explanation, and B shuts it down, so the argument needed it.”
The rebuild. Evidence: a spin class and a discount both started in March, and morning attendance rose by a third. Conclusion: the class is responsible for at least part of the rise. Words in the text that support (B): none. The argument never says the deal did nothing.
Where the point died: the stem. You understood the argument fine. You answered “what would make this argument impossible to doubt?” when the question asked what it needs.
The why-chain. Why did (B) feel safe? Because it rules out the rival explanation. Why did ruling it out completely feel required? Because if the deal did anything, the class’s credit looked shaky. Why did shaky feel like broken? Because I was checking whether the argument was airtight. Why was airtight the standard? That’s the bottom. You’ve been reading “what the argument needs” as “what would make it bulletproof.” Run the stop test on that sentence. It’s a belief about how arguments work, and it would explain other misses, every question where the biggest, most sweeping choice felt like the safe one.
Check it by flipping (B) to its opposite: the deal had some effect on morning attendance. Read the argument again with that in place. It still stands, because the class can own part of the jump while the deal owns the rest. The argument never needed (B).
The rule: when a question asks what an argument needs, I’ll negate my pick before I commit, and if the argument survives, it wasn’t needed. The trigger is any time the choice I like is the biggest claim on the page.
The forecast: this skeleton comes back on any argument with a modest conclusion, “at least part” or “some,” sitting next to a choice that rules out everything else. Next time it could be dressed as a diet or a store’s new hours.
The rematch: a date one week from today, written next to the question, with the skeleton logged as “treated airtight as needed.”
Now flip (C) the same way. The deal caused the entire jump. Then the class is responsible for none of it, and the conclusion falls. That’s what “needs” means. (A) and (E) say nothing about what caused the jump, and (D) is about the future. Chapter 6 has Ben running into this exact belief in a session, and the groan at the end of it.
If you picked (C), don’t skip ahead. A right answer can hide the same gap. Run the confidence audit in writing: one sentence on why (C) wins, and one on why (B) is wrong, sharp enough to convince the version of you from three weeks ago. If the second sentence comes out as “B just seems like too much,” you’ve found a soft spot that a harder question will find under time, and it earns the full twenty minutes. If it comes out as “the argument only claims part of the jump, so the deal can have helped too,” you’re done with this one.
The left column and the right
The left side of the card is yours. You can run it alone, on paper, for as long as you take this test, and nothing in it needs me. The right side is what a coach does, and every question in it is printed above. What a book can’t do is sit across from you, wait through eleven seconds of silence, and then ask a smaller question when the first one didn’t land. That’s the one part of the method that takes a second person, and it’s what every coach on my team does in a lesson.
Now watch it run on people who didn’t know they were in a protocol.
