Join the waitlist
Join the MCAT waitlist
Draft, sample content
CARS methodFor Pre-med stuck at 124 on CARS while the science sections climb

CARS is a structure-reading test: a method for pre-meds

CARS scores a map of the passage, not a memory of it. A paragraph-labeling method shown on an original passage, with timing for 9 passages in 90 minutes.

13 min readUpdated November 9, 2026By Rick, founder
The short version
  • CARS gives you 53 questions on 9 passages in 90 minutes, which is 10 minutes a passage; spend 4 of them building a map and 6 answering.
  • Label each paragraph with its job in the argument (claim, concession, example, pivot) in five words or fewer, never with its content.
  • Write the author’s claim in one sentence before you open the first question; most wrong answers fail against that sentence alone.
  • A 124 almost always comes from reading for facts and re-reading; four weeks of mapped passages, two a day, is enough to feel the change.

Your science sections are moving and CARS is parked at 124, and you have started to suspect that more reading will not fix it. You’re right: CARS scores whether you can see the structure of an argument, not whether you remembered its details, and the fix is to read for the shape of the passage and write that shape down. This article shows the method on an original passage and gives you the timing to run it nine times in 90 minutes.

What a 124 usually means

A 124 on CARS is not a reading problem in the ordinary sense. People who score there read fine. What they do is read the passage the way they read a textbook, trying to hold the content, and then meet questions that ask about the author’s purpose, the function of a paragraph, or how new information would land against the argument. The content they stored does not answer those questions, so they go back and re-read, and the clock goes.

You can check this on your own practice. Look at your last two CARS sections and sort the misses. If most of them are on questions with “the author would most likely,” “the primary purpose,” “which of the following best describes the function,” or “suppose that,” the issue is structure. If they are mostly detail-retrieval questions, it is something else and the method below still helps, but less dramatically.

The second signature of a 124 is time. You run out, guess the last passage, and tell yourself you need to read faster. Speed is not the problem. Re-reading is.

What the section actually is

Per the AAMC, Critical Analysis and Reasoning Skills is 53 questions on 9 passages in 90 minutes, with passages drawn from the humanities and social sciences and no outside knowledge required (the official description is at students-residents.aamc.org). The AAMC sorts the questions into Foundations of Comprehension, Reasoning Within the Text, and Reasoning Beyond the Text, and the last two categories, the ones about argument and application, carry the majority of the questions [verify].

Fifty-three questions in 90 minutes is about 1 minute 42 seconds a question. Nobody thinks that way during the test. The usable figure is 10 minutes a passage, which leaves no slack, so the method has to be fast as well as accurate.

The mapping method

The method has five moves. The whole thing takes about four minutes a passage once it is habit.

Read the first paragraph slowly. The opening usually sets up either the author’s claim or the view the author is about to push against, and you need to know which before anything else. Write a label of five words or fewer in the margin of your noteboard: not what the paragraph says, but what it does. “Common view set up” or “Author’s claim” or “Historical example.”

Read each later paragraph for its job. Ask what the paragraph is doing to the argument: extending it, complicating it, conceding something, offering an example, turning against the opening. Label it the same way. Pivot words carry most of the structure, so circle “but,” “however,” “yet,” “although,” “admittedly,” and “in fact” as you go. A passage with a pivot in paragraph three is a different passage from one with a pivot in paragraph one.

Track voices. Authors quote and paraphrase people they disagree with, and the single most common wrong answer on CARS attributes to the author a view the author was describing in order to reject. Mark anything that belongs to someone else with a different symbol than the author’s own statements.

Write the claim. Before you touch a question, write one sentence that states what the author believes and, if there is one, the qualification the author puts on it. This is the sentence the questions will be tested against. If you cannot write it, you are not ready for the questions and you should re-read the pivot paragraph, not the whole passage.

Go to the questions with the map. Each question tells you where to look. A function question points at one label. A claim question points at your sentence. An application question asks you to hold the sentence steady and see what the new information does to it. The map is a table of contents you wrote yourself.

Four minutes building this feels slow the first week. It is faster than reading twice.

Worked example: an original passage

The passage below was written for this article. It is shorter than a real CARS passage, which typically runs 500 to 600 words [verify], but the structure is the same.

Public clocks are usually described as instruments of control: the factory bell, the railway timetable, the tower that told a town when to work. On this account, standardized time was something imposed on ordinary people by institutions that profited from their punctuality, and the resentment it produced is taken as proof.

The record is more complicated than the account allows. Petitions from nineteenth-century market towns asked for clocks to be installed, repaired, and lit at night, and complained when they were not. Traders wanted a shared time because disputes over when a market opened cost them money. Churchgoers wanted it because arriving late was a public embarrassment. The clock was demanded from below at least as often as it was imposed from above.

This is not to say that standardized time served everyone equally. The same shared hour that let a trader hold a rival to an agreement let an employer dock a worker’s wages, and the worker had less recourse. But the existence of losers does not make a technology an imposition. It makes it a technology.

What the control account gets wrong, then, is not its facts but its direction. It reads the resentment of some users as the intention of the designers, when the more defensible reading is that a tool wanted by many was used, as tools are, by whoever could reach it.

The map, then the questions

Here is the map we would write on the noteboard. Four lines, under five words each.

Paragraph Label Whose voice
1 Control view set up Others’
2 Pivot: demanded from below Author’s
3 Concession, then re-pivot Author’s
4 Author’s claim restated Author’s

The claim sentence: the author thinks standardized time was wanted by ordinary people and that the “control” account mistakes some users’ resentment for the designers’ intent, while conceding it served some people better than others.

Now watch how the usual question types fall against the map.

A main-idea question offers answers like “public clocks were instruments of institutional control.” The map kills it instantly, because that is the paragraph-1 view, marked as someone else’s. The right answer has to contain both the pivot and the concession, and any choice that says the author “rejects” or “disproves” the control account is too strong, because paragraph 3 conceded part of it.

A function question asks why the author mentions the employer docking wages. Content readers answer “to show that clocks hurt workers.” The map says paragraph 3 is a concession, so the wages example is there to grant the opposing side its best point before turning on it. The answer is “to acknowledge a limitation of the author’s position.”

An author-attitude question asks how the author regards the control account. Not hostile, not dismissive. The last paragraph says it gets its facts right and its direction wrong. The answer will be something measured like “partly mistaken” or “insufficiently careful,” and choices with “naive” or “fraudulent” are too hot.

An application question says: suppose a historian found that most town clocks were paid for by factory owners. Does this weaken the author? Hold the claim sentence steady. The author’s point is that clocks were demanded from below, not that they were funded from below, so the new fact is compatible with the argument. A 124 reader feels the word “factory” and picks “strongly weakens.” The map says otherwise.

Timing for nine passages in 90 minutes

The map only works if you can afford it. Here is the clock.

  • Four minutes reading and labeling. That includes the claim sentence. If you are at six minutes, you are reading for content again.
  • Six minutes on five to seven questions. Roughly one minute each, with the extra going to the one application question that deserves it.
  • Zero minutes re-reading the whole passage. Return to one labeled paragraph when a question sends you there, read that paragraph, come back.
  • Nine passages in order. Skipping around costs more than it saves on CARS because every passage is a fresh argument; there is no “easy type” to hunt for.
  • A hard stop at 10 minutes. If a passage is going badly, answer with the map and move on. One lost passage costs you about six questions. A blown clock costs you two passages.

Practice the clock on every passage from the first week, even while the method is slow. A method you cannot run in 10 minutes is a method you do not have.

Four weeks of practice

Here is how we would build the habit if you were sitting across the table.

Weeks one and two: two AAMC passages a day, untimed, map on paper, claim sentence written before the questions. Review every question, right or wrong, by asking which labeled paragraph answered it. Wrong answers get a one-phrase cause: “took paragraph 1 as author’s view,” “chose a too-strong verb,” “missed the concession.” Keep the causes in a running list.

Week three: two passages a day, timed at 10 minutes each, same review. Your cause list from weeks one and two should be short enough now to read before each session. That list is your personal set of traps.

Week four: full 90-minute CARS sections, one every other day, with the map still written out. Review the next morning, not the same day. Count how many misses fall into your old cause categories and how many are new.

Use AAMC passages for all of this. Third-party passages can train speed, but their questions are written by people guessing at the AAMC’s logic, and the wrong-answer patterns are not the same.

What to do this week

  1. Sort the misses on your last two CARS sections by question type. If function, purpose, attitude and application questions make up most of the misses, you have a structure problem and the method applies directly.
  2. Do two AAMC CARS passages a day for five days, untimed, writing a label of five words or fewer per paragraph and one claim sentence before the questions.
  3. Start a cause list. Every wrong answer gets one phrase describing what you did, not what the right answer was.
  4. On day six, time one passage at 10 minutes using the 4-and-6 split. Note where the time went.
  5. If you want to hear when our MCAT program opens, join the waitlist. Until then, the AAMC passages and a noteboard are everything this method needs.

Questions people ask

Should I read the questions before the passage on CARS?

No. CARS questions test the argument as a whole, so previewing them turns reading into a scavenger hunt for details and you lose the structure. Build the map first, then let each question send you back to a specific labeled paragraph.

How many CARS passages a day should I practice?

Two, mapped and reviewed properly, beats six done fast. The review is where the method is learned. Once the map is habit, usually after three or four weeks, move to full timed sections to build pace.

Does CARS require background knowledge in humanities or philosophy?

No. Every question is answerable from the passage and the AAMC says so explicitly. Unfamiliar topics feel harder because you cannot lean on content, which is exactly why mapping structure instead of content works.

What score is realistic on CARS from a 124?

We will not name a number for a student we have not seen. What we can say is that the move from 124 to the high 120s is usually a method change, not a reading-speed change, and method changes show up within weeks when practiced on real AAMC passages.

Written by Rick [Last name], founder and tutor at Giant Prep. 99th percentile on the tests he teaches, in every session himself. About Rick Scores pending