Speed drills on real passages

Reading speed is not a talent you hope arrives before test day — it is built with a stopwatch, on real passages, against targets. But "read faster" is useless advice, because slowness has three different causes and each needs a different drill. This lesson gives you the three timed formats and tells you which one your symptoms prescribe.

Three bottlenecks, three drills

Time a slow candidate and you find the minutes leaking from one of three places:

Diagnose before you train: on your next practice passage, time the three phases separately (first skim / locating / deciding) for a handful of questions. The fattest number is your bottleneck. Training the wrong one feels productive and changes nothing. If you keep an error log, your cause column has probably already told you.

Drill A — paragraph gist sprints (comprehension speed)

The skill: extracting a paragraph's point on one pass, no regression. This is the engine under matching headings and under every fast first skim.

1. Take one passage you have not read. Set a timer for 60 seconds per paragraph.
2. For each paragraph, read once — first and last sentence fully, middle at speed — and write a 3–5 word label: why the theory failed, costs of the project.
3. Hard rule: when the 60 seconds end, the label gets written from whatever you have. No second pass. The discomfort of writing from incomplete reading is the training stimulus — it forces your eyes forward and teaches you how much comprehension one pass actually yields (more than you think).
4. Afterwards, check your labels against the paragraphs. A wrong label means you missed a pivot (however, yet) — note where it hid.

Progression: 60 → 45 seconds per paragraph over two weeks. A full 7-paragraph passage skimmed and labelled in under 5 minutes is test-ready.

Drill B — scan sprints (location speed)

The skill: finding a specific word-shape in 950 words without reading them. Pure mechanics of location discipline.

1. Take any passage, including ones you have done before (comprehension is not the point). From its questions, list five anchors: two names or numbers, three content nouns.
2. Timer: 90 seconds for all five. Scan with a moving finger or cursor, faster than reading pace. Mark each find.
3. Score by finds, not elegance. Missed one? Note whether the anchor was paraphrased in the passage (a keyword-choice error — you picked a scannable word that wasn't verbatim) or simply overlooked (a pure speed issue: repeat the sprint).
4. Advanced round: scan for the paraphrase of an anchor — list funding, find investment. This wires your vocabulary families directly into your eyes.

Progression: 90 → 60 seconds for five anchors. This drill is cheap — three minutes including setup — so it fits into days when a full passage won't.

Drill C — forced-pace passages (decision speed)

The skill: committing. Hoverers lose their minutes after location, re-reading a found sentence four times hoping certainty will arrive. It never does; the drill teaches you to decide at 80% certainty, which is where most correct answers are made.

1. Take a full passage with its questions. Your normal share would be 20 minutes; set the timer at 15.
2. Sit it under one rule: every question gets an answer the moment its located sentence has been read twice. Two readings, then commit — mark a small dot next to answers where you felt under 80% sure.
3. Score it, then look only at the dotted answers. The reliable finding: most dotted answers are right. That is the lesson — your 80% instinct is calibrated, and the third and fourth readings you used to buy were purchasing comfort, not accuracy.
4. Any dotted answer that was wrong goes through the paraphrase-pair drill: the transformation you couldn't see at speed is the one to train.

Progression: 15 minutes → 13. Do not go below 13; past that you are training recklessness, not decision speed.

Scheduling the drills

• One drill per day, not three. A: 8–10 minutes; B: 3 minutes; C: 20 minutes with scoring. Rotate according to your bottleneck — weight it 3:1 over the others.
• Speed drills complement full tests; they never replace them. A good week: two full timed tests, three drill days between them.
• Re-run the three-phase diagnosis every two weeks. Bottlenecks move — fixing location speed often reveals a decision-speed problem that was hiding behind it.

Your drill

Today, the diagnosis plus your first prescribed drill — about 25 minutes.

1. Sit one passage of Reading 2026-05 Test 6 with rough phase timing: note the clock after your first skim, after locating each of the first five questions, and after deciding each. Identify the fat phase.
2. Run the matching drill immediately: Drill A on a fresh passage from Reading 2026-02 Test 4, or Drill B using its questions' anchors, or Drill C on the full passage at 15 minutes.
3. Write down today's numbers — seconds per paragraph, per anchor, or minutes per passage. Speed training without recorded numbers is jogging, not training.
4. Repeat your prescribed drill three times this week on passages from Reading 2025-12 Test 2, then re-diagnose.

The stopwatch is the whole method. Same drill, same measurement, falling numbers — and the hour that used to end at question 33 starts ending at question 40 with minutes to spare.