Reading Pacing: 39 Minutes, Four Parts

The ratios, and why they mislead

Per question, the budget looks almost flat. Part 1 gives about 55 seconds. Part 2 gives about 60. Part 3 gives about 60. Part 4 gives about 72. On that table, Part 4 looks generous and Part 1 looks tightest.

The table lies, because a question is not a unit of work. What a question actually costs depends on how many places you have to look to answer it. A Part 3 statement costs one search of a mapped passage. A Part 2 blank costs one trip into the email for the constraint and one trip into the chart for the fact, inside the same minute.

Part 4's 72 seconds is also less generous than it looks, because it has to absorb the reading of the longest text in the section. Three of those twelve minutes are gone before you answer anything. What remains is closer to 54 seconds a question for ten questions that turn on who said what.

Rank the parts by real pressure rather than by the clock and the order changes: Part 2 is tightest, Part 4 second, Part 1 third, Part 3 the most comfortable. Plan for that order, not for the one on the timer.

A budget for each part

Part 1, about 10 minutes. Ninety seconds reading the message properly, straight through, without hunting. Then about three and a half minutes on questions 1 to 6, which follow the message in order. Then about four and a half minutes on the reply and its five blanks, with the last thirty seconds kept back for the one blank that fights you. The reply is the slower half and it needs the larger share, which is the opposite of how most candidates split the part.

Part 2, about 8 minutes. Forty-five seconds learning the shape of the chart: what the four options are and which dimensions are being compared. Sixty seconds reading the email for the person's constraints. Then about five minutes on the eight questions, leaving a little over a minute of slack. Notice that nearly two of the eight minutes go before you answer anything, and that this is correct rather than wasteful.

Part 3, about 9 minutes. One minute mapping the four paragraphs by topic, no more than that. Then roughly 50 seconds a statement for nine statements, keeping the hardest one for last. The mapping minute feels like a luxury and is the reason the rest of the part goes quickly.

Part 4, about 12 minutes. Three minutes reading the article and building a short ledger of who holds which position. Four minutes on the five sentence stems. Five minutes on the comment and its five blanks. Again the second block gets more time than the first, and again most candidates do the reverse.

The part that quietly eats the section

Part 2 is where the minutes go, and the reason is almost never reading speed. The chart is short. The email is short. The total word count in Part 2 is lower than anywhere else in the section.

What costs the time is reading the chart in the wrong order. Four options that are near-identical, each with three feature bullets and a price line, is a text that means nothing until you know what you are looking for. Candidates read all twelve bullets carefully, retain none of them because there was nothing to attach them to, and then read them again once the questions tell them what mattered. That is the chart read twice inside an eight minute part.

The fix is not to read faster. It is to read the chart for structure first and detail later: what are the four things, and what dimensions separate them. Then take the constraints from the email, then go back into the chart for exactly the two bullets that decide the blank in front of you.

The same failure happens in miniature in Part 4, where candidates read the article, understand it, and then discover at question 3 that they retained the argument but not who made it. Both are the same mistake: reading for meaning when the part needs reading for structure.

What to do when a part is already lost

Two minutes left, four questions to go, and the sensible plan has collapsed. This happens to almost everybody at least once, and what you do in those two minutes decides whether it costs one mark or four.

First, fill every unanswered drop-down immediately, before you do anything else. There is no penalty for a wrong answer, so an empty box is a mark you have chosen not to take. Filling four boxes takes ten seconds and converts a guaranteed zero into roughly one expected mark.

Second, spend the remaining time on the question closest to being solved, not on the first unanswered one. Order is not a virtue here. The question you had half-worked out is worth several times the one you have not read.

Third, do not carry the loss into the next part. The clock resets, the part is scored on its own questions, and a Part 2 that went badly has no arithmetic effect on Part 3 unless you let it have a psychological one. Candidates routinely turn one bad part into two.

Practising the clock rather than the content

Pacing is a separate skill from reading, and it needs separate practice. Doing whole timed tests trains it slowly and expensively, because a full reading section costs you 39 minutes and gives you four data points.

A cheaper drill: take one part, set the real timer for it, and do nothing else that day. Eight minutes on a Part 2. Twelve on a Part 4. You get the same pacing information in a fifth of the time, and you can repeat the part that is actually failing rather than the three that are fine.

Track one number per attempt, which is where you were when the timer hit half. In Part 1 you should be starting the reply. In Part 3 you should be on statement five. If you are consistently behind at halfway, the problem is in the opening move of the part, not in the questions at the end, and that is where to look for the fix.

Practise it

Five questions on the clock: where it runs out, why, and what to do about it.