Exam Day: The Three Hours Decide More Than You Think

Two students walk into the same hall knowing the same amount. One comes out with a noticeably better mark.

The difference is not luck and it is not usually nerves. It is what they did with the three hours - which question they started with, how long they let a hard one hold them, whether they wrote four points for a four-mark question or one dense paragraph, and whether they left anything blank.

None of that is knowledge. All of it is trainable, and almost nobody trains it.

Marks Per Minute Is the Only Thing That Matters

Everything in this post follows from one idea, so it is worth stating plainly.

In the hall you are not trying to answer questions correctly. You are trying to collect the most marks in the time available. Those sound identical and they are not, and the difference is where a great many marks are lost every year.

A student who spends twenty-five minutes perfecting a six-mark question and then leaves two four-mark questions blank has answered one question beautifully and lost more than they gained. A student who wrote reasonable answers to all three collected more.

Work out the paper's rate before the exam. If the paper carries 100 marks in 180 minutes, that is roughly 1.8 minutes per mark, minus reading and checking time. Call it 1.5 minutes per mark of writing. A four-mark question gets six minutes. A ten-mark question gets fifteen. Knowing those two numbers changes how the whole paper feels, because you stop guessing whether you are behind.

Write the times on the paper. In the first minutes, next to each section, write the clock time you should be finishing it. Not the duration - the actual time, so you never have to do arithmetic under pressure. This is the single most effective thing in this post and it takes ninety seconds.

The First Five Minutes

What you do before writing decides a surprising amount.

Read the whole paper. Not skim - read. You are looking for three things: which questions you can definitely do, which you cannot, and whether anything is different from what you expected. Five minutes spent here is not lost time, because it removes the worst outcome in an exam, which is discovering an easy question with four minutes left.

Mark the paper up. Tick the questions you can do straight away. Cross the ones you cannot. Circle the ones you think you can do with effort. You have now built a route through the paper.

Check the instructions properly. How many questions from each section. Whether there is a choice. Whether there is negative marking. Every year, students answer more questions than required, or fewer, or answer both options where only one counted. These are not knowledge failures and they are expensive.

Look at the mark allocation on every question. It tells you how much to write, and it is printed right there. A one-mark question wanting a phrase and a five-mark question wanting five points look similar on the page and are completely different tasks.

Answer in Your Own Order, Not the Paper's

The paper is printed in an order that suits the setter. It has no obligation to suit you.

Start with a question you can definitely do. This is partly about marks in the bank and mostly about state of mind. Starting with something you can handle settles you; opening with the hardest question on the paper creates a fifteen-minute spiral that damages everything after it. Managing your own test anxiety is a strategic act, not a soft one.

Then work in descending order of marks-per-minute. The questions you know well and that carry good marks come next. Save the ones you circled for later, when the certain marks are already collected.

Number your answers clearly and leave space. If you answer question 7 before question 3, write the number large and unambiguously. A marker who cannot tell which question you answered cannot award it.

Never leave a question you can do until last. The end of an exam is the worst part of it - you are tired, the clock is loud, and your judgement is poor. Spend that time on the hard ones, not on marks you had already earned.

The Rule for Getting Stuck

This is where most time is lost and there is a simple discipline that fixes it.

Give a question one attempt, then leave it. If you have not found the way in after the time it deserves, stop, mark it in the margin, and move on. Come back with the rest of the paper done.

Two reasons this works. First, the arithmetic: five minutes overspent on a hard question is five minutes not spent on an easy one somewhere else, and the easy one was worth the same. Second, and less obvious, coming back to a problem after twenty minutes on something else genuinely helps - you re-read the question without the assumption that defeated you the first time.

The Zeigarnik effect is worth knowing here: an unfinished task keeps a hold on your attention. That is normally a nuisance, and in an exam it works for you - your mind keeps turning over the question you abandoned while your pen works on another, and you often see the way in on returning.

Write something rather than nothing. If a numerical defeats you, write the formula and what you were given. Method marks exist on many papers, and a blank scores zero with certainty while a partial attempt sometimes does not.

Match the Answer to the Marks

The most reliable way to waste time in an exam is to over-answer.

One mark wants a phrase. Not a sentence, not a paragraph. Write the phrase and move on.

Four marks want four identifiable points. Not one paragraph containing four ideas - four things a marker can find and tick. Number them or start each on a new line. A marker awarding marks is looking for distinct pieces, and pieces they cannot find are not awarded.

Ten marks want structure. An opening line that answers the question, then the points, then a closing line if the question asked for a conclusion or an opinion. A wall of prose with the same content in it scores lower, reliably.

Do what the instruction word says. "State" is not "explain". "Describe" is not "compare". "Give reasons" means the reasons carry the marks, not the description. This is the most common way a well-prepared student answers a slightly different question from the one asked and loses marks they had earned. If your paper has a published marking scheme or your teacher has one, reading it once tells you more about this than a term of advice.

Diagrams where a diagram helps. In Science and Geography a labelled diagram often earns marks directly and always earns them faster than prose. Draw it, label it, spell the labels.

Read the Question Twice, and Read the Verb

More marks are lost to misreading than most students believe, and it is not a comprehension problem - it is a speed problem under pressure.

Read every question twice before answering. The second read takes four seconds and catches the thing the first read assumed. Under time pressure your brain fills in what it expects the question to say, and expectation is exactly what a well-set paper plays against.

Underline the instruction verb and the quantity words. "State two reasons" - underline two. "Explain why, with reference to the diagram" - underline with reference to the diagram, because an answer that ignores it loses marks that were sitting there. This is the same skill as reading comprehension in a language paper and it transfers to every subject.

Watch for questions that contain two tasks. "Define X and give an example" is two marks in two parts, and students routinely do the first half beautifully and forget the second. If a question has an "and" in it, put a small mark next to each part and tick them off.

Do not answer the question you revised. This is the subtle one. If you prepared a topic thoroughly, a question that touches it will pull you towards writing what you know rather than what was asked. A brilliant answer to a neighbouring question earns very little, and it is one of the commonest ways strong students underperform.

Watch the cognitive load you create for yourself. Trying to hold the question, the method and the arithmetic in your head at once overloads working memory. Writing down the given values as you read offloads part of it onto the paper, which is why the students who scribble first tend to make fewer careless errors - they reduced their own cognitive load before starting.

Multiple Choice and Negative Marking

If your paper has an objective section, two facts change everything about how you handle it.

Without negative marking, never leave a blank. A guess has positive expected value. Answer every single one, even the ones you know nothing about. Students leave objective questions blank every year for no reason at all.

With negative marking, a wild guess is negative expected value. Eliminate what you can. If you can rule out two of four options, guessing is usually still worth it. If you can rule out nothing, leaving it blank is correct. Find out which regime your paper uses before the day, not in the hall.

Answer the objective section fast and do not revisit. Multiple choice rewards recognition, and your first instinct on a recognition task is right more often than your third revision of it. Changing answers repeatedly on objective questions costs marks more often than it gains them.

Watch for the transfer. If answers go on a separate sheet, check the numbering every ten questions. An off-by-one error propagating through thirty answers is a catastrophe and it happens.

Practise the Clock, Not Just the Content

The reason exam timing goes wrong is that almost nobody rehearses it.

Students do practise questions with no clock, in comfortable conditions, with the answers within reach. Then they sit a three-hour paper for the first time on the day it counts. That is like training for a race by walking.

Sit at least three full papers to time before the exam. Full length, one sitting, no phone, no book, no breaks. It is unpleasant and it is the point - the fatigue in the third hour is real and you need to have met it.

Mark them strictly, or better, have someone else mark them. Marking your own work generously is the most common way a preparation that looks strong hides a real gap.

Track two numbers separately: marks lost to not knowing, and marks lost to running out of time. They need completely different responses. More revision fixes the first and does nothing for the second. If your timing losses are not falling across three papers, the problem is your allocation discipline, not your knowledge - and it is more fixable than the other kind.

Practise the whole three hours, including the writing. Physical writing stamina is real. If you have only ever written for forty minutes at a stretch, your hand and your handwriting will both degrade in the third hour, and a marker cannot award what they cannot read.

The Last Twenty Minutes

Plan them in advance, because your judgement at that point will be poor.

Reserve them. Genuinely reserve them - build them into the clock times you wrote at the start. This is the single hardest piece of discipline in an exam and the most valuable.

First, fill every blank. Any question with nothing written gets something. A formula, a definition, a partial attempt, a guess on an objective item where guessing is free. Blanks are the only guaranteed zero on the paper.

Then check the mechanical things, in this order: that you answered the required number of questions; that your answer numbers match the paper; that units are on your final answers; that nothing on a separate sheet is misnumbered. These are checks with a high hit rate and they take seconds.

Then check the arithmetic on the numericals. Recompute, do not re-read. Reading over a calculation almost never finds an error because you follow the same path and make the same mistake. Redo the line.

Do not rewrite a good answer to make it better. The marginal mark from polishing an answer that already earns most of its marks is close to zero, and the time would earn more anywhere there is a blank.

The Practical Things That Go Wrong

Not strategy, just the recurring avoidable disasters.

Arrive early enough to be bored. Rushing in raises your heart rate for the first twenty minutes of a paper you needed to read carefully.

Bring two of every pen, a pencil, a ruler and an eraser. A pen failing in the second hour has ruined papers.

Check what your exam permits - calculator, log tables, geometry instruments - and bring exactly that.

Eat, and sleep the night before. Not because it is nice advice, but because sleep deprivation measurably degrades working memory, which is the faculty you use to hold a multi-step problem together. An all-nighter trades a few facts for the ability to use any of them.

Do not discuss answers between papers. It changes nothing about the paper you finished and it damages the one you have not sat. This is the most common way a good student wrecks their second exam.

Get the actual timetable from the official source. Rumours about dates and pattern changes circulate every year and are frequently wrong; the National Examinations Board publishes the real thing.

The Days Between Papers

In an exam series most of your remaining marks are decided between papers, not in them, and this period is badly used almost universally.

The paper you just sat is finished. Nothing you do changes it. Reconstructing your answers, comparing with friends, and calculating what you probably scored are all activities with a guaranteed return of zero and a real cost to the next paper. This is the discipline that separates a good series from a good first paper.

Sleep before you study. After a three-hour paper you are genuinely depleted. An hour of rest and then two focused hours beats four hours of tired staring, and the difference is larger than it feels.

Revise for the next paper by retrieval, not by reading. With one or two days available, reading a chapter is close to useless. Closed-book self-testing on the highest-weight topics is what moves a mark in a short window, because the testing effect works fast and re-reading does not.

Prioritise ruthlessly by weight. With limited time, the topics that carry the most marks and that you are weakest on are the whole agenda. This is time management at its most brutal and its most useful: something will not get revised, and you should choose which rather than letting the alphabet choose for you.

Do one timed section, not a whole paper. A full mock between papers costs more energy than it returns. Forty minutes of one section under the clock keeps the timing discipline sharp without emptying you.

A One-Page Plan to Take In

Everything above compresses to this. Learn it, and the three hours stop being improvised.

1. Read the whole paper. Tick, cross, circle. Check the instructions and the choice rules.
2. Write clock times next to each section. Actual times, not durations.
3. Start with something you can do. Never with the hardest question.
4. Match the answer to the marks. One mark, a phrase. Four marks, four findable points.
5. One attempt on a hard question, then move on. Mark the margin and come back.
6. Never leave a blank. Formula, definition, guess where guessing is free.
7. Reserve the last twenty minutes. Blanks, then mechanical checks, then recompute the numbers.

The Short Version

The hall does not test what you know. It tests what you can get onto paper in three hours, and that is a separate skill from knowing it.

Work out the paper's marks-per-minute before the day and write actual clock times beside each section. Read everything before you write anything. Start with a question you can do, give a hard one exactly one attempt, and match the length of every answer to the marks printed next to it.

Then reserve the last twenty minutes and spend them on blanks and mechanical checks rather than polishing something that was already fine.

And sit three full papers to time before the real one, tracking marks lost to not knowing separately from marks lost to running out of time. Those two problems look identical on a total and they have nothing in common - and most students spend a whole month treating the second one as the first.