Two days out and you haven't started

No lecture about how you should have started earlier. You know. It's Thursday, the exam is Saturday morning, and you need a plan for the next 48 hours.

The instinct is to open chapter 1 and start reading fast.

Don't. With two days you cannot cover everything, so the only question that matters is which parts you're going to deliberately not cover — and answering it takes thirty minutes and is worth more than any three hours of reading you could do instead.

The first thirty minutes: sort, don't study

Get the topic list. Syllabus, course schedule, lecture titles — whatever gives you the actual list of what's on the exam.

For each topic, two judgments. How many points is it likely worth? And how far am I from being able to do it?

Then sort into three piles.

Do first. Heavy on the exam and reachable from where you are. This is where most of your 48 hours go.

One pass. You half know it. It needs one recall attempt and a fix, not a rebuild. These are the cheapest points on the table and people skip them because they don't feel urgent.

Skip. Low weight, or so far from where you are that you'd burn six hours for four points. Choosing this pile is the entire skill, and it's why you do the sort before you start.

Write the piles down. When it's 1 a.m. and you're tired and drifting toward the interesting-but-worthless topic, the paper is the thing that stops you.

Then: problems, not reading

With two days, reading is close to worthless. It's slow, it produces recognition rather than ability, and recognition is not what you'll be asked for.

Do problems. Old exams if you have them, homework problems, end-of-chapter questions, anything with an answer you can check.

The loop is: attempt closed-book, check, and for anything you got wrong, go read only the part that explains that specific thing. That's targeted reading, and it's the only kind worth doing right now.

If a topic is so new that you can't attempt a problem at all, read the worked example first, then immediately do a different problem of the same type with the example closed. Not "read three examples and move on" — the closed-book attempt is the part that does anything.

The 48-hour shape

Thursday evening. Triage. Then start on the first pile. Two or three hours of problems. Stop at a normal hour.

Friday morning. Best hours of the whole window. Hardest material from the first pile.

Friday afternoon. Finish pile one, start pile two. Faster passes, less depth.

Friday evening. Finish pile two. Then, if you have anything left, one timed practice section under real conditions. Sleep at a normal time.

Saturday morning. Do not learn anything new. Review the things you got wrong yesterday, look over formula sheets or key definitions, eat something. Show up early.

That last block is where people do the most damage to themselves. Cramming a brand-new topic three hours before an exam gives you a shaky version of one thing and destabilizes everything you already had. New material on exam morning is a bad trade every time.

Sleep

The all-nighter is the single most expensive habit in this whole guide, and American campuses have a whole folklore around it.

Two reasons it doesn't work. Sleep is when memory consolidates — the studying you did on Friday partly doesn't take without it. And an exam is a multi-hour reasoning task, which is exactly the thing sleep deprivation destroys first.

The trade is a few hours of low-quality studying against a large hit to the thing doing the exam. If you're deciding at 2 a.m. between one more topic and six hours of sleep, take the sleep. It's not the disciplined-sounding answer and it's the right one.

What to skip entirely in 48 hours

• Rewriting notes. Zero value under time pressure. It feels like work because your hand is moving.
• Making new flashcards. Making cards is a long-horizon investment. Two days out you should be using cards, not building them.
• Reading the textbook cover to cover. You will not finish and you'll have covered the first third four times.
• Reorganizing your files. This is procrastination wearing a costume, and it is remarkably convincing at 11 p.m.
• Group study, mostly. With time this abundant it's great. With 48 hours it usually becomes a two-hour conversation. The exception is a specific, bounded session: one hour, everyone explains one topic, done.

If a lot of it is genuinely gone

Sometimes the honest assessment is that half the course never got learned.

In that case, get very selective. Pick the three highest-weight topics and learn them properly rather than covering nine topics badly. Partial credit is real in most courses, and a well-executed answer on a heavy question beats four half-answers.

And on the exam itself: answer what you know first, all of it, before touching anything hard. Points are points regardless of order, and the classic disaster is spending twenty minutes on question 2 and never reaching questions 8 through 12, which you could have done.

Afterward

When it's over, write down what happened while it's fresh. Which topics you'd misjudged in the triage. Where the time went. Whether the format was what you expected.

That takes five minutes and it's the only thing that makes the next one different. Otherwise you'll be here again in six weeks, doing the same triage with the same bad estimates.