Probability

Probability is the maths of "how likely is this to happen?" — and once you understand the language, the rest of the chapter becomes a recipe rather than a guessing game. By the end of this post you will be able to describe any event in probability terms and write down its probability as a number between 0 and 1. Let us start at the very beginning so nothing in later posts trips you up.

The basic definition in one line

Probability measures how likely an event is to happen. We assign every event a number from 0 to 1, where 0 means the event is impossible and 1 means it is certain. A probability of 0.5 means the event is just as likely to happen as not to happen. Everything else in this chapter builds on that single idea, so it is worth slowing down until you are completely comfortable with it.

In Grade 12 you will not be asked to define probability; you will be asked to calculate it. But you must keep the definition in mind, because it tells you what your final answer should look like. If your working produces a probability of 1.4 or −0.2, you immediately know something has gone wrong. A probability outside the range 0 to 1 is always a signal to check your working, not a sign that you have discovered a new kind of maths.

The sample space, where every probability question begins

The sample space, written S, is the list of everything that could possibly happen in the experiment. If you roll a normal six-sided die, the sample space is {1, 2, 3, 4, 5, 6}. If you toss one coin, the sample space is {H, T}. If you draw a card from a standard pack, the sample space is the 52 cards. Everything we do in probability starts by identifying S, even when the question does not explicitly ask for it.

A common exam trick is to give a question that seems to be about one event but where the sample space has more outcomes than learners expect. For example, "What is the probability that a number rolled on a die is even?" looks like it involves only the three even numbers. But the denominator is still 6, because the sample space is all six possible rolls. You compare the favourable outcomes against the full sample space, not against a shortened version of it.

Writing probability as a number

There are three acceptable ways to write a probability in Grade 12: as a fraction, as a decimal, or as a percentage. A probability of one half can be written as 1/2, 0.5, or 50%. In the final exams most mark schemes accept any of these, but you must use the same form throughout a single question. Mixing fractions and decimals in one line of working will cost you marks even if your final answer is correct.

Two rules about the number itself will save you repeatedly: the probabilities of all the outcomes in the sample space always add up to 1, and the probability of the complement of an event (everything that is not the event) is 1 minus the probability of the event. These two facts are extremely useful for checking your work in long questions.

A simple worked example with a die

Let us calculate the probability of rolling a number greater than 4 on a fair six-sided die. We will use the three-step method that almost every basic probability question follows: count the sample space, count the favourable outcomes, and divide.

What probability is not

Probability is not a guess. It is a calculation based on the assumption that every outcome in the sample space is equally likely. When you read "a fair die" or "an unbiased coin", the word "fair" tells you that assumption is in place. When the question says a die is biased, the calculation changes — you would be given the probabilities of each outcome, and you would no longer divide by the size of the sample space. Most Grade 12 questions assume fairness unless told otherwise, but always check for the word "biased" before you start.

Probability is also not the same as odds. "Odds" is a separate idea (a ratio of favourable to unfavourable outcomes) that you may see in the CAPS ATP for background reading, but you will not be asked to compute odds in the final exam. Stick to probability as a fraction, decimal, or percentage, and you will stay inside what the mark scheme expects. Once you are happy with this foundation, the next posts will show you how to handle more interesting questions: combinations, dependent events, and the words that confuse learners most.

Before you read any further, take a piece of paper and try the question below. It looks short, but it tests the most important skill in the whole chapter: turning a word description into a probability calculation. Spend two minutes on it before you scroll down to the solution — that two minutes of struggling is where the real learning happens.

The question to try first

Here is the question. Try it on your own first, with a calculator if you like, and write down your final answer as a simplified fraction. Do not peek at the section below until you have at least made an attempt.

How to think about it before calculating

Before you reach for a formula, slow down and ask yourself two questions. First: what is the sample space? In other words, how many marbles are there in total? Second: what does "not red" actually include? Many learners freeze on the word "not" because they want a positive description, not a negative one. The trick is to convert "not red" into "blue or green", which is something you can count directly.

There is a second way to think about it that is just as valid: use the complement rule. The probability of "not red" is 1 minus the probability of "red". Both routes should give you the same answer, and in the exam you can use whichever one feels faster on the day. Checking the two routes against each other is also an excellent way to confirm your final answer.

Reveal: the worked solution

Let us walk through both routes so you can see they agree. Route 1 (direct counting) first.

Now Route 2, using the complement rule.

Both routes give 1/2, which means we can be confident the answer is correct. In the exam, choose whichever route feels faster for the question in front of you. If the question had asked for "not red" in a bag of 5 red, 3 blue and 1 green, the complement route would still be quick, but the direct route would also work — there is no wrong choice between the two.

Why this question matters

This question is short, but it tests three skills at once. You had to identify the sample space correctly (ten marbles, not three colours). You had to interpret "not red" as a countable set. And you had to simplify your final fraction. Every harder probability question in Grade 12 will use those same three skills in a more complicated setting, so make sure you are confident with the basics before moving on.

If you got 1/2, well done. If you got something else, check whether you used 3 as the denominator (you treated the colours as the sample space instead of the marbles), or whether you added instead of multiplying when you counted "blue and green". Both are very common slips, and they are exactly the kind of mistake that disappears with a few more practice questions.

What to take into the next post

Carry three things forward from this question. First, always count the full sample space before you do anything else. Second, when you see "not", ask whether the complement rule would be quicker than direct counting. Third, simplify your final fraction — examiners expect it, and an unsimplified fraction can cost you a mark even when the value is correct.

In the next post we will look at the single mistake that costs Grade 12 learners more probability marks than any other. If you have ever lost marks on a probability question and not been sure why, that post is for you.

If I had to name one mistake that costs Grade 12 learners more probability marks than any other, it would be this: forgetting that "without replacement" changes the second probability. It is easy to do, and the working looks correct, which is exactly why it slips past the marker. Let us look at where it comes from and how to never make it again.

The mistake in one sentence

Here is the mistake, said plainly: when a question says "without replacement", learners reach the second draw and use the same fraction they used for the first draw. They have forgotten that the bag has got smaller, so the denominator must change. The first draw was, say, "4 out of 10", and the second draw should be "3 out of 9", but they write "4 out of 10" again. The multiplication looks neat, the answer looks reasonable, and the mark is gone.

This mistake is so common that some examiners actively design questions to catch it. If you see the words "without replacement", "drawn one after the other", or "and is not replaced", treat them as a warning bell. The denominator of every fraction after the first one must drop by one each time.

Why it is so easy to make

There are two reasons. The first is that the language is sneaky: "without replacement" is a short phrase, and it does not feel as important as it actually is. The second is that the multiplication step looks the same whether you do it correctly or not. You still multiply two fractions together; the difference is hidden in the second denominator, which is one number off. A marker scanning your work might not even notice, especially if you have not labelled which draw each fraction comes from.

The third reason is that "with replacement" questions do exist, and they use the same fraction twice — so a learner who has done a few "with replacement" questions can fall into autopilot when they meet a "without replacement" one. Treat the two as different species. With replacement means same fraction; without replacement means changing fractions.

The wrong version vs the right version

Let us make this concrete. A bag contains 4 red balls and 6 blue balls. Two balls are drawn, one after the other, without replacement. Find the probability that both balls are red.

Notice the difference: 0.16 versus roughly 0.133. The wrong answer is bigger because it pretends the bag still has four red balls left after one has been removed. The right answer is smaller because, having taken one red out, there are fewer reds left to draw.

The rule of thumb that fixes it

Here is a rule you can write on the inside of your exam pad: without replacement means every fraction after the first one has its denominator reduced by one. So in a four-draw sequence from a bag of 12, your denominators go 12, 11, 10, 9. Your numerators change according to what was drawn — they go down by one if you drew the kind of object you are tracking (a red after a red, say), but they stay the same if you drew a different kind (a blue after a red). This rule is not a substitute for thinking, but it is a useful sanity check: if your denominators are all the same and the question says "without replacement", you have probably made the mistake.

If the question says "with replacement", the denominators stay the same. If the question does not say either, assume "without replacement" for objects drawn from a bag, because that is the default in CAPS exam questions unless told otherwise.

How to check yourself in the exam

The fastest check is to label each fraction with the draw it represents. Instead of writing "(4/10)(3/9)", write "Draw 1: 4/10. Draw 2: 3/9." That tiny bit of labelling makes it impossible to use the same fraction twice by accident, because the labels would shout at you. It also makes your working easier for the marker to read, which is always good for your mark.

Another check is to ask, after every fraction, "How many are left in the bag now?" If your second draw uses a denominator of 10 in a ten-ball bag, ask yourself: did I put the first one back? If the answer is no, your denominator must be 9. This is a five-second check that catches almost every instance of this mistake. Make it a habit and you will stop losing these marks for good.

If probability has felt like a guessing game so far, this is the post where it becomes a recipe. There is a single three-step method that solves most of the basic probability questions you will meet in Grade 12, and once you can write it down without thinking, the topic suddenly gets a lot easier. Let us walk through it together with a card example you will see in almost every test.

The three-step method

Almost every basic probability question in Grade 12 can be solved with the same three steps. Step one: identify the sample space, S, and count how many outcomes it contains. Step two: identify the event the question is asking about and count how many outcomes belong to that event. Step three: divide the favourable count by the total count, and simplify your fraction.

That is it. Most of the marks in a basic probability question are won or lost in steps one and two, because those are the steps where learners miscount. Step three is just arithmetic. Spend your mental energy on identifying S and the event correctly, and the rest follows naturally.

Identifying the sample space correctly

The sample space is everything that could possibly happen. For a single card drawn from a standard pack, that is 52 outcomes. For a single roll of a die, that is 6 outcomes. For two coins tossed together, that is 4 outcomes (HH, HT, TH, TT — note that HT and TH are different). For a letter chosen at random from the word "PROBABILITY", that is 11 outcomes, because the word has 11 letters even though some of them repeat.

Two tips here. First, if the question involves a deck of cards, remember that a standard deck has 52 cards in 4 suits (hearts, diamonds, clubs, spades), with 13 cards per suit. Of those 13 cards, 10 are number cards (Ace through 10) and 3 are face cards (Jack, Queen, King). Some schools count the Ace as a face card, which would give 9 number cards (2 through 10) and 4 face cards (Ace, Jack, Queen, King). Check which convention your teacher uses and stick with it throughout the question. Second, when letters repeat in a word, you still count every physical position as a separate outcome, because you are choosing a position, not a unique letter.

Counting the favourable outcomes

Once the sample space is identified, count the favourable outcomes. This is where reading carefully matters: "a face card" is different from "a card that is a King or a Heart", and "an even number on a die" is different from "a prime number on a die". Translate the words of the question into an explicit list or count before you write any fractions.

For cards, the four suits each have the same structure, so multiplying usually works: 4 suits × 3 face cards = 12 face cards, 4 suits × 1 King = 4 Kings, and so on. For dice, list the outcomes explicitly if you are not sure: the even numbers on a die are {2, 4, 6}, so 3 outcomes. For words, count by hand and check by re-reading the word letter by letter.

A worked example with cards

Let us put the method into action with a standard card question you will see again and again. A card is drawn at random from a standard pack of 52 cards. Find the probability that the card is a face card, where a face card is a Jack, Queen, or King.

If the question had asked for "a King or a Heart", you would have to be more careful, because the King of Hearts is in both groups and would be counted twice if you simply added 4 + 13. We will deal with that "or" rule in a later post — for now, notice that the straightforward three-step method works when the favourable outcomes form a single, well-defined group.

What to do when the question is in words

Word problems are the same maths wrapped in extra language. The trick is to underline the total number of objects (the sample space size) and the specific group being asked about (the favourable count) before you write anything. Many learners lose marks not because they cannot do the maths, but because they did not notice that the bag has 20 marbles rather than 10, or that the question is asking about "blue" marbles rather than "green" ones.

Once the three-step method is automatic, you will find that most basic probability questions take less than a minute. Save your thinking time for the harder questions later in the paper, where the language becomes trickier and the events start to combine.

One of the quickest ways to lose marks in probability is to treat dependent events as if they were independent, or vice versa. The two ideas look similar on the surface but the maths is different. This post will give you a side-by-side comparison you can keep on your desk whenever you practise, plus a simple test for telling which one a question is using.

What 'independent' actually means

Two events are independent when the outcome of the first one has no effect on the probability of the second one. The classic example is tossing a coin twice: the first toss does not change the coin, so the probability of Heads on the second toss is still 0.5 no matter what happened on the first. Another example is rolling a die twice, drawing a card with replacement, or asking two different learners in a class whether they take Mathematics.

The mathematical signature of independence is the multiplication rule: if A and B are independent, then P(A and B) = P(A) × P(B). You simply multiply the two probabilities, with no adjustment to either one. This is the cleanest calculation in the chapter, and it is also the easiest one to misuse, because not every pair of events is independent.

What 'dependent' actually means

Two events are dependent when the outcome of the first one changes the probability of the second one. The classic example is drawing two balls from a bag without replacement: after the first ball is taken out, the bag is smaller, so the probability of the second ball being a particular colour is different from what it was before the first draw. Other examples include drawing cards without replacement, choosing learners from a class for a team, and selecting people from a small group.

The mathematical signature of dependence is the conditional multiplication rule: P(A and B) = P(A) × P(B|A), where P(B|A) means "the probability of B given that A has already happened". The vertical bar is read as "given". In practice, P(B|A) is just the new fraction after the first event has changed the situation — for the bag example, if you have 4 red and 6 blue and the first one drawn is red, then P(second red | first red) = 3/9, not 4/10.

The contrast in one place

Here is the side-by-side comparison you can keep in your notes.

How to tell which one the question is using

The quickest way is to look for the words "with replacement" or "without replacement". If the question says "with replacement" or describes the object going back in (the card is put back, the ball is returned, the die is rolled again), the events are independent. If it says "without replacement" or describes the object staying out, the events are dependent.

When the question does not use either phrase, fall back on this test: does the second event take place in the same conditions as the first one? If a die is rolled again, nothing has changed — independent. If a card is drawn from a deck that no longer has the first card in it, the deck has changed — dependent. The same test works for choosing people from a small group: once one person has been chosen, the group is smaller, so the events are dependent.

Why this distinction matters in exams

Examiners love this distinction because it is invisible to a learner who is on autopilot. A single word — "with" or "without" — can change the answer completely, and the working looks almost identical either way. That is why so many learners lose marks here: they do the calculation correctly for the wrong type of event. Once you have made the side-by-side comparison above your working habit, this kind of mistake disappears.

Practise telling the two apart before you do any calculations. Read the question, decide "independent or dependent?", write your decision in the margin, and only then start multiplying fractions. That tiny extra step takes five seconds and saves the kind of mark that is otherwise very hard to win back later in the paper.

Mutually exclusive events are the easier cousin of independent events. Once you know what the phrase means, the rule for them is the simplest probability rule in the chapter. Let us break the definition into pieces, see why the addition rule works so neatly for them, and finish with a die example you will recognise.

The definition, plainly

Two events are mutually exclusive when they cannot both happen at the same time. The classic example is rolling a die once and asking, "Is the number rolled even or odd?" A number cannot be both even and odd at the same time, so "even" and "odd" are mutually exclusive. Another example is tossing a coin once: "Heads" and "Tails" are mutually exclusive, because the same toss cannot be both.

The phrase is sometimes written as "disjoint events" in textbooks, but the CAPS ATP and South African exam papers use "mutually exclusive", so stick with that wording. The key idea is that the two events have no outcomes in common — their intersection is empty. If you drew them on a Venn diagram, the two circles would not overlap.

Why mutually exclusive events add so neatly

When two events A and B are mutually exclusive, the probability of "A or B" is simply the sum of the two individual probabilities: P(A or B) = P(A) + P(B). This works because there is no overlap to subtract. If you rolled a die, the probability of "even or odd" is P(even) + P(odd) = 3/6 + 3/6 = 6/6 = 1, which makes sense because every roll is one or the other.

Contrast this with the general addition rule, which says P(A or B) = P(A) + P(B) − P(A and B). The minus term subtracts the overlap, so we do not double-count it. For mutually exclusive events, P(A and B) = 0, so the general rule collapses to the simple rule above. That is the whole story: mutual exclusivity is just the special case where the overlap is zero.

A worked example with a die

Let us make this concrete. A fair six-sided die is rolled once. Let A be the event "rolling a 1" and B be the event "rolling a 6". Find P(A or B).

If A and B had not been mutually exclusive, we would have had to subtract the overlap. For example, if A = "rolling an even number" and B = "rolling a number greater than 3", both events include 4 and 6. So P(A or B) would be P(A) + P(B) − P(A and B) = 3/6 + 3/6 − 2/6 = 4/6 = 2/3. The subtraction step is the difference between the two situations.

Mutually exclusive vs independent

These two ideas are constantly confused, and the confusion costs marks. Mutually exclusive means the two events cannot both happen — they fight each other. Independent means the two events do not affect each other — they ignore each other. The two ideas are almost opposites, in fact, because if two events are truly independent and both have positive probability, they cannot be mutually exclusive (the first happening does not prevent the second, so the second can still happen too).

A useful memory hook: mutually exclusive events have an empty overlap (the circles do not touch). Independent events have an overlap that is exactly the size you would expect from multiplying their probabilities (the overlap is the right size). One is about "can they happen together"; the other is about "does one change the other".

How to spot them in exam questions

The clearest tell-tale phrase is "or": when a question asks for "A or B", the next question you should ask yourself is whether A and B can happen together. If they cannot, use the simple addition rule. If they can, use the general addition rule with the subtraction. The wording "either ... or ..." also signals this, as does "but not both", which makes the mutual exclusivity explicit.

A second tell-tale phrase is "drawing one card" or "rolling one die" or "tossing one coin" — when the experiment is a single action, many natural pairs of events are mutually exclusive. Once the experiment becomes "drawing two cards" or "rolling twice", the events are usually not mutually exclusive, because both can happen across the two draws. Read the question carefully before deciding which rule applies.

Here is a question that catches out learners every year. Two events are described, and you have to decide whether they can both happen at the same time, and if so, what the probability of "both" actually is. Try it on your own first, then we will walk through it together.

The question to try first

A fair six-sided die is rolled once. Let A be the event "the number rolled is even" and B be the event "the number rolled is greater than 3". Your tasks are below. Take two minutes before scrolling down to the solution.

How to think about it before calculating

First, write out the two events as sets. A is the set of even numbers on a die, which is {2, 4, 6}. B is the set of numbers greater than 3 on a die, which is {4, 5, 6}. Once you have the sets in front of you, the question becomes a simple matter of looking at the overlap.

The overlap of two sets is the set of outcomes that belong to both. If the overlap is empty, the two events cannot both happen on the same roll, so they are mutually exclusive. If the overlap has at least one outcome, the two events can both happen, and the probability of "both" is the probability of the overlap.

Reveal: the worked solution

Let us write out the sets and look at the overlap.

So the answer to part (a) is: yes, both A and B can happen at the same time, because 4 and 6 are in both sets. The events are not mutually exclusive.

The probability that both events happen on a single roll is 1/3. Notice that we did not multiply anything here, because we are not talking about two draws — we are talking about one draw where two descriptions must both be true. The favourable outcomes are simply the outcomes that satisfy both descriptions at once.

What this teaches you about 'and'

This question is a small example of a big idea: the word "and" has two different meanings in probability, and you must not mix them up. When the question is about two separate draws (or two separate experiments), "A and B" usually means "A happens on the first, and B happens on the second", and you multiply. When the question is about one experiment where two descriptions must both be true, "A and B" means "the outcome is in both sets at once", and you look at the overlap.

Many learners reach for the multiplication rule whenever they see "and", and they lose marks in single-experiment questions like this one. The fix is to ask yourself, before you multiply, "Is this two separate experiments or one experiment with two conditions?" If it is one experiment, find the overlap. If it is two experiments, multiply.

What to take into the next post

Carry forward two skills. First, write events as sets whenever you can — it makes the overlap visible and the calculation almost automatic. Second, when the question asks "can both happen?", check whether the two sets share any outcomes. If yes, they are not mutually exclusive; if no, they are. This single habit will help you across the next several posts, where we will put all these rules together.

There are really only four probability rules you need for most Grade 12 questions: the addition rule, the multiplication rule, the complement rule, and the rule that probabilities sum to 1. Once you know which rule fits which phrase, the chapter becomes a matching exercise. Let us see all four in one place, then combine them in a worked example.

The four rules you actually need

Here is the shortlist. The addition rule handles "or" — when you want the probability that one event or another happens. The multiplication rule handles "and" — when you want the probability that one event and another both happen. The complement rule handles "not" — when you want the probability that an event does not happen. The sum rule tells you the probabilities of all possible outcomes add up to 1, which is more of a check than a calculation rule.

Every Grade 12 probability question uses at least one of these. The harder ones combine two or three. The skill is not in memorising the rules — they are short — but in spotting which rule to use from the wording of the question.

The addition rule for 'or'

The general addition rule is P(A or B) = P(A) + P(B) − P(A and B). The minus term subtracts the overlap so we do not double-count outcomes that belong to both events. If A and B are mutually exclusive, P(A and B) = 0, so the rule collapses to the simpler P(A or B) = P(A) + P(B). Always ask yourself first whether the two events can both happen — that single check decides whether you subtract or not.

Tip: write down "P(A and B) = ..." explicitly, even if it is zero. Many learners lose marks by silently skipping the subtraction step. If you write P(A and B) = 0, the marker can see you considered the overlap and correctly concluded it was empty.

The multiplication rule for 'and'

The general multiplication rule is P(A and B) = P(A) × P(B|A), where P(B|A) is the probability of B given that A has happened. If A and B are independent, P(B|A) = P(B), so the rule collapses to P(A and B) = P(A) × P(B). For two separate experiments (two coin tosses, two die rolls, two draws with replacement), the events are independent. For two draws without replacement, the events are dependent, and the second probability changes.

Always label which fraction is the "first" and which is the "second" when you multiply. The labels protect you from the most common mistake in the chapter, which is using the same fraction twice when the question says "without replacement".

The complement rule for 'not'

The complement rule is P(not A) = 1 − P(A). It is the single most useful rule for tricky questions, because "at least one", "at least", "not all", and many other awkward phrasings can be turned into a single "not" statement and solved in one line. We will see this in detail in later posts about "at least" and "at most".

The complement rule is also the cheapest way to check your answer in any probability question: if you have just calculated P(A) = 3/10, quickly confirm that P(not A) = 7/10 makes sense in the context of the question. If it does, you have probably not made a slip; if it does not, you have caught yourself before the marker does.

A worked example combining the rules

In a class of 30 learners, 18 take Mathematics, 15 take Physical Sciences, and 10 take both subjects. A learner is chosen at random. Find the probability that the learner takes Mathematics or Physical Sciences.

This question used the general addition rule. It also tells you something useful by the complement rule: 7 learners out of 30 take neither Mathematics nor Physical Sciences, because P(neither) = 1 − 23/30 = 7/30, and 7/30 of 30 is 7. The four rules work together more often than they work alone, and the more you practise, the more naturally you will see the connections between them.

Probability is one of those chapters where the maths is small but the language is large. Learners who can do the calculations perfectly at home still lose marks in the exam, because a single word can change the entire question. This post is for both learners and parents — to help learners know what to look out for, and to help parents understand why this chapter deserves a little extra patience.

They look simple but read carefully

Most probability questions are short, which is part of the problem. A long question gives a learner time to settle into the wording; a short question is over before they have finished reading it. A question like "A die is rolled. What is the probability of an even number?" can be answered in seconds, but if "even" is swapped for "even or a 3", the calculation changes completely. The brevity hides the difficulty.

The trick is to slow down on the short questions. Read every probability question twice before you write anything. The first read is to identify what is being asked; the second read is to notice the small words that change the meaning. Most of the marks lost in this chapter are lost to small words, not to big ideas.

They test reading, not only maths

Probability is the most reading-heavy chapter in the Grade 12 syllabus. The maths itself — counting, adding, multiplying, simplifying — is straightforward, but the question is wrapped in language that has to be decoded. Phrases like "at least one", "at most two", "given that", "without replacement", and "given the first was red" each carry a specific mathematical instruction, and missing any of them turns the whole question in the wrong direction.

Parents who watch their child practise this chapter sometimes wonder why a child who "knows the maths" keeps getting questions wrong. The answer is almost always that the child is reading the question too quickly and missing the small words. Slowing the reading down by even ten seconds per question is usually enough to recover the lost marks.

They use words that hide the maths

Some words in probability have a technical meaning that does not match their everyday use. "And" in everyday speech can mean "and then" (a sequence) or "both at once" (a single event). In probability, "and" can mean either of these things too, and the rule is different for each one. Similarly, "or" in everyday speech often means "one or the other but not both", but in probability, "or" usually means "one or the other or both". These mismatches catch learners all the time.

The way through is to translate the words into maths before calculating. When you see "and", ask: is this two experiments or one? When you see "or", ask: are these mutually exclusive or not? When you see "given", ask: what has changed about the situation? Once the translation is done, the calculation is usually simple.

They change meaning with one word

The single biggest source of lost marks is one-word switches. "With replacement" versus "without replacement". "At least one" versus "exactly one". "Even number" versus "even number or a 3". "First is red" versus "given the first is red". Each of these is one word, and each of them changes the calculation. Examiners love these switches because they sort the learners who read carefully from the learners who do not.

For learners, the fix is mechanical: underline the key word in every probability question before you start. For parents, the fix is patience — your child is not being careless, they are up against a chapter that punishes speed. A calm, slow, second read is worth more in this chapter than another hour of practice.

Tip for parents

Why this chapter rewards patience

Probability is one of the most coachable chapters in Grade 12. The rules are short, the calculations are simple, and the only real skill is reading carefully. Once a learner treats this chapter as a reading exercise with a small amount of arithmetic at the end, the marks come quickly. Encourage your child to think of probability as "reading comprehension with fractions" — that small reframing is often the difference between a frustrating topic and a confident one.

If you do one thing differently in your next probability test, make it this: underline the key words in every question before you write anything else. It sounds too simple to matter, but it is the single habit that separates learners who score in the 80s from learners who score in the 60s in this chapter. Let us break down exactly what to underline and why.

Why underlining works

Probability questions are short and dense. A single word like "without" or "given" can flip the entire calculation, and the rest of the question gives no warning that the word matters. Underlining forces you to slow down on the first read, which is exactly when most learners lose their marks. The visual cue on the page also helps when you check your work later — you can see at a glance which phrases you have already accounted for.

This is not a "study skill" gimmick. It is a method used by learners who consistently score in the top band of the probability paper. The reason it works is that it makes the small words impossible to skip past. Once they are underlined, they sit on the page in front of you, and you have to deal with them.

What to underline, in four categories

There are four categories of words worth underlining in every probability question. The verbs tell you what to calculate. The connectors tell you which rule to use. The conditions tell you which probabilities change. The totals tell you what your sample space is. Get into the habit of underlining one word from each category before you start writing.

Underline the verbs first

Verbs are the words that tell you what to find. "Find", "calculate", "determine", "show that", "write down". The verb "write down" means no working is needed; "show that" means you must produce a full solution that arrives at the given value; "find" or "calculate" is the normal working. Underlining the verb stops you from giving too much or too little working for the marks on offer.

Also underline any word that tells you what the event is — "even", "red", "King", "at least one", "both", "neither". These are not verbs, but they are the description of the event you are calculating about, and missing one of them is the most common reason learners solve the wrong question.

Underline the connectors

The connectors are "and", "or", "given", "if". Each one points to a rule. "And" points to the multiplication rule (or to finding the overlap, in a single experiment). "Or" points to the addition rule. "Given" or "if" points to a conditional probability, where the situation has changed. Underline these words every time, and write the rule next to them in the margin: "and = multiply", "or = add", "given = conditional". The rule will be in front of you when you start calculating.

If you find yourself writing a calculation without knowing which rule you are using, stop. The connector should already be underlined and labelled. If it is not, go back and underline it before you write any fractions.

Underline the conditions and totals

Conditions change the calculation: "without replacement", "with replacement", "the first was red", "the die is biased". Each of these is a condition that alters the probabilities in the question. Underline them, and next to each, write what it changes. "Without replacement" means denominators decrease. "First was red" means the second draw has one fewer red and one fewer ball.

Totals are the numbers that decide your sample space: "a bag contains 4 red and 6 blue", "a class of 30 learners", "a standard pack of 52 cards". Underline the total number of objects in the bag, class, deck, or sample. If you misread the total, every fraction you write afterwards will be wrong, no matter how clean your method is.

Tip: a one-line checklist before you start

What this habit does to your marks

Learners who adopt this habit report two changes in their next probability test. First, they finish the section faster, because they do not waste time redoing questions they misread. Second, they pick up marks on the trickier questions at the end of the section, because the small words are already in front of them. The habit costs nothing and gains real marks, which is the best kind of exam technique there is.

Practise it on every probability question you do from now on, even the easy ones. The easy ones are where you build the habit so it is automatic by the time the harder ones arrive.

"At least" is one of those tiny phrases that scares learners, but the meaning is straightforward once you slow down. In this post we will pin down exactly what "at least" means, see why the complement rule is usually the fastest way to handle it, and contrast it with "at most" so you never confuse the two in an exam again.

The plain meaning of 'at least'

"At least n" means "n or more". So "at least 3" means "3 or more", which is 3, 4, 5, 6, and so on up to the maximum. In probability, the phrase always refers to a count: "at least one Heads in three coin tosses", "at least two red balls drawn", "at least one learner who takes Mathematics". The phrase tells you to include the number itself and everything above it.

The trick is that "at least one" sounds like a small, single case, but mathematically it includes every possibility except the case of zero. So "at least one Heads in three tosses" includes one Heads, two Heads, and three Heads — three separate probabilities to add up. Adding them all is one way to solve the question, but there is a much faster way.

Why it usually means use the complement

Here is the shortcut. The complement of "at least one" is "none", and the probability of "none" is usually a single calculation. So P(at least one) = 1 − P(none). For the coin example: P(no Heads in three tosses) = (1/2) × (1/2) × (1/2) = 1/8, so P(at least one Heads) = 1 − 1/8 = 7/8. Notice that we did not have to calculate the probabilities of one, two, or three Heads separately — the complement rule gave us the answer in one line.

This is why "at least" almost always points to the complement rule. The case of "none" is usually one clean multiplication, while the cases of "one or more" require you to add several probabilities together. Going via the complement turns a long calculation into a short one. Whenever you see "at least one" in a question, write "1 − P(none)" in the margin before you do anything else.

'At least' vs 'at most'

The two phrases are mirror images, and learners mix them up constantly. Here is the side-by-side comparison you can keep in your notes.

A worked comparison

Three coins are tossed. Find the probability of (a) at least one Heads, and (b) at most one Heads.

Part (a) uses the complement. P(no Heads in three tosses) = (1/2) × (1/2) × (1/2) = 1/8. So P(at least one Heads) = 1 − 1/8 = 7/8. The complement route took one short multiplication.

Part (b) requires direct addition. P(0 Heads) = 1/8 (just TTT). P(1 Heads) = 3/8 (HTT, THT, TTH — three arrangements). So P(at most one Heads) = P(0) + P(1) = 1/8 + 3/8 = 4/8 = 1/2. There is no shortcut here, because the complement "two or more Heads" is a bigger sum, not a smaller one.

Notice how the two routes differ: "at least" pulled us toward the complement, "at most" pushed us toward direct addition. That pattern is the whole reason the distinction matters — pick the wrong route and you end up with more calculation, and more chances to slip, than you needed.

How to remember which one to use

Here is a simple test. Ask yourself: is the case of zero in the set I am calculating, or in its complement? If "zero" is in the complement (as with "at least one"), the complement is small and you should use 1 − P(complement). If "zero" is in the set I am calculating (as with "at most one"), the set itself is small and you should add the small cases directly. This single test will tell you in five seconds whether to use the complement rule or direct addition.

One last caution: "at least one" is the easiest case, but the same logic applies to "at least two", "at least three", and so on. The complement of "at least two" is "zero or one", which is two cases to add instead of many. Always check which set has fewer cases, and calculate that one.

If "at least" is the phrase that scares learners, "at most" is the one that quietly catches them out, because it does not have the same tidy complement shortcut. In this post we will pin down what "at most" means, see why it usually means adding the small cases directly, and contrast it carefully with "at least" so the two never get muddled in your exam again.

The plain meaning of 'at most'

"At most n" means "n or fewer". So "at most 3" means "3 or fewer", which is 0, 1, 2, or 3. In probability, the phrase always refers to a count: "at most two Heads in three tosses", "at most one red ball drawn", "at most three learners who take Mathematics". The phrase tells you to include the number itself and everything below it, down to zero.

The trap with "at most" is that it looks similar to "at least", and learners sometimes apply the same complement shortcut. The shortcut does work in principle — the complement of "at most 2" is "3 or more" — but in practice the complement is usually a larger set of cases, not a smaller one, so the shortcut does not save you time. For "at most", the route of least work is almost always direct addition of the small cases.

Why it can mean a sum of small probabilities

Here is the rule. For "at most n", calculate the probabilities of the cases 0, 1, 2, ..., n and add them. The number of cases is n + 1, which is usually small. For example, "at most 2 Heads in three tosses" means P(0) + P(1) + P(2) = 1/8 + 3/8 + 3/8 = 7/8. Three small additions, and you are done.

Compare this with "at least 2 Heads in three tosses", which is P(2) + P(3) = 3/8 + 1/8 = 4/8 = 1/2. In this case the "at least" calculation has fewer cases, so it is the faster route. The lesson is: whichever phrase you meet, count the cases first, then pick the route that has the smaller list.

'At most' vs 'at least'

Here is the side-by-side comparison you can keep on your desk.

A worked comparison

Three coins are tossed. Find the probability of (a) at most one Heads, and (b) at least one Heads.

Part (a) uses direct addition. P(0 Heads) = 1/8 (just TTT). P(1 Heads) = 3/8 (HTT, THT, TTH — three arrangements). So P(at most one Heads) = P(0) + P(1) = 1/8 + 3/8 = 4/8 = 1/2. The direct route took two small additions.

Part (b) uses the complement. P(no Heads in three tosses) = 1/8. So P(at least one Heads) = 1 − 1/8 = 7/8. The complement route took one short multiplication. Notice how different the routes are: "at most" pulled us toward direct addition, "at least" pushed us toward the complement. Mixing them up gives you more work and a higher chance of a slip.

Spotting it in exam questions

"At most" tends to appear in questions about counting events — coin tosses, die rolls, balls drawn from a bag, learners chosen from a class. Whenever you see the phrase, your first step is to list the cases that count, then count them. If the list is short (three or fewer cases), add them directly. If the list is long, look at the complement and see if it is shorter.

A common exam variation is "at most one", which is just "zero or one". This is a two-case addition, the simplest version of the phrase. Get comfortable with this version first, because it appears in nearly every Grade 12 probability paper. Once you can do "at most one" in your sleep, "at most two" and "at most three" are just longer versions of the same addition.

One final caution: read "at most" carefully. It includes the number itself. "At most 2" includes 2, not "fewer than 2". This single off-by-one slip costs learners a mark every year. When in doubt, write the list of cases out explicitly: 0, 1, 2 — and confirm 2 is in the list before you add.

"And" and "or" are the two words that drive every probability calculation. Get them straight, and the chapter falls into place. Confuse them, and the rest of the chapter feels like guesswork. In this post we will pin down what each word tells you to do, see when the simple rules need adjusting, and finish with a worked example that uses both.

'And' means multiply

The word "and" almost always points to multiplication. When you see "A and B", you are being asked for the probability that both A and B happen. If the two events are about two separate experiments (two draws, two rolls, two tosses), and the events are independent, then P(A and B) = P(A) × P(B). If the events are dependent (one draw affects the next, like drawing without replacement), then P(A and B) = P(A) × P(B|A), where P(B|A) is the probability of B given that A has happened.

There is one exception to keep in mind: if A and B are about a single experiment — for example, "the card is a King and a Heart" from one draw — then "and" means the overlap, not multiplication. In that example, only the King of Hearts satisfies both, so P(King and Heart) = 1/52, which is a count of the overlap, not a product of two probabilities. The rule is: two experiments means multiply; one experiment means find the overlap.

'Or' means add

The word "or" almost always points to addition. When you see "A or B", you are being asked for the probability that A happens, or B happens, or both. The general addition rule is P(A or B) = P(A) + P(B) − P(A and B). The minus term subtracts the overlap so you do not double-count outcomes that satisfy both A and B. If the two events are mutually exclusive (they cannot both happen), then P(A and B) = 0, and the rule simplifies to P(A or B) = P(A) + P(B).

One note about everyday meaning: in normal speech, "or" often means "one or the other but not both". In probability, "or" is inclusive — it means "one or the other or both". The subtraction term in the addition rule is there precisely because "or" is inclusive. If you ever need the exclusive version ("exactly one of A or B but not both"), you calculate P(A or B) − P(A and B), which removes the "both" case. This comes up rarely in Grade 12, but it is worth knowing the option exists.

When the simple rules need adjusting

The two simple rules — multiply for "and", add for "or" — work in the easy cases, but each has a small adjustment you must remember. For "and", the adjustment is to use the conditional probability when the events are dependent. For "or", the adjustment is to subtract the overlap when the events can both happen. Failing to make these adjustments is the source of most lost marks in the chapter.

The safest habit is to write the general version of the rule every time, then simplify when you are sure the special case applies. Write P(A and B) = P(A) × P(B|A) and only replace P(B|A) with P(B) once you have confirmed the events are independent. Write P(A or B) = P(A) + P(B) − P(A and B) and only drop the subtraction once you have confirmed the events are mutually exclusive. Two extra seconds of writing saves a full mark of carelessness.

A worked example with both

A bag contains 4 red and 5 blue balls. Two balls are drawn, one after the other, without replacement. Find the probability that both balls are the same colour (both red OR both blue). Notice the "or" at the end, joining two "and" calculations.

Notice how the question combined both words. Each "both" calculation used the multiplication rule with the "without replacement" adjustment (the second denominator was 8, not 9). The two results were then added, because the two events "both red" and "both blue" are mutually exclusive — you cannot draw two balls that are both entirely red and both entirely blue at the same time. So no subtraction was needed in step 3, and we simply added.

How to read questions correctly

When you meet a probability question with both words in it, your first job is to spot them. Underline every "and" and every "or" before you write anything. Then, in the margin, label what each one tells you: "and = multiply", "or = add". The labelling takes five seconds and protects you from the common slip of mixing up the two rules under pressure.

Once the words are labelled, decide whether any adjustments apply. For "and", ask whether the events are dependent; if yes, use the conditional version. For "or", ask whether the events can both happen; if yes, subtract the overlap. The whole chapter becomes a sequence of small decisions: identify the word, identify the rule, check for the adjustment, write the calculation. Practise this rhythm and probability becomes one of the most predictable topics in the Grade 12 paper.

Probability trees look like a lot of work, but they are actually one of the most organised ways to keep track of "without replacement" questions. Here is a tree question for you to try first, with the solution revealed below. Take five minutes to draw the tree and calculate before you scroll down.

The question to try first

A bag contains 3 red balls and 5 blue balls. Two balls are drawn, one after the other, without replacement. Draw a probability tree for the two draws and use it to find the probability that the two balls drawn are different colours (one red and one blue, in any order).

How to set up the branches

Start by listing what the first draw can produce. From a bag of 8 balls (3 red and 5 blue), the first draw is either red (probability 3/8) or blue (probability 5/8). These are the two branches coming out of the starting point. Label each branch with its probability.

From each of those branches, draw two more branches for the second draw. If the first ball was red, the bag now has 2 red and 5 blue (7 balls total), so the second-draw probabilities are 2/7 for red and 5/7 for blue. If the first ball was blue, the bag now has 3 red and 4 blue (7 balls total), so the second-draw probabilities are 3/7 for red and 4/7 for blue. Label every branch with the updated fraction — the denominators all drop to 7, and the numerators depend on what was taken out first.

Reveal: the worked answer

There are two routes through the tree that give "one red and one blue": red first then blue, or blue first then red. Let us calculate each route by multiplying along its branches.

Notice two things. First, the two routes happen to have the same probability (15/56), which is not always the case but happens here because the numbers are symmetric. Second, the two routes are mutually exclusive — they cannot both happen on the same two draws — so we add them, not multiply. The most common mistake on tree questions is to multiply the two route probabilities together instead of adding them. Adding is correct here, because the routes are alternative ways to get the result, not additional conditions.

What this teaches you about trees

Probability trees are a way to make "without replacement" questions visible. Without a tree, you have to remember that the second draw changes; with a tree, the change is written on the page in front of you. The updated denominator (7 instead of 8) is on every second-level branch, so you cannot accidentally use 8 again. This is the main value of drawing the tree — it externalises the calculation so your memory does not have to hold it.

Trees also make the "and/or" distinction obvious. Multiplication happens along a single route from start to finish (a sequence of "and"s). Addition happens across different routes that lead to the same outcome (a sequence of "or"s). Once you see the tree that way, the rules of probability become a visual pattern, not a list of formulas to memorise.

What to take into the next post

Carry two habits forward. First, always draw the tree for any "without replacement" question with two draws — the visual check protects you from the most common mistake in the chapter. Second, multiply along a route, add across routes. If you find yourself adding along a single route or multiplying across routes, stop and re-read the question. Those two habits cover most of what trees are for, and they will be useful in the final practice post that follows.

If you have only ten minutes to revise probability today, spend them on this single question. It brings together three of the skills this chapter has been building: the multiplication rule for "without replacement", the complement rule for "at least one", and the simplification of fractions. Try it on your own first, then walk through the solution.

The question

Read the question carefully, then take a sheet of paper and try both parts before scrolling down. Each part uses a different rule, so they will tell you which skills you have mastered and which still need a little work.

Solution: part (a)

Part (a) uses the multiplication rule for "without replacement". The first draw has 6 red out of 10 marbles. After one red has been drawn, the bag has 5 red out of 9 marbles. Multiply the two fractions and simplify.

Notice the labels. We wrote "first red" and "second red | first red" so the marker can see exactly which fraction is which. That labelling is the cheapest way to protect yourself from the "without replacement" mistake we discussed earlier in the chapter.

Solution: part (b)

Part (b) uses the complement rule for "at least one". The complement of "at least one blue" is "no blue", which means "both red". We have already calculated P(both red) in part (a), so we can use it directly here.

This is the power of the complement rule: a question that looks like it has many cases (one blue + two blues, plus the order matters...) collapses to a single subtraction. The reason it works so neatly here is that "at least one" includes every case except the one case of "none", and "none" was already calculated in part (a). Always look for these connections between parts of a question — they are not coincidences, they are the question writer pointing you to the fastest route.

What to take from this question

Three things should stick. First, label every fraction with the draw it represents, especially in "without replacement" questions. Second, when you see "at least one", reach for the complement rule before you start adding cases; it is almost always faster. Third, look at the connection between parts of a question. Part (b) used the answer from part (a); if you had not noticed that, you would have spent extra time calculating "no blue" from scratch.

This question is a good self-test. If you got both parts right on your first attempt and used the rules above without prompting, your probability is in good shape for the exam. If you slipped on either part, go back to the relevant post in this chapter and try two more practice questions on the same skill. Most learners find that two extra focused questions are enough to convert a near-miss into a confident mark.