
Probability and Outcomes
Probability in MYP2 is really one idea wearing six different costumes: dice, coins, spinners, marbles, trees, Venn diagrams. Every question reduces to counting — how many outcomes are possible, and how many count as a 'win'. Get the total number of outcomes wrong (writing 6 instead of 36 for two dice, or adding instead of multiplying for combined events) and the rest of the answer collapses even with perfect method afterwards.
Overview — The Shape of This Topic
What this chapter is really testing
- Probability questions always come back to counting: how many total outcomes are there, and how many of them count as favourable.
- You'll meet in six costumes — dice, coins, spinners, marbles — but the structure never changes.
- The jump from Year 1 is combining events: a die AND a coin, two dice together, a spinner AND a marble draw. Combined totals multiply, they never add.
Examiners like writing a wrong student claim straight into the question (see the sum-of-7 dice problem) and asking you to judge it. That means you need to defend your counting in words, not just produce a fraction.
The shape of the chapter
Command terms this topic uses constantly
| Command term | What it demands | AO | Mark-earning move |
|---|---|---|---|
| State | Give a short answer with no working required. | AO1 | 1 mark for the correct value/fraction alone — justification not needed, but the value must be correctly written (e.g. as a fraction, not left as a raw ratio like 1:6). |
| Identify | Pick out or name a specific feature from given information. | AO1 | 1 mark per correct item named — for combined events this usually means naming both individual sample spaces before anything is calculated. |
| Calculate | Show the method and reach a numerical answer. | AO2 | Marks for correct substitution into AND the final simplified value — an unsimplified fraction can lose the last mark. |
| Explain | Give reasons connected to the maths, not a restated observation. | AO3 | No credit for 'because it's random' alone — must reference sample size, independence, or the specific rule being tested. |
| Compare | State similarities/differences between two given quantities, with values. | AO3 | Needs BOTH a numerical statement (which is bigger/smaller) and a reason — one without the other usually caps the mark. |
Key point
Theoretical vs Experimental Probability
Two different questions about the same event
Theoretical probability is what should happen if the world matched the maths exactly — worked out before you touch a die or a bag. Experimental probability is what actually happened when trials were run and results counted. They usually don't match exactly, and that's not an error — it's just how randomness behaves in small samples.
- Theoretical probability needs no data collection: , based purely on counting what's possible.
- Experimental probability comes from real trials: (number of times the event occurred) ÷ (number of trials) — it changes if you repeat the experiment.
- As the number of trials grows very large, experimental probability tends to settle close to the theoretical value — this is why insurers and casinos can rely on probability over millions of events even though any single outcome is unpredictable.
- Theoretical probability
- The probability of an event calculated from reasoning about equally likely outcomes, with no trials needed.
- Experimental probability
- The probability estimated from the actual results of repeated trials.
- Expected frequency
- The number of times an event is predicted to occur in a given number of trials.
Theoretical probability
predicted count over n trials
- — number of trials
Worked example
Why 'expected 2 sixes, got 1' doesn't mean the die is broken
Theoretical vs Experimental Probability
| Feature | Theoretical Probability | Experimental Probability |
|---|---|---|
| Needs actual trials? | No — calculated from reasoning alone | Yes — based on real recorded results |
| Formula | (favourable/total outcomes) | (observed successes) ÷ (number of trials) |
| Changes if repeated? | Never — fixed by the structure of the event | Yes — different trial runs give different results |
| Best used when | Structure is known (fair die, equal sectors) | Fairness is unknown, or a claim needs testing |
Common mistake
Assuming an experimental result must exactly equal the theoretical prediction every single time.
Consequence: Leads to wrongly concluding a fair die/coin/spinner is 'unfair' from a tiny sample — loses the explain mark even if the probability calculation itself was correct.
Fix: State explicitly that small samples vary randomly around the theoretical value, and fairness is only meaningfully tested over a large number of trials.
Examiner tip
Where theoretical probability actually gets used
Key takeaways
- Theoretical probability = calculated from structure (); experimental = measured from real trials.
- Expected frequency — a prediction, not a guarantee for any single run.
- Small sample sizes cause natural variation from the expected value — not evidence of unfairness.
- As trials increase, experimental probability converges toward the theoretical value.
Single and Combined Events
One event vs. two (or more) happening together
A single event has one action and one set of outcomes — rolling one die, drawing one marble. A combined event links two or more actions together — a die AND a coin, a marble draw AND a spin. The moment you combine events, the total number of outcomes multiplies rather than adds, because every outcome of the first action pairs with every outcome of the second.
- Single events: , with from that one action only.
- Combined independent events: total outcomes , and .
- 'Independent' means the outcome of one event has zero effect on the other — a die roll can never influence a coin flip, so those two are always independent of each other.
- Independent events
- Two events where the outcome of one has no effect on the probability of the other.
- Combined event
- An event made up of two or more single events considered together, e.g. rolling a die and flipping a coin.
- Mutually exclusive events
- Two events that cannot both happen at the same time, e.g. rolling a 2 and a 5 on the same roll.Different concept from independence — the vocabulary contrast is commonly tested.
multiplication rule for independent combined events
- — two independent events
total outcomes for two combined actions
Worked example
Die and coin together — the classic independent pairing
Frequently confused
| Concept A | Concept B | The distinction that matters |
|---|---|---|
| Independent events | Mutually exclusive events | Independent = one event doesn't affect the other's probability (multiply probabilities); mutually exclusive = the two events can never happen together at all. |
Common mistake
Adding instead of multiplying when finding total combined outcomes (e.g. thinking die+coin gives 8 outcomes, not 12).
Consequence: Every probability calculated afterwards using that denominator is wrong, even if the favourable-outcome count was correct.
Fix: Build or imagine the full grid of combined outcomes first — count rows × columns, don't just add the two totals.
Key takeaways
- Combined events multiply totals: .
- Independent events multiply probabilities: .
- Independent ≠ mutually exclusive — don't mix up the vocabulary in an explain answer.
- Identify each single sample space before combining — examiners award marks for this step separately.
Probability with Dice, Coins and Spinners
Where the total-outcomes trap usually strikes
Dice, coins and spinners are the classic models because their outcomes are equally likely by design — but combining two dice is where most marks are lost, because the natural (wrong) instinct is to count possible SUMS (2 to 12, i.e. 11 outcomes) instead of possible OUTCOME PAIRS (36).
- Two dice rolled together: each die still shows 1–6, but because they're rolled separately, and are two different equally likely outcomes, giving total.
- A sum like 7 can be made multiple ways: — six different outcome-pairs, all giving the same sum.
- Spinners follow the same rules once you know the number of equal sectors — a spinner with 4 equal sections behaves exactly like a 4-sided die for probability purposes.
Sample space grid for two dice — sums of 7 highlighted
36 equally likely outcomes total; exactly 6 of them sum to 7, giving .
Worked example
Judging the student's claim about rolling a 7
total outcomes for two dice rolled together
Examiner tip
Key takeaways
- Two dice together always give 36 equally likely outcomes, never 11 (sums aren't equally likely) and never 6.
- List combined outcomes as ordered pairs — (1,6) and (6,1) are different outcomes.
- Spinners and dice follow identical rules once you know the number of equally likely sections.
Using Tree Diagrams and Venn Diagrams (Introductory)
Two visual tools for combined events
A tree diagram shows combined events as branching paths — multiply the probabilities ALONG a path to get that path's probability, then add DOWN separate paths if combining several different results. A Venn diagram instead shows how two categories of outcomes overlap within a single sample space — useful for 'and'/'or'/'not' questions, like numbers that are even AND greater than 3.
- Tree diagrams suit sequential or combined independent events (coin then coin, spinner then die).
- Venn diagrams suit classifying a single group of outcomes by two overlapping properties.
- Rule of thumb: along a single branch/path — MULTIPLY. Across separate complete paths — ADD.
- Tree diagram
- A branching diagram where each branch is a possible outcome of a stage, and each complete path represents one combined outcome with probability equal to the product of its branch probabilities.
- Venn diagram
- A diagram using overlapping circles inside a rectangle (the sample space) to show how outcomes are grouped and shared between categories.
- Intersection
- The overlapping region of a Venn diagram — outcomes belonging to BOTH categories.
- Union
- All outcomes belonging to EITHER category (or both) — the combined area of both circles.
Tree diagram: flipping a fair coin twice
Multiply along each path to get its probability; the four end probabilities sum to 1.
Venn diagram: even numbers and numbers greater than 3 (die faces 1–6)
Overlap (intersection) contains 4 and 6 — numbers that are both even and greater than 3.
Worked example
Reading probabilities off a tree diagram
Common mistake
Adding probabilities along a single tree branch path instead of multiplying.
Consequence: Produces impossible probabilities in some cases, and always loses the calculation mark even if the branch probabilities were correct.
Fix: Say the rule out loud while working: 'along a path, multiply; between different end-paths, add.'
Key takeaways
- Tree diagrams: multiply along a path, add across separate complete paths.
- Venn diagrams sort a single sample space by two overlapping properties; overlap = intersection (), total covered area = union ().
- All end-path probabilities in a tree diagram must sum to 1 — a useful self-check.
Introduction to Probability and the Probability Scale
Probability as a number between 0 and 1
Every probability, however it's calculated, lands somewhere on a fixed scale from 0 (impossible) to 1 (certain). The scale is identical whether you express the value as a fraction, a decimal, or a percentage — they're just three costumes for the same number.
- means the event genuinely cannot happen (rolling a 7 on a standard die).
- means the event is guaranteed (rolling a number less than 7 on a standard die).
- (or , or 50%) sits at 'evens' — exactly as likely to happen as not.
- Words like 'unlikely' or 'likely' are rough zones, not exact values — always also give the precise fraction/decimal/percentage.
F = number of favourable outcomes, T = total number of possible outcomes
- — favourable outcomes
- — total possible outcomes
every probability lies on this scale, inclusive of both ends
complement rule — a shortcut for 'not' questions
The probability scale, 0 to 1
Fractions, decimals and percentages are the same probability in three formats — check which format the question asks for.
Same probability, three formats
| Fraction | Decimal | Percentage | Likelihood word |
|---|---|---|---|
| 0 | 0 | 0% | Impossible |
| 1/4 | 0.25 | 25% | Unlikely |
| 1/2 | 0.5 | 50% | Evens |
| 3/4 | 0.75 | 75% | Likely |
| 1 | 1 | 100% | Certain |
Mnemonic
I Use Every Last Chance
Impossible, Unlikely, Evens, Likely, Certain — the five likelihood zones on the probability scale from 0 to 1, in order.
Worked example
Comparing probabilities between two different bags
Examiner tip
Key takeaways
- Every probability sits between 0 (impossible) and 1 (certain) — fractions, decimals and percentages are just different formats for the same value.
- ; complement rule: .
- Compare probabilities as proportions (fractions), never as raw favourable counts, when totals differ.
Listing Outcomes and Sample Space Diagrams
Systematic listing beats guessing
A sample space is the complete, organised list of every possible outcome of an experiment. The skill tested here isn't the probability calculation — it's whether you can list or table the outcomes systematically enough to be certain nothing was missed or duplicated.
- For a single action, the sample space is just the list of outcomes: a die's sample space is .
- For combined actions, use the counting principle: action 1 with outcomes and action 2 with outcomes gives combined outcomes.
- A two-way table (rows = one action, columns = the other) is the safest way to list a combined sample space without missing any pair — exactly the grid used for two dice.
- Sample space
- The complete set of all possible outcomes of an experiment.e.g. for a coin
- Outcome
- One single possible result of an experiment, e.g. rolling a 3.
- Event
- A collection of one or more outcomes sharing a property, e.g. 'rolling an even number' = .
counting principle — total combined outcomes when action 1 has m outcomes and action 2 has n outcomes
Sample space table: rolling a die and flipping a coin
12 equally likely outcomes total () — every one appears exactly once, which is why tables are safer than mental listing.
Worked example
Listing outcomes for a 3-colour spinner and a coin
Which listing method to use
| Method | Best for | Risk if used carelessly |
|---|---|---|
| Simple list {..} | Single action only (one die, one spinner) | None — outcomes are few and obvious |
| Two-way table | Two combined actions with small outcome counts (die+coin, two dice) | Guarantees every pair shown exactly once — safest for counting totals |
| Tree diagram | Combined/sequential events, especially with unequal branch probabilities | Slower to draw for large sample spaces, but shows probabilities directly |
Common mistake
Missing an outcome or double counting when listing combined sample spaces without a system (e.g. forgetting (6,T) in a die-coin list).
Consequence: The total used as denominator becomes wrong, so every probability calculated from that incomplete list is wrong, even if the favourable count was fine.
Fix: Use a two-way table or a fixed cycling order instead of listing from memory — check rows × columns against the counting principle .
Key takeaways
- Sample space = complete list of all possible outcomes; event = a subset sharing a property.
- Counting principle: combined total outcomes — use this to check any list or table.
- Two-way tables are the safest listing method for two combined actions; trees suit sequential/unequal-probability situations.
Quick Reference
Probability and Outcomes — everything on one page
Formulas
- theoretical probability
- prediction over n trials
- independent combined events
- counting principle for combined sample spaces
- complement rule
Key facts
- Two dice total outcomes
- 36
- Die + coin total outcomes
- 12
- Probability scale range
- 0 to 1 inclusive
Mnemonics
- I Use Every Last ChanceImpossible, Unlikely, Evens, Likely, Certain — the probability scale from 0 to 1
Don't confuse
- Independent events vs Mutually exclusive events: Independent = multiply probabilities, one doesn't affect the other; mutually exclusive = can't both happen at once.
Command terms
- Calculate:
- Show substitution into and give a simplified final value
- Explain:
- Reference sample size, independence, or the specific rule — not just 'it's random'
- Compare:
- Give a numeric comparison AND a reason, not just two fractions side by side
Exam checklist
- Have I counted the TOTAL outcomes correctly before dividing? (36 for two dice, not 6 or 11)
- Have I simplified my final fraction?
- For combined independent events, did I multiply totals/probabilities, not add them?
- For a tree diagram, did I multiply along each path and check all end-path probabilities sum to 1?
- For a 'compare' or 'explain' question, have I written a full sentence with a reason, not just a number?