AP Statistics · Unit 2 · Probability, Random Variables and Probability Distributions
2.3 Simulation
Describe how to design a simulation, carry out its steps, and interpret the results as an estimate of a probability.
1/5A simulation imitates a random process with a model
Part 1 of 5: A simulation imitates a random process with a model
A basketball player makes 70% of her free throws. You want the chance that she makes at least 3 of her next 4. Watching many games would take far too long.
A simulation imitates a random process using a model and repeats it many times. The model relies on a random device. Examples are a coin, a die, a calculator’s random number function or a table of random digits. The device must match the real chances.
Simulation is useful when the real process is slow, costly or hard to calculate exactly.
Key terms · tap one for its definition
simulation
A simulation is a model that uses random numbers or devices to imitate a random process and repeats it many times.
random device
A random device is a tool, such as a coin, a die or a random number generator, that produces outcomes by chance.
Part 2 of 5: A simulation follows a short sequence of design steps
Before using any random numbers, plan the model. The process is repeated many times. Each complete run is a trial, and each trial ends with a recorded result. The design steps are:
- State the real situation and the chances of its outcomes.
- Assign digits to outcomes so their shares match those chances.
- Define one trial: how many digits to read and what to record.
- Run many trials and record each result.
- Find the proportion of trials in which the event occurred.
Key term
trial
A trial is one complete run of a simulated random process, ending in a recorded result.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: C · Worked explanation
Part 3 of 5: Assigned digits must match the probabilities exactly
The second step is where most errors happen. For a 70% shooter, use the ten digits 0 through 9. Seven of the ten digits should stand for a made shot, because 7 out of 10 is 70%. The other three stand for a miss.
Some digits may be left unassigned and skipped. When choosing people without repeats, skip any repeated digit and keep reading. Each shot is separate, so repeated digits are allowed for shots.
Part 4 of 5: Repeating many trials gives an estimated probability
Now run the model. Each trial uses four digits, one for each shot. After many trials, count how often the event occurs. The proportion of trials with the event is the empirical probability, which estimates the true probability. A short run gives a rough estimate, and a long run gives a steadier one.
Look at the example that follows. The digits were assigned first, each trial used four digits, and the proportion was found at the end. Every step of the plan appears in the working.
Worked example
Estimating a probability from trials
- Word equation: empirical probability = trials with the event divided by total trials.
- Count the makes in each trial: 3, 2, 4, 3, 1, 4, 2, 4, 2, 4.
- Trials with at least 3 makes are trials 1, 3, 4, 6, 8 and 10, which is 6 trials.
- Substitute: .
Answer: The estimated probability of at least 3 makes is 0.60.
Key term
empirical probability
An empirical probability is the proportion of trials in which an event occurs, used as an estimate of its probability.

Quick check
One question on the part you just read
Easy
Show answer and explanation
Answer: A · Worked explanation
Part 5 of 5: A simulation result is an estimate that varies
Ten trials is a small run, so treat 0.60 as a rough guess. Another set of ten trials would probably give a different proportion. Chance variation affects every simulation. More trials make the estimate more stable, so results tend to settle near the true probability.
A simulation never gives the exact probability. It gives an estimate that can be close or far from the true value.
Check your recall[12 marks]
5 written questions. Answer each one in your head or on paper, then open the mark scheme.
- 1.Describe[2 marks]
Describe what a simulation is and why it can be useful.
Show mark schemeHide mark scheme
Model answer
A simulation uses a random device to imitate a random process, repeating it many times to estimate a probability. It is useful when the real process is slow, costly or hard to calculate exactly.
- 1 mark for stating that it imitates a random process with a random device over many trials
- 1 mark for giving a reason, such as the real process being slow, costly or hard to calculate
- 2.Describe[3 marks]
A student guesses on each of 3 multiple-choice questions with 4 equally likely options. Describe how to use a table of random digits to simulate one trial that counts correct guesses.
Show mark schemeHide mark scheme
Model answer
Let digit 1 be a correct guess and digits 2, 3 and 4 be incorrect guesses, ignoring digits 0 and 5 through 9. Read three digits for one trial. Record the number of 1s.
- 1 mark for assigning one digit to a correct guess and three digits to incorrect guesses
- 1 mark for ignoring the unassigned digits
- 1 mark for reading three digits per trial and recording the number of correct guesses
- 3.Calculate[2 marks]
A simulation of 50 trials had the event occur in 14 of them. Calculate the empirical probability.
Show mark schemeHide mark scheme
Model answer
Empirical probability = trials with the event divided by total trials = .
- 1 mark for dividing 14 by 50
- 1 mark for giving the answer 0.28
- 4.Explain[2 marks]
Explain why a simulation result is an estimate rather than an exact probability.
Show mark schemeHide mark scheme
Model answer
The result is based on a limited number of random trials. Another run would give a different proportion because of chance variation.
- 1 mark for stating that it is based on a limited number of random trials
- 1 mark for explaining that chance variation would change the result in another run
- 5.Justify[3 marks]
One class runs 20 trials and finds an empirical probability of 0.30. Another class runs 2,000 trials and finds 0.34. Justify which result is more trustworthy.
Show mark schemeHide mark scheme
Model answer
The 2,000-trial result is more trustworthy. More trials reduce the effect of chance variation, so the estimate tends to be closer to the true probability. A run of only 20 trials can vary a lot.
- 1 mark for choosing the 2,000-trial result
- 1 mark for stating that more trials reduce the effect of chance variation
- 1 mark for noting that a small number of trials can vary widely or give a less reliable estimate

Practice this topic
10 quick-fire multiple-choice questions, easy to hard. Tap an answer to see if you're right and the worked explanation. (The written questions with mark schemes are under “Check your recall” above.)
0 right0 of 10 answered
Includes the 2 quick checks in the note above.
EasyIdentify[1 mark]
Show answer and explanation
Answer: B · Worked explanation
MediumCalculate[2 marks]
Show answer and explanation
Answer: C · Worked explanation
MediumDetermine[2 marks]
Show answer and explanation
Answer: A · Worked explanation
MediumDetermine[2 marks]
Show answer and explanation
Answer: D · Worked explanation
MediumInterpret[2 marks]
Show answer and explanation
Answer: B · Worked explanation
HardDetermine[2 marks]
Show answer and explanation
Answer: A · Worked explanation
HardExplain[2 marks]
Show answer and explanation
Answer: C · Worked explanation
HardDetermine[2 marks]
Show answer and explanation
Answer: B · Worked explanation