design and run a simulation to estimate the probability of a compound event, and compare it with the exact probability.
Phase 4: Build
Do Now. A family plays the Marble Bag twice. P(yellow) = 1/5 each time, and the marble goes back in the bag after each draw.
Is the family more likely to win at least one prize, or to win none?
The reading opens over the lesson; close it to come back to exactly where you were.
What is a simulation? A simulation is a model of a chance process. We use a coin, a number cube, a spinner or random digits that have the same probabilities as the real situation. Simulations are useful when the real thing is slow, costly or hard to count.
Four steps. (1) Choose a tool with the right probability. (2) Decide what one trial is. (3) Run many trials and record each one. (4) Find the relative frequency and compare it with the exact probability if you know it.
Worked example 1: Hana simulates two Marble Bag plays. P(yellow) = 1/5 = 2/10, so she lets the random digits 0 and 1 stand for yellow and 2 to 9 for other colors. One trial is a pair of digits (two plays). Her sample digits were: 91 74 00 29 75 50 47 36 88 13 98 49 16 51 56 47 13 41 09 35 73 89 80 53 96. The pairs with a 0 or a 1 are 91, 00, 50, 13, 16, 51, 13, 41, 09 and 80. That is 10 of 25 trials, so the estimate is 10/25 = 0.4. The exact probability is 0.36, so her simulation is close.
Worked example 2: Tyrell simulates a carnival puzzle. Our phenomenon asks about 7 heads in a row. Tyrell uses a calculator’s random number tool to make 7 coin flips per trial. The exact probability of 7 heads in a row is (1/2)7 = 1/128, about 0.008. In 300 trials he expects only about 2 or 3 runs of 7 heads. Rare, but it happens.
A simulation is only as good as its tool. If the tool’s probabilities do not match the real game, the results mean nothing. That is why NIST makes its random numbers public and checkable.
Plan each simulation before you run it.
Simulate a question for your booth card.
Choose one question about your booth, such as “What is the chance a family wins at least once in 2 plays?” Write the four simulation steps, run at least 25 trials, and give the relative frequency. If you can, find the exact probability with a tree or the complement and compare.
Plan a simulation for Spin and Flip (P(win) = 1/8) using three coins. Hint: how many outcomes do three coins have? Which outcome will stand for a win?
A simulation lets us play a game a thousand times without spending a dime.
Exit ticket: Jada simulates the Marble Bag (P(yellow) = 1/5) with a coin, heads for yellow. What is wrong with her plan? Suggest a better tool.
Pacing: warm-up 5 minutes, reading 5, mini lecture 10, guided practice 13, independent 5, closure 2.
Likely misconceptions: using a tool with the wrong probability (Jada’s coin); counting the number of yellow digits instead of trials that contain one; thinking a simulation must match the exact answer.
Supports: printed random-digit strips; calculators with a random-integer key; number cubes. Stretch: estimate P(at least one 6 in 3 rolls); exact 1 − (5/6)3 = 91/216 ≈ 0.42.
Answer guide: warm-up 16/25 = 0.64 none, 0.36 at least one; number cube 1 or 2; 0.4; 128; true; digits 0-6. Reading: public, signed pulses let anyone check that the numbers were not changed; recording every trial lets others check our estimate too. Spin and Flip plan: three coins have 8 outcomes, so let HHH stand for a win (1/8). Exit ticket: a coin gives 1/2, not 1/5; use random digits 0-1 for yellow, a 10-section spinner with 2 yellow parts, or the bag itself. Hana’s digit row is fixed sample data so every student can check the count of 10.
National Institute of Standards and Technology, Interoperable Randomness Beacons project page · U.S. government work, public domain
The National Institute of Standards and Technology (NIST) is a U.S. government science agency. Since 2011, NIST has run a project called the Randomness Beacon. A beacon is a public service that sends out fresh random numbers on a schedule, for example once a minute.
Each pulse from the beacon holds a new 512-bit random string. A bit is a single digit that is either 0 or 1. NIST builds each string by combining randomness from at least two separate random number generators. Every pulse is numbered, time-stamped and signed, and any past pulse can be looked up by anyone.
Because every past pulse is public and signed, anyone can check later that the random numbers were not changed. NIST says this helps the auditability and transparency of services that depend on random processes, such as choosing a random sample of records to check.
A simulation works the same way. We replace a real game with random numbers that have the same probabilities. If the random numbers are fair and we record every trial, anyone can repeat our test and check our results.
Source: National Institute of Standards and Technology, Interoperable Randomness Beacons project page (nist.gov)