Random Numbers in Computing

Section 1

Quick Answer

Random numbers are values with no predictable pattern. In computing they are usually pseudo-random: a deterministic algorithm starts from a seed and produces a sequence mixed up enough to behave like real chance for most tasks. Reach for this concept when a problem needs a value chosen by chance — a die roll, a shuffle, a sampled value. The cue separating it from Simulation is that the concept is the SOURCE of the random value, not the larger process that uses it; and from a known seed the same sequence repeats.

Section 2

Why This Matters

Randomness appears in simulations, games, testing, sampling, and security. Students need to know that random behavior in software is usually generated, not magical.

Section 3

Intuitive Explanation

Imagine a program that needs to roll a die. It cannot truly throw plastic across a table, so it runs an algorithm that spits out a number from 1 to 6 that you cannot easily predict. That number is a pseudo-random number: the computer followed an exact rule, but the rule scrambles its output so well that, for a game or a test, it behaves just like genuine chance.

The word 'pseudo' is the key idea. Because the values come from a deterministic algorithm, they start from a seed. Give the generator the same seed and it replays the exact same sequence — handy for reproducing a bug, dangerous if you assumed the numbers were truly unpredictable (as in security). Recognizing that software randomness is generated, not magical, is what this concept is really teaching.

Don't confuse the source of a random value with what you do with it. A simulation of a board game or a weather forecast pulls many random numbers to model chance events, but the simulation is the whole model; Random Numbers is just the faucet the chance values flow from. When a question turns on getting one unpredictable value — and on whether a seed makes it repeatable — you are looking at Random Numbers, not the larger process around it.

Core idea

Random numbers help models include uncertainty, variation, and chance.

Section 4

When to Use

Use Random Numbers when a task needs an unpredictable value — a die roll, a coin flip, a shuffle, a random sample — generated by the computer rather than computed from the inputs. The deciding cue is the source of the value: it comes from chance, and in software that usually means a pseudo-random generator driven by a seed. Do not reach for Random Numbers when the focus is the whole chance process being modeled (that is Simulation) or the step-by-step procedure itself (that is Algorithm); here the point is just where the unpredictable value comes from.

Pro tip

Ask what range of values is possible, whether each outcome should be equally likely, and whether repeating the same starting seed should reproduce the same sequence.

Section 5

How to Recognize It

Before using Random Numbers, check whether the task hinges on generating a value by chance — and whether it is pseudo-random from a seed. These questions test that recognition move from a few angles.

Does the task need a value chosen by chance rather than computed from the given inputs?

If you need a die roll, a coin flip, a shuffle, or a random sample, Random Numbers is in play. If every output is determined by the inputs, it is not about randomness.
Is the randomness pseudo-random — produced by an algorithm from a seed?

If the same seed reproduces the same sequence, you are dealing with pseudo-random numbers, not true randomness. Recognizing this is the heart of the concept.
What is the nearest confusion?

Simulation is the trap: it models a whole chance process (a game, a queue, a forecast) and often CONSUMES random numbers, but the concept here is just the source of the unpredictable values. If the prompt is about the overall model, it is Simulation.
Is there a target distribution or range the values should follow?

Asking for a value from 1 to 6, or a uniform pick, signals that generating values to fit a distribution is the point — that is the evidence for Random Numbers.
What would make this NOT Random Numbers?

If the task is to design the procedure (Algorithm) or to run and interpret a full chance experiment (Simulation), the source of the random value is no longer the focus — switch concepts.

Section 6

Random Numbers vs Simulation vs Modeling vs Algorithm

These get mixed up because a simulation often USES random numbers, and an algorithm may generate them. The deciding question is the focus: Random Numbers is about where an unpredictable VALUE comes from, while the others are about the process, the representation, or the procedure around it.

Random Numbers vs Simulation vs Modeling vs Algorithm
Type	Meaning	Key test	Formula	Example
Random Numbers	Use this when the task needs a single unpredictable value — a die roll, coin flip, shuffle, or random sample — that the computer generates rather than computes from the inputs. The deciding question is: does this task need a value chosen by chance, generated rather than derived?	Does this task need a value chosen by chance, generated rather than computed from the inputs (often pseudo-random from a seed)?	$P(r=i)=\tfrac{1}{n}$	Picking a number from 1 to 6 to simulate one die roll, or seeding a generator so a shuffle is repeatable.
Simulation	Fits when the focus is modeling a whole real-world process as it unfolds over time, which may merely USE random draws as one ingredient.	Use when the task is running and observing a whole chance process, not generating one value.	$S_{t+1}=f(S_t, P)$	Modeling how a disease spreads across a population over many days, drawing random infections each step.
Modeling	Fits when the focus is building a simplified representation of a real system — deciding what to include and ignore — so you can study or predict its behavior.	Use when the task is choosing what a representation should capture, not generating a chance value.	$\text{output}=f(\text{inputs}, \text{assumptions})$	Representing traffic by car speed and road capacity while ignoring each car's color.
Algorithm	Fits when the focus is the step-by-step procedure itself — the recipe of instructions — rather than where a chance value comes from.	Use when the task is describing the ordered steps that solve a problem.	$\text{output}=f(\text{input})$	The ordered steps of long division, or directions for getting from one place to another.

Random Numbers

Meaning: Use this when the task needs a single unpredictable value — a die roll, coin flip, shuffle, or random sample — that the computer generates rather than computes from the inputs. The deciding question is: does this task need a value chosen by chance, generated rather than derived?
Key test: Does this task need a value chosen by chance, generated rather than computed from the inputs (often pseudo-random from a seed)?
Formula: $P(r=i)=\tfrac{1}{n}$
Example: Picking a number from 1 to 6 to simulate one die roll, or seeding a generator so a shuffle is repeatable.

Simulation

Meaning: Fits when the focus is modeling a whole real-world process as it unfolds over time, which may merely USE random draws as one ingredient.
Key test: Use when the task is running and observing a whole chance process, not generating one value.
Formula: $S_{t+1}=f(S_t, P)$
Example: Modeling how a disease spreads across a population over many days, drawing random infections each step.

Modeling

Meaning: Fits when the focus is building a simplified representation of a real system — deciding what to include and ignore — so you can study or predict its behavior.
Key test: Use when the task is choosing what a representation should capture, not generating a chance value.
Formula: $\text{output}=f(\text{inputs}, \text{assumptions})$
Example: Representing traffic by car speed and road capacity while ignoring each car's color.

Algorithm

Meaning: Fits when the focus is the step-by-step procedure itself — the recipe of instructions — rather than where a chance value comes from.
Key test: Use when the task is describing the ordered steps that solve a problem.
Formula: $\text{output}=f(\text{input})$
Example: The ordered steps of long division, or directions for getting from one place to another.

Section 7

Formula & Notation

P(r = i) = \frac{1}{n}

A random number generator produces values intended to approximate a target probability distribution. In many programs, the generator is pseudo-random and controlled by an initial seed value.

Section 8

Worked Examples

Example 1 — Recognize the model

Easy

Problem

A class sees this computing situation: students convert a small image or sound into numbers and explain what information is kept, simplified, or lost. How should a student decide whether Random Numbers is the right model?

Solution

Identify the target of the reasoning.

The target might be a problem, data representation, code state, system component, user need, or stakeholder.
List the process or relationship that matters.

Random Numbers is useful when the problem asks for a data explanation with representation, units or structure, transformation rule, possible loss, and interpretation stated.
Apply the recognition test: Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information?

This separates random numbers from raw real-world object and algorithm.
State the evidence that would prove the answer.

A trace, test, diagram, input-output pair, or impact argument prevents a vague answer.

Answer

Use Random Numbers only if the task is asking for a data explanation with representation, units or structure, transformation rule, possible loss, and interpretation stated and the situation passes the recognition test. Otherwise, choose the nearby model that better matches the computing structure.

Takeaway: Model choice comes before definitions. The same words can belong to different CS ideas depending on the problem structure.

Example 2 — Avoid the vocabulary trap

Standard

Problem

A student says, "This prompt contains the word data, so I should use random numbers." Explain why that shortcut is risky.

Solution

Treat the word as a clue, not proof.

CS vocabulary overlaps across problem solving, programming, data, systems, design, and impact questions.
Check whether the target and process match Random Numbers.

The computing structure decides the model.
Compare with Raw real-world object and Algorithm.

A computer stores a representation of the object, not the object itself. An algorithm processes data; the representation decides what data the algorithm can see.
State what the final result would mean.

If the final result would not mean a data explanation with representation, units or structure, transformation rule, possible loss, and interpretation stated, the model is probably wrong.

Answer

The shortcut is risky because data can appear in several related CS models. The student must first show that the task answers "Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information?" with yes.

Takeaway: A CS thinking concept is a reasoning tool, not just a vocabulary match.

Example 3 — Write the computing conclusion

Application

Problem

After solving a Random Numbers problem, a student writes only a definition. What should be added to make the answer useful?

Solution

Name the specific case.

The answer should identify the input, data, program state, system component, user, or stakeholder being described.
Show the process or evidence.

A trace, test, example, diagram, or tradeoff explains why the concept applies.
Connect the result to the goal.

The final sentence should say how the concept helps solve, test, design, represent, protect, or evaluate the computing situation.
Mention limits or edge cases.

Computing answers are stronger when they state where the method might fail, scale poorly, exclude users, or require a different design.

Answer

A complete answer should say what random numbers controls in the specific situation, include evidence such as a trace or test, and state any condition needed for the model to apply.

Takeaway: The final explanation is part of CS thinking, not an optional sentence after the term.

Section 9

Common Mistakes

Common slip-up

Assuming pseudo-random numbers are truly unpredictable in every context

The right idea

Fix this by naming the input, process, output, evidence, and checking "Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information?" before using the concept.

Common slip-up

Using random values without checking whether the intended distribution is uniform

The right idea

Fix this by naming the input, process, output, evidence, and checking "Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information?" before using the concept.

Common slip-up

Forgetting that the same seed can recreate the same sequence

The right idea

Fix this by naming the input, process, output, evidence, and checking "Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information?" before using the concept.

Common slip-up

Using random numbers from a keyword alone

The right idea

Signal words like data, binary, bits only point to a possible model; the computing structure must match too.

Section 10

Mini Practice

Try these on your own. Tap Reveal when you want to check.

Which concept fits: a card game needs to deal a fresh, unpredictable hand each round, so it must produce a value to pick which of 52 cards comes next?

Hint: Is the task generating a chance value, or modeling a whole process over time?
Why is this a contrast case (Simulation, not Random Numbers): a program models the spread of a flu across a town over 30 days, each day randomly infecting some neighbors, and you must report the total infected at the end.

Hint: Is the point one chance value, or the whole process the chance values feed?
Why does seeding matter: a tester runs a dice game twice with the same seed and gets the exact same sequence of rolls. Is that a bug?

Hint: Recall how pseudo-random generators produce their sequence.

Want the full set?

50 practice questions for this concept — free to try, every one with a complete worked solution showing the why, not just the answer.

Practice this concept → Take a mastery check

Section 11

Frequently Asked Questions

What are random numbers in simple terms?

They are values chosen with no predictable pattern — the kind of value you'd get from rolling a die or flipping a coin. On a computer they are usually pseudo-random: a deterministic algorithm starts from a seed value and produces a sequence mixed up enough to behave like real chance for most purposes. So 'random' in software usually means 'looks random, but is actually generated by a rule from a seed.'

How do I know a problem wants Random Numbers rather than Simulation?

Ask: is the point the unpredictable VALUE itself, or the whole process it feeds? If the task is to produce one chance value — roll a die, shuffle a deck, pick a random sample — that's Random Numbers. If the task is to model an entire process over time that merely consumes those values, like disease spread or weather, that's Simulation. The simulation is the big machine; random numbers are one part it draws on.

Why are computer random numbers called pseudo-random?

Because they are not truly unpredictable — a deterministic algorithm produces them from a starting seed, so the same seed always yields the same sequence. That repeatability is useful for testing and reproducing results, but it also means they aren't suitable everywhere. The common mistake is assuming pseudo-random values are genuinely unpredictable in every context, such as security, where that assumption can fail.

What is the most common mistake with Random Numbers?

Trusting that the values are uniform or truly unpredictable without checking. A generator is meant to approximate a target distribution, but you should confirm the intended distribution is actually what you need — for a fair die each face should have probability $\tfrac{1}{6}$ , $P(r=i)=\tfrac{1}{n}$ . Also remember pseudo-random sequences are reproducible from their seed, so don't treat them as if no one could predict them.

Section 12

Learning Path

← Before

Simulation

Random Numbers

You are here

Next →

Modeling

Before this, students should be comfortable with Simulation. This page focuses on the recognition cue: Am I explaining how data is encoded, organized, transformed, or interpreted rather than only naming the information? That cue connects earlier computing descriptions to later problem solving because students first choose the model, then choose the representation, code, test, diagram, or explanation. After this, Modeling become easier to recognize.

Section 13