Chapter I · 18 min

Sample Spaces

From counting to chance

“The most important questions of life are, for the most part, really only problems of probability.”
— Pierre-Simon Laplace, Théorie analytique des probabilités (1812)

You already know how to count. You can list the ways to seat four people, the five-card hands in a deck, the binary strings of length $n$ . Probability begins the moment you decide that some of those lists are lists of things that might happen, and that — for a while — nothing distinguishes one item from another except its name.

If this subject has ever made you feel slow, that was the writing, not you. Every intimidating word in later chapters is a plain idea with a fancy badge. We will pin the badge on only after the idea is sitting in a sentence you could say over chai.

That last clause is the whole classical theory. A fair die is fair because its six faces are interchangeable. A shuffled deck is fair because every ordering is interchangeable. Once you believe that, chance is a ratio of counts.

The sample space

A sample space $Ω$ is the set of mutually exclusive, exhaustive descriptions of how an experiment can come out. Each description is an outcome. We write $ω$ for a typical one.

Exhaustive means: something in $Ω$ will happen. Mutually exclusive means: two of them cannot happen together. If you are sloppy about either, later arithmetic will quietly lie to you.

A fair die as a sample space. The event “even” is the teal subset {2,4,6}. Classical probability is three cells out of six.

For one fair die, $Ω = {1, 2, 3, 4, 5, 6}$ . For two distinguishable dice, $Ω$ has thirty-six ordered pairs. For a coin flipped until heads appears, $Ω = {H, TH, TTH, \dots}$ — already infinite, and already a warning that “count the points” will not be the last word.

The classical rule

An event is a subset $A \subseteq Ω$ . In the classical world, where every outcome is equally likely and $Ω$ is finite,

P (A) = \frac{∣ A ∣}{∣Ω∣} .

The numerator is the count of favourable outcomes; the denominator is the count of possible ones. This is not a theorem. It is a modelling choice: the choice that the labels on the points of $Ω$ carry no extra weight.

Two distinguishable dice. Each of the 36 cells is one outcome. The terracotta diagonal is the event {sum = 7}.

Where counting does the work

The classical rule is only as good as your enumeration. Three habits pay rent for the rest of the book:

Decide whether order matters before you count. Ordered lists and unordered hands are different sample spaces.
Make outcomes equally likely by construction. “A random five-card hand” means each of the $(5 52)$ hands has the same weight. “A random card, then another” is a different experiment.
When a story has stages, write the sample space as a product, or draw a tree. Do not jump to the event.

When equally likely fails

A loaded die still has six faces, but $P ({6})$ is no longer $1/6$ . A thumbtack lands up or down, and nobody gave you a reason to call those two outcomes interchangeable. A waiting time can be any positive real number — there is no “number of points” to divide.

The rest of this book is the repair. We will keep the sample space, keep events as subsets, and replace the ratio-of-counts with a function $P$ that assigns weights. Counting remains the special case in which every singleton weighs the same.

If you want to feel the classical rule with your hands, open the die lab and build events by clicking faces.

History

Pascal, Fermat, and a match that would not finish

People gambled for millennia without a sample space. Gerolamo Cardano’s Liber de Ludo Aleae (written around 1564, printed only in 1663) already counted faces of dice, but it mixed arithmetic with superstition. The idea that “what might happen” should be written as a set of interchangeable atoms is Pascal and Fermat’s, in their letters of 1654: they listed remaining futures of an interrupted match, treated them as equally likely, and took a ratio. Christiaan Huygens’s De ratiociniis in ludo aleae (1657) made that counting public. The word “sample space” itself is twentieth-century — von Mises and then Kolmogorov — but the object is the 1654 list. Whenever this book writes Ω, it is doing what Pascal did with a pot of money.

Foundations studio: make the idea yours

This extended studio deliberately slows the pace. It is for a first-time learner who wants to recognize the idea in a new story, not merely reproduce a formula. Work with pencil and paper. Predict before calculating; redraw the pictures; and finish every numerical answer with a sentence in ordinary language.

A mental map before more algebra

A sample space is not discovered like a fossil; it is chosen to answer a question. The same physical experiment can support several useful sample spaces.

A working map for Sample Spaces. Cover the labels and reconstruct the chain from memory.

Do not treat the arrows as a theorem. They are a study aid. A strong probability habit is to move back one box whenever a formula feels unmotivated: ask what the experiment is, what information is available, and what quantity the question actually requests.

Three formulas worth being able to narrate

O m e g a = t e x t a l l o u t co m es a tt h ec h ose n r eso l u t i o n

Read this line from left to right and explain what every symbol refers to in the experiment. If a symbol has no story, the model is not finished.

A s u b se t e q O m e g a

Now read the statement backwards: what would have to be known to use it? Backwards reading is often the difference between recognizing a formula and knowing when it applies.

ma t hbb P (A) = ∣ A ∣/∣ O m e g a ∣ q u a d t e x t w h e n o u t co m es a r ee q u a l l y l ik e l y

Test the expression at an edge case or simple symmetric case. Probability formulas should survive sanity checks before you trust the arithmetic built on them.

Worked example ladder

A small experiment you can actually do

A five-step habit for every example in this chapter.

What usually goes wrong

When you notice this mistake, do not merely correct the final number. Return to the first line where the model became ambiguous. Probability errors are often representation errors wearing arithmetic clothing.

Questions beginners are right to ask

Is there one correct sample space?

Usually there are many mathematically valid choices. The best one is fine enough to express the events of interest and simple enough to work with.

Why label coins or dice?

Labels preserve order and identity. You may later forget them deliberately, but you cannot recover distinctions that were never represented.

When is favourable over total valid?

Only when the elementary outcomes being counted have equal probability under the model.

Where the abstraction earns its keep

For each application, ask what counts as an outcome, what the model treats as random, and which assumptions are approximations. This is how probability becomes a modelling language instead of a catalogue of formulas.

Connections: do not store chapters in separate boxes

Problem-solving clinic: from recognition to fluency

There is a stage where every worked example looks clear but a fresh problem still feels foreign. The cure is not another formula; it is practice choosing the representation. Before equations, do a sixty-second scan: identify the experiment, what is known, what remains uncertain, the quantity being asked for, the assumption doing the heavy lifting, and one impossible answer that gives you a sanity bound.

The expert loop returns every calculation to the original story.

Case clinic A: Two coins, two questions

For two labelled tosses, use HH, HT, TH, TT. If the question asks only for the number of heads, the random variable collapses those four outcomes to 0, 1, 2. The physical trial has not changed; only the description has.

Case clinic B: Cards and order

Drawing two cards without replacement can be represented by ordered pairs when order matters, or by two-card subsets when it does not. Mixing these representations halfway through a calculation is a classic source of factors of two.

Solve or reason about it twice: once exactly and once with a rough estimate, simulation, or symmetry argument. If the two approaches disagree dramatically, investigate before trusting the more sophisticated calculation.

Debug a confident wrong answer

Two questions to answer without notes

Is there one correct sample space? Usually there are many mathematically valid choices. The best one is fine enough to express the events of interest and simple enough to work with.

Why label coins or dice? Labels preserve order and identity. You may later forget them deliberately, but you cannot recover distinctions that were never represented.

A notebook protocol for proficiency

Give this chapter one notebook page divided into four quadrants: picture, formula, example, mistake. Redraw the main visual from memory, narrate one formula in English, invent a fresh story using the same mathematics, and record the most tempting wrong move. Revisit the page after two days and again after a week.

A mastery check before you move on

Try these without looking back. If one item feels slippery, return to the corresponding example and rebuild it rather than memorizing the answer.

Give a one-minute explanation of the chapter title to a curious teenager without a formula.
Invent a tiny example with at most six elementary outcomes and solve it completely by enumeration.
State one assumption that would make your example invalid and identify exactly which line would break.
Draw the mental map from memory and connect at least two boxes to an earlier or later chapter.
Write one question whose answer you still do not know. Good questions show that the concept has become active rather than passive.