Historical Context & Motivation
Imagine you're offered a bet: pay $5 to roll a die, and you win $20 if you roll a 6. Should you take it? Your gut might say "maybe," but mathematicians wanted a precise way to answer questions like this. The concept of expected value was born from exactly this kind of question — how do we assign a fair numerical value to uncertain outcomes? The story begins with gamblers, letters between friends, and a puzzle about dice.
The central question that expected value answers is deceptively simple: if I could repeat a random experiment forever, what single number would my average result settle toward? That number might not even be a possible outcome of the experiment — and that's perfectly okay. The rest of this lesson shows you how to compute it and why it matters.
Core Principles & Definitions
Before diving into calculations, let's nail down the key ideas that make expected value work. Each principle below builds on the previous one, moving from basic probability language to the full concept of expected value.
Discrete Random Variable
Probability Distribution
Weighted Average
Long-Run Average Interpretation
Expected Value May Not Be a Possible Outcome
Visual Explanation — The Balance Point
The diagram below shows the probability distribution for a single roll of a fair six-sided die. Each outcome (1 through 6) has an equal probability of 1/6. The bars represent those equal probabilities, and the dashed line marks the expected value of 3.5 — the exact balance point of the distribution. Notice that 3.5 sits right in the middle because the distribution is perfectly symmetric.
Now consider what happens when a distribution is not symmetric. If certain outcomes are more probable, the balance point shifts toward those heavier outcomes. That's why expected value is a weighted average — the probabilities act as weights. In Section 5 you'll see a diagram where the probabilities are unequal, and the expected value is pulled away from the center.
Mathematical Framework
Let's formalize the idea. Suppose a discrete random variable X can take on the values x₁, x₂, …, xₙ with corresponding probabilities P(x₁), P(x₂), …, P(xₙ). The expected value of X, written E(X) or sometimes μ (the Greek letter mu), is defined by the formula below.
Using summation notation, this can also be written more compactly.
There's one important requirement for the formula to be valid: the probabilities must form a legitimate distribution. That means every P(xᵢ) must satisfy 0 ≤ P(xᵢ) ≤ 1, and all probabilities must sum to 1.
Detailed Breakdown — Unequal Probabilities
Expected value gets really interesting when the probabilities aren't all the same. Consider a carnival game where you spin a wheel divided into unequal sections. The table below shows the prize amounts and their probabilities.
| Prize (x) | Probability P(x) | Product x · P(x) |
|---|---|---|
| $0 | 0.50 | $0 × 0.50 = $0.00 |
| $2 | 0.30 | $2 × 0.30 = $0.60 |
| $5 | 0.15 | $5 × 0.15 = $0.75 |
| $20 | 0.05 | $20 × 0.05 = $1.00 |
| Total | 1.00 ✓ | E(X) = $2.35 |
Even though $20 is the biggest prize, it only contributes $1.00 to the expected value because it's so unlikely (5% chance). Meanwhile, the $0 outcome has the largest probability and contributes nothing. The expected value of $2.35 tells you that if you played this game hundreds of times, your average prize would settle near $2.35 per spin. If the game costs $3 to play, you'd lose money in the long run — a fact the carnival is counting on.
Notice how the dashed green line is pulled strongly to the left — toward the outcomes with the biggest probabilities. The rare $20 prize barely shifts the balance. This is the power of the weighted-average perspective: probability determines influence.
Worked Example — Should You Play the Game?
A school fundraiser sells raffle tickets for $5 each. The prizes and probabilities are shown below. Let's compute the expected value of the raffle ticket and decide whether it's a good deal for the buyer.
| Prize | Value | Probability |
|---|---|---|
| Grand Prize | $100 | 1/200 = 0.005 |
| Second Prize | $25 | 5/200 = 0.025 |
| Small Prize | $5 | 20/200 = 0.10 |
| No Prize | $0 | 174/200 = 0.87 |
Strengths and Limitations of Expected Value
Expected value is one of the most useful numbers in all of probability, but it doesn't tell you everything. Understanding both its strengths and its limitations will help you use it wisely.
| Strengths | Limitations |
|---|---|
| Summarizes a distribution in a single number, making comparisons easy. | Ignores spread — two distributions can have the same E(X) but very different variability. |
| Accurately predicts long-run averages over many repetitions. | Doesn't predict any single trial — your result could be far from E(X). |
| Works for any discrete random variable, regardless of the shape of its distribution. | Can be misleading when distributions are highly skewed (e.g., lotteries). |
| Foundation for decision-making in business, insurance, and game theory. | Doesn't account for risk tolerance — losing $1,000 once may matter more than gaining $10 one hundred times. |
Connection to Advanced Theory
The expected value you're learning now is the foundation for a tower of more advanced ideas in statistics and probability. Here's a preview of how this concept grows as you move into AP Statistics, college courses, and real-world applications.
| What You Know Now | Where It Leads |
|---|---|
| E(X) for finite, discrete outcomes | E(X) for continuous random variables, computed using integrals (calculus-based probability) |
| Long-run average interpretation | The Law of Large Numbers (formal proof that the sample mean converges to E(X)) |
| Expected value of a single variable | Linearity of expectation: E(aX + bY) = aE(X) + bE(Y), a powerful rule for combining random variables |
| Deciding if a game is fair (E(X) = 0) | Expected utility theory in economics — adjusting for risk preferences, not just averages |
| Variance as a companion to E(X) | Var(X) = E(X²) − [E(X)]², a formula that itself relies on expected value |
One particularly elegant result is the linearity of expectation. It says that the expected value of a sum of random variables is always the sum of their expected values — even if the variables are dependent. This property makes expected value far easier to work with than many other statistical measures. It's the reason you can compute the expected total of rolling five dice by simply multiplying 3.5 × 5 = 17.5.
Practice Problems
Test your understanding with these five problems, arranged from foundational reasoning to critical thinking. Show your work and check that probabilities sum to 1 before computing E(X).
Lesson Summary
The expected value of a discrete random variable is computed by multiplying each possible outcome by its probability and summing all the products: E(X) = Σ xᵢ · P(xᵢ). This weighted average gives outcomes with higher probabilities more influence over the result. Before computing, always verify that the probabilities in your distribution sum to 1.
The expected value represents the long-run average — the value your sample mean approaches as the number of trials grows, a result guaranteed by the Law of Large Numbers. It may not equal any actual possible outcome (like 3.5 on a die). While E(X) is powerful for comparing options and evaluating fairness, it doesn't capture variability or risk — for that, you'll soon learn about standard deviation and variance.