Opening subject page...
Loading your content
Learn how to calculate what you can expect to win — or lose — on average when probability and money collide.
The idea that you can assign a numerical value to the "average" outcome of an uncertain event may seem obvious today, but it took centuries of mathematical thinking to formalize. The concept of expected value — the foundation for calculating an expected payoff — was born from a simple human activity: gambling. Long before anyone wrote down a formula, players at card tables and dice games intuitively sensed that some bets were better than others. Turning that intuition into a precise, repeatable calculation changed not only mathematics but also economics, insurance, and everyday decision-making.
The central question that drove all of this work remains the same one you'll learn to answer in this lesson: If I play a game of chance many, many times, how much can I expect to win or lose per play on average? That answer is the expected payoff.
Before you can calculate an expected payoff, you need a solid understanding of a few key ideas. Each one builds on the previous, so take them in order. If you're comfortable with basic probability (you know how to find the probability of rolling a 4 on a standard die, for instance), you already have the prerequisite knowledge to learn this concept.
To see how expected payoff works visually, consider a simple spinner game. A circular spinner is divided into four colored regions of different sizes, and each region is labeled with a dollar payoff. The larger a region, the higher its probability of being selected. The diagram below shows the spinner alongside a bar chart of each outcome's weighted contribution to the total expected value.
Notice how the large red bar (the loss) nearly cancels out the three positive bars, but not quite. The player comes out ahead by fifty cents per spin on average. This doesn't mean you win fifty cents every time — in fact, no single spin can yield exactly $0.50. What it means is that if you played this game hundreds of times, your total winnings divided by the number of spins would approach $0.50. That is the power of the expected payoff: it gives you a single number that summarizes the long-run behavior of a random game.
The expected payoff is calculated using a straightforward formula. Suppose a game has n possible outcomes. Each outcome i has a payoff value xi (positive for a gain, negative for a loss) and a probability P(xi) of occurring.
In compact summation notation this is written as:
Let's break this down step by step so the formula feels concrete rather than abstract.
Step 1 — List all possible outcomes. Identify every distinct result the game can produce. Make sure you haven't left any out; the probabilities must cover the entire sample space.
Step 2 — Assign a payoff value to each outcome. Use positive numbers for money you receive and negative numbers for money you pay. If you must pay $5 to play and you win $20 in one outcome, the net payoff for that outcome is $20 − $5 = +$15. If you lose, the payoff is −$5 (the entry fee you don't get back).
Step 3 — Determine each outcome's probability. These might come from the rules of the game, from counting equally-likely outcomes, or from given data. Verify that all probabilities add up to 1.
Step 4 — Multiply and sum. For each outcome, multiply the payoff by the probability. Add all the products together. The result is E(X), the expected payoff per play.
Step 5 — Interpret. If E(X) > 0, the game favors you over the long run. If E(X) < 0, you can expect to lose money over time. If E(X) = 0, the game is mathematically fair.
In practice, the cleanest way to organize a payoff calculation is with a probability distribution table. The table below shows every outcome, its probability, and its weighted contribution. This format makes it nearly impossible to miss a term in the sum.
| Outcome | Payoff (xᵢ) | Probability P(xᵢ) | xᵢ × P(xᵢ) |
|---|---|---|---|
| Win big | +$50 | 0.05 | +$2.50 |
| Win small | +$10 | 0.15 | +$1.50 |
| Break even | $0 | 0.30 | $0.00 |
| Lose | −$5 | 0.50 | −$2.50 |
| Totals: | 1.00 ✓ | +$1.50 |
The final column sums to +$1.50, which is the expected payoff. This means that, over many repetitions, you'd average a profit of $1.50 per play. Should you play? From a purely mathematical standpoint, yes — a positive expected value favors the player.
Here's a complete worked example that walks through every step of the expected-payoff calculation. Read through it carefully, and try to anticipate each step before reading the answer.
Expected payoff is a powerful decision-making tool, but like any mathematical model, it has boundaries. Understanding both what it does well and where it falls short will help you use it wisely.
| Aspect | Strength | Limitation |
|---|---|---|
| Long-run prediction | Accurately predicts the average result over many repetitions (Law of Large Numbers). | Tells you nothing about what will happen on any single play. |
| Comparing options | Gives a clean, apples-to-apples number for comparing two different games or bets. | Ignores your personal risk tolerance — a $10,000 loss may matter more to you than a $10,000 gain, even if E(X) = 0. |
| Simplicity | Requires only basic multiplication and addition — no advanced math. | Requires knowing the exact probabilities, which aren't always available in real life. |
| Fairness check | Immediately reveals whether a game favors the house, the player, or neither. | Doesn't account for variance — two games with the same E(X) can feel very different if one is high-risk, high-reward. |
The expected payoff calculation you've learned is a first step into a much larger world of probabilistic decision-making. As you continue in statistics and mathematics, you'll encounter more sophisticated tools that build directly on this foundation.
| This Lesson | Advanced Extension | What It Adds |
|---|---|---|
| Expected Value E(X) | Variance & Standard Deviation | Measures how spread out outcomes are around E(X). Two games may have the same expected payoff but very different levels of risk. |
| Single-game E(X) | Law of Large Numbers | Proves rigorously that the sample mean converges to E(X) as the number of trials → ∞. This is why expected value works. |
| Payoff × Probability | Expected Utility Theory | Replaces raw dollar values with a "utility function" that reflects how much a person values each dollar. Explains why people buy insurance even though E(X) of insurance is negative. |
| Discrete payoffs | Continuous Distributions | When outcomes form a continuum (e.g., stock returns), the sum becomes an integral: E(X) = ∫ x · f(x) dx. The idea is identical; only the technique changes. |
For now, the discrete expected-payoff calculation is the tool you'll use most often on standardized tests and in introductory statistics courses. Master it, and these advanced concepts will feel like natural extensions rather than unfamiliar territory.
Work through these five problems in order. They increase in difficulty, starting with a conceptual check and building to a problem that asks you to synthesize multiple ideas from the lesson.
The expected payoff of a game of chance is the long-run average amount you can expect to win or lose per play. To find it, you list every possible outcome, assign each outcome a net payoff (prize minus any entry cost), determine the probability of each outcome, and then compute the sum E(X) = Σ xᵢ · P(xᵢ). The result is a single number: positive means the game favors the player on average, negative means it favors the house, and zero means the game is mathematically fair.
This concept, rooted in the 17th-century work of Pascal, Fermat, and Huygens, remains one of the most practical tools in all of probability and statistics. It powers real-world decisions in insurance pricing, investment analysis, game design, and everyday choices under uncertainty. Remember: the expected value doesn't predict what will happen on one play — it predicts the average across many plays. Always verify that your probabilities sum to 1, always use net payoffs (after subtracting any cost to play), and always interpret your result in context.