On this concept: Formula

Not sure which lesson to do first? Find your level and get a plan →

Statistics · Grade 9-12

Expected Value

The expected value of a random variable is the long-run average outcome of a random process, calculated as the weighted sum of each possible outcome times its probability.

E(X)=∑x⋅P(x)
Venn diagram of Event A and Event BA two-event view of expected value.ABA ∩ B
A: Event A
B: Event B
A two-event view of expected value.

Quick Answer

The expected value of a random variable is the long-run average outcome of a random process, calculated as the weighted sum of each possible outcome times its probability. Use it to compare choices under uncertainty: whether a game, raffle or warranty pays off on average, or which strategy scores more points per attempt. Multiply each outcome by its probability and add, losses included: a $2 game that pays $10 with probability 0.15 is worth 0.15×10−2=−0.50 dollars per play. The expected value need not be a possible single result; a fair die's is 3.5.

Why This Matters

Expected value is the tool for decisions that repeat: it tells a basketball coach that a 35% three-point shot (1.05 points per attempt) beats a 50% two-point shot (1.00 points), tells a store whether a warranty is priced to profit, and shows why lotteries lose money on average despite their jackpots. Without it, people judge a bet by the size of the prize and ignore how unlikely the prize is.

Intuitive Explanation

A carnival game costs $1. You roll a die: a 6 wins $4, and anything else wins nothing. Picture 600 plays. About 100 of them are sixes, paying $400 in total, while you paid $600, so you are down about $200, or about $0.33 per play. That per-play average is the expected value: 16(4)+56(0)−1=−13 dollars.

The formula E(X)=∑x⋅P(x) is a weighted average: each outcome counts in proportion to how often it happens. That is why a huge prize can matter less than it looks. A $2 raffle ticket with a 1-in-500 chance at $600 gives 1500(600)=1.20 dollars of expected winnings; minus the $2 cost, that is −0.80 dollars per ticket. And the expected value need not be a result you can actually get: a fair die's is 1+2+3+4+5+66=3.5, a number no roll shows.

Expected value is not the most likely outcome. In the raffle, by far the most likely result is losing $2, which happens 499 times in 500, yet the expected value is −0.80. It describes the average over many plays, so it guides repeated decisions well. For a one-time choice with a ruinous downside, like betting the rent, how widely the outcomes spread matters as much as the average, and that spread is what Standard Deviation measures.

Think of it like this. Expected value is the balance point of a seesaw: place each outcome on a number line and stack a weight on it proportional to its probability, and the board balances at the expected value. A heavy stack near a small loss and a tiny weight far out at a big prize balance close to the loss, which is why rare jackpots add so little.

Core idea. The expected value is the average result per trial over a long run, found by weighting every possible outcome, losses included, by its probability; it guides repeated decisions even when no single trial can produce that value.


When to Use

Use expected value when a question asks about results on average, per play or in the long run, whether a game is fair, or which of two risky options is better over many tries. You need every outcome's value and probability, losses included. If the question asks only how likely one outcome is, use Basic Probability; if it asks how much the results vary around the average, or how risky one try is, use Standard Deviation.

Tip. Ask: Does each possible outcome carry a value, such as points or dollars, so you need the probability-weighted average per try?

How to Recognize It

Before calculating, list every outcome with both its value and its probability, and check that the question wants an average per play.

  1. Does each outcome have a number attached, like a prize, a loss or a score?

    Yes means you can weight those numbers by their chances. If the question only wants how likely one outcome is, such as the chance of winning, that is Basic Probability.

  2. Is the question about the long-run average per play or per decision rather than the most common result?

    Yes fits Expected Value. The single most frequent result is the Mode; for a raffle it is usually losing the ticket price, while the expected value is a different number.

  3. Are the weights probabilities that add to 1 rather than counts or credit hours?

    Probabilities make this an expected value. Weights such as credit hours, or the number of students in each class, give a Weighted Average of data instead.

  4. Does the decision depend on how much the results swing, not only on their average?

    If two choices have the same expected value but one is far riskier, compare their spread with Standard Deviation as well. Expected value alone cannot tell a sure $10 from a coin flip for $0 or $20.

Expected Value vs Basic Probability vs Weighted Average vs Mode

Expected value uses probabilities, is built like a weighted average, and has a name that sounds like 'most likely', so all four get confused. Ask whether you want an average payoff, the chance of one result, an average of data, or the most common value.

Expected Value

Meaning
Expected value is the long-run average result per trial, found by multiplying each outcome's value by its probability and adding.
Key test
Needs a value and a probability for every outcome; the answer is in points or dollars.
Formula
E(X)=∑x⋅P(x)
Example
A game costing $0.50 that pays $4 on a roll of 6 is worth 16(4)−0.50≈0.17 dollars per play.

Basic Probability

Meaning
Basic probability measures how likely one event is, as a number from 0 to 1, without attaching any values to the outcomes.
Key test
Answers 'how likely', never 'how much'; it always lies between 0 and 1.
Formula
P(E)=favorabletotal
Example
The chance of rolling the 6 in that game is 16, whatever the prize is.

Weighted Average

Meaning
A weighted average combines data values using weights that show how much each one counts, such as credits or frequencies.
Key test
Weights can be any sizes; expected value is the case where they are probabilities.
Formula
xˉw=∑wixi∑wi
Example
A grade from 90 on homework worth 30% and 70 on exams worth 70% is 0.3(90)+0.7(70)=76.

Mode

Meaning
The mode is the value that occurs most often, which is the single most likely outcome when outcomes have probabilities.
Key test
The most frequent result, which can sit far from the expected value.
Example
For a $2 raffle ticket with a 1 in 1,000 chance at $1,000, the most likely result is a $2 loss.

Formula & Notation

E(X)=∑x⋅P(x)
For a discrete random variable X with outcomes x1,x2,…,xk and probabilities p1,p2,…,pk, the expected value is E[X]=∑i=1kxi pi. For a continuous random variable with density f(x), E[X]=∫−∞∞x f(x) dx.

Worked Examples

Example 1 — A card game's value per play

Easy

A game costs $3 to play. You draw one card from a standard 52-card deck: an ace wins $20, a face card (jack, queen or king) wins $5, and any other card wins nothing. Find the expected net gain per play.
Show the steps and the answer
  1. List the outcomes with their probabilities: ace 452, face card 1252, other 3652; check 4+12+36=52.

    Every card falls into exactly one group, so the probabilities must add to 1.

  2. Weight each payout: 20×452=8052, 5×1252=6052, 0×3652=0.

    Each prize counts in proportion to how often it is won.

  3. Add: expected payout =14052≈2.69 dollars per play.

    This is the average amount the game pays back each time.

  4. Subtract the cost: 14052−3=140−15652=−1652≈−0.31 dollars.

    The $3 is paid on every play, win or lose.

  5. Interpret: over 1,000 plays, a player loses about $310.

    Expected value is a per-play average, so it scales with the number of plays.

Answer: Expected net gain = −4/13 ≈ −$0.31 per play.

Takeaway: Weight every payout by its probability, add, then subtract what every play costs.

Example 2 — Usually lose, still a good game

Standard

In a game you roll a fair die. On a 1 through 5 you lose $1; on a 6 you win $8. You lose on most rolls. Is the game good or bad for the player in the long run?
Show the steps and the answer
  1. Find the most likely result: losing $1, with probability 56≈0.83.

    That is the mode, the single result seen most often, and it is a loss.

  2. Compute the expected value: 56(−1)+16(8)=−56+86=36=0.5.

    The rare win is big enough to outweigh the frequent small losses.

  3. Check with 600 rolls: about 500 losses cost $500 and about 100 wins pay $800, a net gain of $300, or $0.50 per roll.

    Imagining many plays turns the weighted average into a running total.

  4. Note the short run: the chance of losing all of your first 6 rolls is (5/6)6≈0.33.

    A positive expected value is reliable over many rolls, but a short session often loses.

Answer: Good for the player in the long run: the expected value is +$0.50 per roll, even though 5 out of 6 rolls lose.

Takeaway: The most likely outcome and the expected value answer different questions; judge a repeated game by the expected value.

Example 3 — Find the mistake: a missing outcome

Application

A company sells a one-year phone plan for $90. It pays $600 if the phone is lost (probability 0.05) and $200 for a cracked screen (probability 0.15). Priya writes: "Expected payout = 600 × 0.05 = $30, so the company's expected profit is $90 − $30 = $60 per plan." Find the mistake and correct it.
Show the steps and the answer
  1. List every outcome for the company: lost phone 0.05 (pays $600), cracked screen 0.15 (pays $200), no claim 0.80 (pays $0); 0.05+0.15+0.80=1.

    Expected value needs all the outcomes, and their probabilities must add to 1.

  2. Spot the gap: Priya used only the lost-phone outcome, whose probability is 0.05.

    Her probabilities add up to 0.05, not 1, which signals that outcomes are missing.

  3. Recompute the expected payout: 600(0.05)+200(0.15)+0(0.80)=30+30+0=60 dollars.

    The cracked screens cost the company as much on average as the lost phones.

  4. Expected profit: $90 − $60 = $30 per plan.

    Over 10,000 plans that is about $300,000 of profit, half of Priya's figure.

Answer: The expected profit is $30 per plan, not $60.

Takeaway: Before adding up, check that the probabilities of the outcomes you used add to 1.

Where You'll See It

  • Carnival ring-toss game

    A $1 game pays $5 with probability 0.1 and $0 otherwise. Expected value of a play is 5(0.1)+0(0.9)−1=−0.50, so on average you lose 50 cents per play even though winning is possible.

  • Raffle ticket worth it?

    If a $2 raffle ticket has a 1/1000 chance at a $1000 prize, E(X)=1000⋅11000−2=−1, meaning each ticket loses a dollar on average over many draws.

  • Choosing a spinner bonus in a video game

    A bonus spinner gives 0, 10, or 100 coins with probabilities 0.5, 0.4, 0.1. Its expected payout is 0(0.5)+10(0.4)+100(0.1)=14 coins — the long-run average that lets you compare it to a flat reward.

Common Mistakes

  1. The slip: The expected value is the outcome you're most likely to get.

    Why it's tempting: The name 'expected' suggests it's what you should expect to see each time.

    Instead: Expected value is a long-run weighted average and may be a value that can never occur on a single trial (e.g., EV of a die is 3.5). The 'most likely' outcome is the mode, a different idea.

  2. The slip: If a game can pay out a big prize, its expected value must be positive.

    Why it's tempting: A large possible win feels like it makes the game worthwhile.

    Instead: Multiply each payoff by its probability and subtract the cost. A rare big prize usually has tiny probability, so most gambling games have negative EV despite the jackpot.

  3. The slip: You only need to count the winning outcome to find expected value.

    Why it's tempting: Students focus on the prize and forget the (often more likely) losing outcomes.

    Instead: Sum x⋅P(x) over ALL outcomes, including the zero/loss cases whose probabilities are usually largest; leaving them out overstates the EV.


Mini Practice

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

  1. A spinner pays $10 with probability 0.2, $2 with probability 0.5 and $0 otherwise. Find the expected value of one spin.

    Hint: Multiply each payout by its probability, then add.

  2. A game costs $2 to play. You roll a number cube: a 6 wins $9, and anything else wins nothing. Find the expected net gain per game.

    Hint: Find the expected winnings, then subtract the cost to play.

  3. A team scores 0 goals with probability 0.3, 1 goal with 0.4, 2 goals with 0.2 and 3 goals with 0.1. Find the expected number of goals per game.

    Hint: Use E(X)=∑x⋅P(x).

  4. An insurer sells a phone policy for $200. There is a 3% chance it pays out $1,500 and a 5% chance it pays out $400; otherwise it pays nothing. What is the insurer's expected profit per policy?

    Hint: Find the expected payout first, then compare it with the price.

  5. A raffle sells 500 tickets at $5 each, with one $1,000 prize. You want to know if a ticket is worth its price on average. Is P(win) enough, or do you need expected value?

    Hint: Does the probability alone include the size of the prize?

  6. A game pays $0, $2 or $5 per play, and its expected value is $1.40. Ari says, "So I will win $1.40 on my next play." Correct Ari.

    Hint: Which amounts can a single play actually pay?

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.

Frequently Asked Questions

Why do people buy insurance if its expected value is negative?

Because expected value ignores how much a rare loss would hurt. A policy costing $300 a year against a 1% chance of a $20,000 loss has an expected payout of $200, so on average you pay $100 a year more than you get back. But a sudden $20,000 bill could be ruinous, and trading a small sure cost for protection from it is a reasonable choice.

Can expected value be used for things other than money?

Yes, any outcome with a number attached works. A basketball coach compares a 2-point shot made 50% of the time, worth 2×0.5=12 \times 0.5 = 1 point per attempt, with a 3-point shot made 36% of the time, worth 3×0.36=1.083 \times 0.36 = 1.08 points. That small edge is one reason teams take far more three-point shots than they did twenty years ago.

Is the expected value of two games the sum of their expected values?

Yes, and this is one of its most useful properties: E(X+Y)=E(X)+E(Y)E(X + Y) = E(X) + E(Y), even when the games affect each other. If one carnival game loses $0.50 per play on average and another loses $0.30, playing both loses $0.80 on average. Ten plays of the first lose about $5, which is why small negative values add up quickly.

Why is a fair game one with an expected value of zero?

A fair game is one where neither side gains in the long run, so the average net result per play is 0. Betting $1 on a coin toss to win $1 is fair: 12(1)+12(−1)=0\frac{1}{2}(1) + \frac{1}{2}(-1) = 0. Casino games are set slightly below zero for the player; in American roulette, a $1 bet on red is worth about −0.053-0.053 dollars, a loss of about 5 cents.

Learning Path

Expected Value

You are here

You need Basic Probability (each outcome's chance) and Weighted Average (values that count in proportion to their weights) first. Expected value combines them into one number for the long-run result of a random process, the basis for judging games, insurance and strategy. Next, Standard Deviation measures how far single results typically land from that expected value, which tells you how risky the process is.

See Also