Expected Value from Outcomes and Probabilities

Expected value summarizes a chance process with one number: the average outcome you would approach over many repetitions under the same probabilities. For a discrete set of outcomes, multiply each outcome by its probability and add the products. In symbols, E(X) = Σxᵢpᵢ. The possible outcomes may be counts, points, dollars, or another measured quantity.

Calculate a weighted average of outcomes

Suppose a game pays $10 with probability 0.2 and $0 with probability 0.8. The expected payout is ($10)(0.2) + ($0)(0.8) = $2. The game does not pay $2 on each play; a player receives either $10 or $0. Across many plays, the average payout per play tends toward $2 if the probabilities stay the same.

A fair six-sided die has outcomes 1 through 6, each with probability 1/6. Its expected result is 1(1/6) + 2(1/6) + 3(1/6) + 4(1/6) + 5(1/6) + 6(1/6) = 21/6 = 3.5. The die cannot show 3.5 on one roll. The value describes the long-run average across rolls.

Check that the probabilities form a distribution

Before multiplying, confirm that each probability is between 0 and 1 and that the probabilities for all listed outcomes add to 1. If outcomes are given as counts, divide each count by the total number of observations to find its probability. A missing outcome or a denominator error can distort the expected value even when the arithmetic is correct.

For a table with outcomes 0, 1, and 2 and probabilities 0.25, 0.50, and 0.25, the check is 0.25 + 0.50 + 0.25 = 1. Then E(X) = 0(0.25) + 1(0.50) + 2(0.25) = 1. The middle outcome has the greatest probability, but the expected value comes from weighting every outcome, not selecting the most likely one.

Include costs and net outcomes

A payout is not always the same as profit. If a ticket costs $3 and pays $10 with probability 0.2 or nothing with probability 0.8, the net outcomes are $7 and −$3. Expected net value is $7(0.2) + (−$3)(0.8) = $1.40 − $2.40 = −$1. The equivalent calculation is expected gross payout, $10(0.2) + $0(0.8) = $2, minus the certain $3 cost, which also gives −$1.

Clearly distinguish gross return, net gain, and cost in a word problem; the question may ask for any one of them. A fixed cost can be subtracted once after the expected payout is calculated, or included in each possible net outcome before weighting. Both methods produce the same answer.

What expected value does and does not say

Expected value is useful for comparing repeated choices, but it does not describe the spread of outcomes or guarantee an individual result. A low-probability loss can make an average unattractive to someone who cannot absorb that loss. Two choices can also have the same expected value but very different risks. Questions on basic probability typically test the weighted calculation and interpretation rather than a complete decision analysis.

Common errors

• Adding the outcomes without weighting them by their probabilities.
• Multiplying by percentages such as 20 instead of converting 20% to 0.20.
• Leaving out an outcome or using probabilities that do not total 1.
• Using the most likely outcome as if it were the expected value.
• Forgetting an entry fee or confusing gross payout with net gain.
• Claiming that the expected value must occur on the next trial.

A useful final check is units: if outcomes are measured in dollars, expected value is also dollars. If every possible outcome is the same constant c, the expected value should be c because probabilities sum to 1. If one outcome becomes more probable while the others change, the weighted average should shift in a direction consistent with that change.

Work from a frequency table

A frequency table can be converted into probabilities by dividing each frequency by the total number of trials. Suppose a delivery service records 20 orders: 12 arrive on time, 6 are one day late, and 2 are two days late. If lateness is the outcome of interest, its expected number of late days per order is 0(12/20) + 1(6/20) + 2(2/20) = 10/20 = 0.5 day. You can get the same result by multiplying each late-day value by its frequency, adding, and dividing by 20.

This form makes the interpretation concrete: across many comparable orders, the average delay is expected to be about half a day per order. It does not imply that an individual order will arrive exactly half a day late. Check whether the table lists all possible outcomes; if on-time orders were omitted, the probabilities would not represent the complete process.

Compare choices using the same basis

Expected values can compare repeated opportunities only when outcomes are measured on the same basis. One game might have an expected net gain of $1 per play, while another has an expected net gain of $5 per ten plays. Convert both to the same unit before comparing. Also make sure the outcome includes every relevant cost; a high gross payout can have a low or negative expected net value after fees.

When possible, use a quick bound as a reasonableness check: a weighted average must lie between the smallest and largest outcomes. If the outcomes are 0, 1, and 2, the expected value cannot be 3. A result outside that interval means a probability, multiplication, or addition error occurred.

Exam takeaway

Make a two-column list of outcomes and probabilities, verify the probabilities total 1, multiply row by row, and add. Include costs as negative outcomes when the question asks for expected net gain. Describe the result as a long-run average, not a certain next result.

Use the concept in a question

Expected value is the long-run average outcome when the same chance process is repeated under comparable conditions. If a game pays $10 with probability 0.2 and $0 with probability 0.8, its expected payout is 10(0.2) + 0(0.8) = $2. If entry costs $3, expected net value is $2 − $3 = −$1 per play. That does not predict the result of one play; a player may win $10 or nothing. Include every possible outcome and its probability, check that probabilities sum to 1, and distinguish gross payout from net gain when the problem includes a cost.