Linearity of Expectation

Linearity of expectation is the rule that the expected value of a sum equals the sum of the expected values. In Intro to Probability, it lets you find average totals without needing variables to be independent.

What is Linearity of Expectation?

Linearity of expectation is the rule that if you add random variables, you can add their expected values too. In notation, E[X + Y] = E[X] + E[Y], and more generally E[X1 + X2 + ... + Xn] = E[X1] + E[X2] + ... + E[Xn].

The big payoff in Intro to Probability is that this works even when the random variables are dependent. That surprises a lot of people, because many probability rules do require independence. For expectation, though, you do not need the outcomes to be separate or unrelated. The average of the total still comes from adding the averages.

This is why the rule shows up so often in counting problems. If you define one random variable for each object, event, or outcome, then the expected total is just the sum of the individual expected values. For example, if a game has several dice rolls, the expected total score is the sum of the expected score from each roll, even if some rule in the game links the rolls together.

A useful way to think about it is that expectation is an averaging operator. Averaging is linear, so it respects addition and constant scaling. That means E[aX + bY] = aE[X] + bE[Y] as well. Negative values do not break anything, either, because the rule is about the algebra of averages, not about whether the outcomes are positive.

What linearly of expectation does not do is simplify everything about random variables. It does not tell you how variances add unless you have extra conditions, and it does not let you multiply expectations in the same easy way. A common mistake is to assume that if E[X + Y] is easy, then E[XY] should work the same way. It does not.

In practice, this rule is one of the fastest tools in probability because it lets you avoid building a full distribution for a total. Instead of listing every possible combined outcome, you can break the problem into pieces, compute each expected value, and add them up.

Why Linearity of Expectation matters in Intro to Probability

Linearity of expectation is the shortcut that makes many Intro to Probability problems manageable. When a problem asks for the expected number of heads, the expected total score, or the expected count of something happening across many trials, you often do not need to find the whole probability distribution first.

That matters because full distributions can get messy fast. A sum of several random variables may have lots of possible outcomes, especially when the variables are dependent or when the sample space is large. Linearity lets you work one piece at a time, which is easier to set up and easier to check.

It also teaches a good modeling habit. In many probability problems, you define indicator variables, one for each event you want to count, then use expectation to find the average total. That technique shows up in problems about games, random processes, selection without replacement, and other situations where counting directly would be clunky.

The term also connects to later ideas like variance and covariance. Once you know expectation adds neatly, you can see why variance needs extra care and why dependence matters there. So this concept is not just a formula, it is a way of organizing probability work.

How Linearity of Expectation connects across the course

Expected Value

Linearity of expectation is built on expected value itself. You use it when a question asks for the average outcome of a sum, such as total points, total winnings, or total successes. If you can find the expected value of each part, the linearity rule lets you combine them without rebuilding the entire distribution from scratch.

Variance

Expectation adds cleanly, but variance does not behave the same way. A lot of students mix up these two rules and assume totals work the same way for both. In Intro to Probability, this contrast is useful because it shows why mean calculations are often simple while spread calculations usually need more structure.

Covariance

Covariance becomes relevant when random variables are dependent. Linearity of expectation does not care about dependence, but variance does, and covariance is part of that story. If a problem asks about the spread of a sum, covariance helps explain why the answer may change when variables move together.

Law of the Unconscious Statistician (LOTUS)

LOTUS is another tool for finding expected values, especially when a random variable is transformed. Linearity is different because it is about sums of random variables, not just a single transformation. Together, the two ideas cover a lot of expected value problems in probability homework.

Is Linearity of Expectation on the Intro to Probability exam?

A quiz or problem-set question will usually ask you to find an expected total, not the full distribution. Your move is to break the total into pieces, identify the expected value of each piece, and add them. If the problem uses indicator variables, you turn each event into a 0/1 random variable and sum their expectations.

A common trap is thinking you need independence before you can add expectations. You do not. Independence matters for many other probability steps, but linearity of expectation still works for dependent variables, which makes it especially useful in counting and game problems.

You may also be asked to explain why a shortcut is valid. In that case, say that expectation is linear and therefore distributes over addition and constant multiples. If the question asks about variance instead, do not reuse this rule blindly, because variance of a sum is a different calculation.

Linearity of Expectation vs Variance of a Sum

Linearity of expectation tells you how averages combine: the expected value of a sum is the sum of the expected values. Variance of a sum is about spread, not average, and it does not simplify the same way unless you know more about the variables. This is one of the most common mix-ups in Intro to Probability.

Key things to remember about Linearity of Expectation

• Linearity of expectation means you can add expected values when you add random variables.

• You do not need independence for the rule to work, which is what makes it so useful.

• This idea is often the fastest way to find the expected total in counting and game problems.

• The rule works with negative values and constant multiples too, not just positive counts.

• Do not confuse expectation with variance, because variance does not add the same way.

Frequently asked questions about Linearity of Expectation

What is linearity of expectation in Intro to Probability?

It is the rule that the expected value of a sum equals the sum of the expected values. In notation, E[X + Y] = E[X] + E[Y]. In Intro to Probability, that lets you find expected totals without working out every possible combined outcome.

Do the random variables have to be independent?

No. That is one of the main reasons this rule is so useful. Even if the variables are dependent, the expected value of their sum is still the sum of their expected values.

How do you use linearity of expectation on a counting problem?

Break the total into smaller random variables, often indicator variables that equal 1 when something happens and 0 when it does not. Find the expected value of each piece, then add them. This is usually much easier than finding the whole distribution first.

Is linearity of expectation the same as adding variances?

No. Expectation adds cleanly, but variance does not behave the same way. If a problem asks about spread instead of average, you need the variance rules, and dependence can change the answer.