Understanding Expected Value in Investments: Definition & Formula

Search Fintrev

Search our guides, reviews and glossary.

Expected value (EV) is the probability-weighted average of all possible outcomes of an uncertain event. In investing, it is the return or profit you would earn on average if you could make the same bet many times: each possible result is multiplied by its probability, and the products are added together. EV does not predict what will happen on any single investment. It shows whether the odds and payoffs, as you have estimated them, lean in your favor.

How expected value works

In probability and statistics, expected value describes the center of a random variable’s distribution. It is also called the expectation, the mean of the distribution, or the first moment, and it is written E(X).

EV is meaningful because of the law of large numbers: as an experiment is repeated more and more times, the average of the observed results converges toward the expected value. A single coin flip lands heads or tails, but the share of heads over thousands of flips settles near one half. The same logic applies to an investment process repeated across many positions or many years.

The expected value formula

For a discrete random variable:

EV = Σ P(Xᵢ) × Xᵢ

where:

• X is the random variable (for example, the dollar gain on a trade)
• Xᵢ is each specific value X can take
• P(Xᵢ) is the probability of that value occurring

The probabilities across all outcomes must add up to 1 (100%). The sum applies to discrete variables, which have a countable set of outcomes such as forecast scenarios. For a continuous variable (a full range of possible returns) with probability density function f(x), the equivalent is the integral E(X) = ∫ x · f(x) dx over all possible values of x.

EV comes out in the same units as the outcomes: dollar outcomes give an EV in dollars, and percentage returns give an expected return in percent. Every outcome in the calculation must use the same unit.

The classic die example

A fair six-sided die shows 1, 2, 3, 4, 5 or 6, each with probability 1/6:

(1/6 × 1) + (1/6 × 2) + (1/6 × 3) + (1/6 × 4) + (1/6 × 5) + (1/6 × 6) = 3.5

No roll can produce 3.5, yet the average of a very large number of rolls approaches it.

How to calculate expected value for an investment

1. List the possible outcomes. Define each scenario and the gain or loss it produces.
2. Assign a probability to each. Make sure they sum to 100%.
3. Multiply each outcome by its probability. Losses enter as negative numbers.
4. Add the products. The total is the expected value.

Worked example: expected profit (hypothetical)

Suppose an investment has a 60% chance of gaining $10,000 and a 40% chance of losing $5,000.

• Gain scenario: 0.60 × $10,000 = $6,000
• Loss scenario: 0.40 × (−$5,000) = −$2,000
• EV = $6,000 − $2,000 = $4,000

This is the expected profit formula in its simplest two-outcome form:

Expected profit = (P(win) × average gain) − (P(loss) × average loss)

Traders apply the same equation to a strategy: a win rate of 40% with an average win of $300 and an average loss of $150 gives (0.40 × $300) − (0.60 × $150) = $120 − $90 = $30 per trade, before commissions and slippage. A strategy can lose more often than it wins and still have a positive EV if its winners are large enough relative to its losers.

Worked example: is waiting for the dip worth it? (hypothetical)

EV can frame a timing decision. Assume a stock trades at $100 and you expect it to reach $110 by your target date. You can buy now, or place a limit order at $95 and wait. You estimate a 40% chance the price dips to $95 before rising to $110, and a 60% chance it rises without dipping, in which case your order never fills.

• Buy now: gain of $10 per share in both scenarios, so EV = $10.
• Wait: (0.40 × $15) + (0.60 × $0) = $6.

Under these assumptions, buying now has the higher expected value. Waiting wins only if you believe the dip is more likely: at 67% or above, 0.67 × $15 ≈ $10 matches buying now. The answer depends entirely on the probabilities you assign.

How investors use expected value

Scenario analysis. Analysts often build bull, base and bear cases for a company or project, attach a probability to each, and compute a probability-weighted value. The technique also shows whether the downside scenarios carry more risk than the reward justifies.

Comparing assets. A stock, a bond and an exchange-traded fund each have different expected returns and risk profiles. Calculating EV for each, with consistent units and the same time horizon, puts them on a common footing. After a portfolio is built, the same comparison can support switching from a holding whose EV has fallen to a replacement with a higher EV for similar risk.

Portfolio construction. Modern portfolio theory uses mean-variance optimization, where the “mean” is the expected return and risk is measured by the standard deviation of returns. A portfolio’s expected return is the weighted average of its holdings’ expected returns:

E(Rₚ) = Σ wᵢ × E(Rᵢ)

For example, a hypothetical portfolio of 60% in an asset expected to return 8% and 40% in one expected to return 4% has an expected return of (0.60 × 8%) + (0.40 × 4%) = 6.4%. Unlike expected return, portfolio risk is not a simple weighted average; it depends on how the assets move relative to each other, which is why diversification can lower risk without lowering expected return proportionally.

Expected value vs. related concepts

Expected value vs. arithmetic mean. The arithmetic mean averages outcomes that have already happened (a distribution of occurrence). Expected value averages possible outcomes weighted by their probabilities (a distribution of probability). With enough observations of a stable process, the sample mean approaches the EV, but one is a historical statistic and the other is a forward-looking estimate.

Expected value vs. most likely outcome. The most likely outcome is the mode, the single result with the highest probability. In the 60/40 example above, the most likely result is a $10,000 gain, while the EV is $4,000. Accounting estimates sometimes distinguish between an “expected value” method and a “most likely amount” method for the same reason: they answer different questions.

Expected value vs. expected cost. Expected cost is EV applied to losses or expenses. If there is a 1% chance of a $20,000 loss, the expected cost is $200. Comparing that with, say, a $300 insurance premium shows the premium exceeds the expected loss, which is normal, since buyers pay to remove the variance rather than to profit.

Statistical EV vs. a stock’s “expected value.” In equity valuation, a stock’s expected or intrinsic value usually means an estimate of what the shares are worth. Examples are the present value of future dividends, or a peer multiple applied to its earnings per share using the P/E ratio. That is a valuation method, not a probability-weighted average, although an analyst can blend several valuation scenarios using EV.

Limitations of expected value

• Inputs are estimates. Probabilities for investment outcomes are rarely known; they are judgments. Small changes to a probability or payoff can flip the sign of the EV.
• It ignores dispersion. Two investments with the same EV can have very different ranges of outcomes. EV must be paired with a risk measure such as standard deviation or maximum drawdown.
• The long run may never arrive. EV assumes many repetitions. An investor who makes one concentrated bet, or who risks losses large enough to end their ability to keep investing, does not get to average out bad results. A positive-EV bet sized too large can still lead to ruin.
• Rare extreme events are hard to price. Outcomes with tiny probabilities but huge losses are easy to underweight or omit, and they can dominate real results.
• No time value or preferences. A plain EV calculation does not discount future payoffs or reflect that most people feel losses more than equal gains.

The Bottom Line

Expected value helps investors and traders check whether their estimated odds and payoffs lean in their favor across many decisions. Its results are only as reliable as the probabilities behind them, so the figure should be paired with a risk measure such as standard deviation or maximum drawdown. Position sizing matters too, because a positive-EV bet that is too large can still end in ruin.