Expected Value Calculator Guide: Formula, Examples & Real Decisions

What Is Expected Value?

Expected value is the probability-weighted average outcome of a decision made under uncertainty. It answers one question: if I made this choice a thousand times, what would my average result be?

The formula is straightforward:

EV = Σ (Outcome × Probability)

You multiply each possible outcome by its probability of occurring, then sum all those products. The result is the “expected” payoff per trial — not a guaranteed result on any single trial, but the average you'd converge toward over many repetitions.

A Simple Example: The Coin Flip

Flip a fair coin. You win $1 on heads, $0 on tails. What's the expected value?

EV = (0.5 × $1) + (0.5 × $0) = $0.50

You'll never actually receive $0.50 on a single flip — you get $1 or $0. But if you played 1,000 times, you'd average very close to $0.50 per flip. That's the expected value at work.

Expected Value in Gambling: Why the House Always Wins

Casinos are in the business of offering games with negative expected value for the player and positive expected value for the house. Understanding the math makes clear why this is unavoidable for most players.

American Roulette

An American roulette wheel has 38 pockets: numbers 1–36, plus 0 and 00. A $1 straight-up bet on one number pays 35-to-1 if it hits.

EV = (1/38 × $35) + (37/38 × −$1) = $0.921 − $0.974 = −$0.0526

The house edge on American roulette is 5.26%. Every $1 you bet costs you an average of $0.0526. Bet $100 per spin for three hours at 40 spins per hour and your expected loss is $631. The actual result will vary, but that's the math (source: American Gaming Association, 2024).

Slot Machines

Slot machine house edges typically range from 5% to 15%, with an average closer to 8% in most U.S. casinos. That translates to an expected loss of roughly $0.08 per $1 wagered. Speed amplifies this: slots are designed to run 400–600 spins per hour, meaning your expected hourly loss on a $1-per-spin machine is $32–$48 (source: UNLV Center for Gaming Research, 2023).

The Lottery: Extreme Negative EV

The Powerball jackpot can reach hundreds of millions of dollars. But the expected value of a $2 ticket is roughly −$1.50, even when the jackpot is large. Why?

• The probability of winning the jackpot is approximately 1 in 292 million.
• Federal taxes take roughly 37% of large winnings; state taxes add more.
• The lump sum is typically ~60% of the advertised jackpot.
• Smaller prizes add some positive EV back, but not enough to overcome the rake.

At a $300 million advertised jackpot, the after-tax lump sum is roughly $113 million. EV from the jackpot prize alone: (1/292,201,338) × $113,000,000 ≈ $0.39. Add in expected value from smaller prizes (~$0.11) and your total EV is roughly $0.50 on a $2 ticket — an expected loss of about $1.50 per ticket. Even jackpots above $500 million rarely push EV positive once taxes are factored in (source: Lottery USA, 2025 analysis).

Expected Value in Business Decisions

EV thinking is most valuable when you have multiple paths with different probabilities of success. Rather than going on gut feeling, you can compare options numerically.

Product Launch Decision

You're deciding between two product launches:

• Product A: 60% chance of $100K profit, 40% chance of breaking even ($0).
• Product B: 30% chance of $250K profit, 70% chance of breaking even ($0).

EV(A) = (0.60 × $100,000) + (0.40 × $0) = $60,000
EV(B) = (0.30 × $250,000) + (0.70 × $0) = $75,000

Product B has a higher expected value ($75K vs $60K) despite a lower probability of success. A pure EV maximizer chooses B. But context matters: if your company's survival depends on generating at least some revenue, the more reliable Product A may be strategically preferable — EV is one input, not the whole picture.

Portfolio Management

Portfolio expected return is the weighted sum of each asset's expected return multiplied by its weight in the portfolio. If a portfolio holds 60% stocks (expected return 7%) and 40% bonds (expected return 3%):

EV(portfolio) = (0.60 × 7%) + (0.40 × 3%) = 4.2% + 1.2% = 5.4%

According to MIT OpenCourseWare probability and statistics materials, this framework — probability-weighted outcomes — is foundational to modern portfolio theory and underlies tools like the Sharpe ratio and Value-at-Risk models used by institutional investors.

When Negative EV Is the Rational Choice

Insurance is the clearest example. The expected payout of an insurance policy is always less than the premium — otherwise insurers couldn't stay in business. Yet buying insurance is often completely rational.

If your house is worth $400,000 and burns down, that's a catastrophic, possibly unrecoverable loss. Paying $2,000/year in homeowner's insurance is negative EV in strict dollar terms — but the expected value calculation misses something: the utilityof avoiding financial ruin. You're not trying to maximize expected dollars; you're trying to maximize expected wellbeing.

Kahneman and Tversky's Prospect Theory (1979) formalized this asymmetry. Their research showed that the pain of losing $100 is psychologically about twice as powerful as the pleasure of gaining $100 — a phenomenon called loss aversion. This is why rational people routinely accept negative EV to hedge against large losses. The American Statistical Association notes that utility-based decision models, rather than pure EV maximization, better predict human behavior in risk scenarios.

Medical Decision Making

EV frameworks also inform clinical decisions. A treatment with a 70% chance of full recovery, a 20% chance of partial recovery, and a 10% chance of no improvement can be compared against an alternative using quality-adjusted life years (QALYs) as outcomes. Physicians and health economists use this structure to evaluate surgical interventions, chemotherapy regimens, and screening programs — always weighing expected outcomes against costs and risks.

Variance, Risk, and Why High EV Isn't Everything

Two bets can have the same expected value but wildly different variance:

Same EV, completely different risk profile. If you can't afford to lose $1,000, Bet 2 is irrational regardless of EV. This is why standard deviation and risk tolerance are inseparable from EV analysis in practice. Richard Thaler's work in behavioral finance showed that people systematically misevaluate variance — taking on too much risk in some domains and too little in others — often because they frame individual bets rather than thinking about long-run expected outcomes.

The Law of Large Numbers: Why Casinos Always Win

A casino might lose to a lucky player on any given night. But the Law of Large Numbers guarantees that as the number of bets increases, the actual average result converges toward the expected value. With hundreds of thousands of bets placed each day across a casino floor, the house's 5.26% edge on roulette will produce results extremely close to 5.26% of total money wagered — as predictably as a manufacturing process.

This is also why individual variance doesn't protect you long-term. A lucky run at the slots doesn't change the math. Play long enough and your results will converge toward the negative EV of the game.

5 Real-World Decisions Where Expected Value Thinking Helps

1. Job offer comparison: Weight salary, equity, bonus probability, and career growth probability to compare total expected compensation across offers.
2. Startup investment: Estimate exit probabilities across failure, modest acquisition, and breakout success to gauge expected return on an angel investment.
3. Insurance coverage level: Compare expected claim payouts at different deductibles against premium differences to find the EV-optimal deductible.
4. Marketing channel allocation: Assign conversion rates and revenue per conversion to each channel to identify where a dollar of spend has highest expected return.
5. Legal settlement decisions: Estimate trial win probability, potential award, legal costs, and settlement offer to determine whether settling is EV-positive relative to litigation.

Frequently Asked Questions

What is expected value?

Expected value (EV) is the probability-weighted average of all possible outcomes of a decision. It is calculated as the sum of each outcome multiplied by its probability: EV = Σ (Outcome × Probability). A positive EV means the average result is a gain; a negative EV means the average result is a loss over many repetitions.

How do you calculate expected value?

Multiply each possible outcome by its probability of occurring, then sum all those products. For a coin flip where you win $1 on heads and $0 on tails: EV = (0.5 × $1) + (0.5 × $0) = $0.50. For more complex decisions, list every scenario, assign a dollar value and a probability to each, and add them up.

Is gambling ever positive expected value?

Rarely, and only in specific situations. Card counting in blackjack can shift the house edge negative (positive EV for the player) when the deck is rich in high cards. Certain promotional casino bonuses or matched-betting opportunities on sports can also be positive EV. But standard casino games — roulette, slots, lottery — are always negative EV for the player. The house edge ensures the casino wins over time.

Why do people make decisions with negative expected value?

Kahneman and Tversky's Prospect Theory (1979) showed that people weight losses more heavily than equivalent gains (loss aversion) and overestimate the probability of rare events. This explains why people buy lottery tickets (overweighting the tiny chance of a jackpot) and why they pay for insurance even when the expected payout is less than the premium. Behavioral finance researchers like Richard Thaler further documented how cognitive biases routinely cause decisions that deviate from pure EV maximization.

How is expected value used in business?

Businesses use EV to compare projects, prioritize investments, and set prices. A product launch with a 60% chance of $100K profit has an EV of $60K. A riskier launch with a 30% chance of $250K profit has an EV of $75K — higher EV despite lower probability of success. EV analysis also underlies pricing insurance policies, structuring loan portfolios, and evaluating R&D investments where outcomes are uncertain.