Expected value:
chance of win × win + chance of loss × loss
Weigh a two-outcome choice with the expected value formula, find the break-even chance, and see why one positive result does not guarantee a good decision.
Calcylator Editorial Team
Updated · 5 min read
The expected value formula for a win or a loss
For a choice with two outcomes, expected value equals the chance of winning times the win payoff, plus the chance of losing times the loss payoff. With a 30 percent chance of winning 500 and a 70 percent chance of losing 100, the expected value is 0.3 × 500 + 0.7 × (−100) = 80.
- p:
- probability of winning, as a fraction (30% = 0.3)
- win:
- payoff if you win
- loss:
- payoff if you lose; enter it as a negative number for a loss
Win probability
30 %
Win payoff
500
Loss payoff
-100
Calculation
0.30 × 500 + 0.70 × (−100)
Expected value
80
Expected value is a long-run average: you never actually get 80 in a single trial.
Finding the break-even probability
The break-even probability is the win chance at which the expected value is exactly zero. Rearranging the formula gives p = −loss ÷ (win − loss). Take a business bid that earns 300 if won and loses 150 on proposal costs if not. Break-even is 150 ÷ (300 + 150) = 33.3 percent.
| Win chance | Expected value |
|---|---|
| 20% | -60 |
| 33.3% | 0 |
| 40% | 30 |
| 50% | 75 |
| 60% | 120 |
If you believe your real chance is above 33.3 percent the bid has a positive expected value, and if it is below, you are likely to lose money over many bids.
A ticket that looks cheap
A ₹100 ticket gives a 1 percent chance at a ₹5,000 prize. Your net result is a gain of ₹4,900 if you win and a loss of ₹100 if you do not. EV = 0.01 × 4,900 + 0.99 × (−100) = 49 − 99 = −50, so on average each ticket costs you ₹50.
Another way to see it: the prize is worth 0.01 × 5,000 = ₹50 in expectation, and you paid ₹100 for it. Always include the price of the ticket as part of the loss.
The same method helps a small business owner. A quote that wins 40 percent of the time with a profit of 300 and loses 150 in preparation costs otherwise has EV = 0.4 × 300 + 0.6 × (−150) = 30. It is positive, but thin, so a small rise in preparation cost could wipe it out.
What expected value does not tell you
- It ignores risk. A positive EV can still ruin you if you can only afford one try.
- It assumes the probability is right. A rough guess gives a rough answer.
- It treats money as linear, but losing 10,000 can hurt more than winning 10,000 helps.
- It describes the average over many repeats, not the next single outcome.
This is a general maths tool, not financial or investment advice. Real decisions have more than two outcomes and often cannot be repeated.
Common mistakes
- Entering the loss as a positive number, which turns a cost into a gain.
- Typing 0.3 into a field that expects a percentage; use 30.
- Forgetting that the loss probability is 100 minus the win probability.
- Leaving out the cost of playing when the payoff is only the prize.
Common questions
What is the expected value formula?
For two outcomes, EV = p × win + (1 − p) × loss, where p is the probability of winning. For a 30 percent chance of winning 500 and a 70 percent chance of losing 100, the expected value is 80.
How do I calculate expected value?
Multiply each outcome by its probability and add the results. With two outcomes, take the win chance times the win payoff, then add the loss chance times the loss payoff. Enter the loss as a negative number so it reduces the total.
What does a negative expected value mean?
It means that, on average, you lose money or value each time you repeat the choice. A ₹100 ticket with a 1 percent chance at ₹5,000 has an expected value of −₹50. Single outcomes can still be wins, but the long-run average is negative.
How do I find the break-even probability?
Use p = −loss ÷ (win − loss). If winning pays 300 and losing costs 150, break-even is 150 ÷ 450 = 33.3 percent. Above that win chance the expected value is positive; below it the expected value is negative.
Is expected value the same as the most likely result?
No. Expected value is a probability-weighted average, so it may be a figure that never actually occurs. In the 30 percent example above you either win 500 or lose 100, yet the expected value is 80, a result you can never get in one try.
Was this guide helpful?
Continue reading
View all blogsNumber needed to treat: formula and worked example
Number needed to treat converts a risk difference into how many people need an intervention to prevent one event. Formula, rounding rule, example.
4 min read
Absolute risk reduction: formula and worked example
Absolute risk reduction shows how many percentage points a treatment cuts from the control group's event rate. Formula, 4-point example, pitfalls.
4 min read
Odds ratio from contingency table: formula and example
The odds ratio from a 2x2 contingency table compares odds of an outcome between two groups. See the cross-product formula, a 1.71 example and pitfalls.
5 min read




