Type in the decimal odds, your stake, and your estimate of how often the bet wins. You get back the expected value of the bet in cash, your edge over the bookmaker’s price in probability points, and the fair odds for your estimate. A positive EV means the price pays more than your probability needs to break even.
What Expected Value Measures
Picture placing the same bet a thousand times at the same price. Some settle as wins, more of them as losses, and the running total lands on an average result per bet. Expected value is that average, worked out from three numbers you already have:
EV = (win probability × profit on a win) – (lose probability × stake)
Profit at decimal odds is stake × (odds – 1), so a $100 bet at 2.50 returns $150 profit when it lands. The win probability is your own estimate. The lose probability is whatever remains after it.
A Worked Example
Stake $100 at decimal odds of 2.50, with a win estimate of 45%.
- Profit on a win: $100 × (2.50 – 1) = $150.
- Breakeven probability: 1 ÷ 2.50 = 40%. Below a 40% win rate, this price loses money over time.
- Expected value: (0.45 × $150) – (0.55 × $100) = $67.50 – $55.00 = +$12.50.
- EV as % of stake: +12.5%, or $12.50 per $100 staked on average.
- Edge: 45% – 40% = 5 percentage points.
Fair odds for a 45% chance come out at 1 ÷ 0.45 = 2.222. At 2.50 the bookmaker is paying you above fair value. Drop the same estimate into the calculator at odds of 2.10 and the EV line turns negative: (0.45 × $110) – (0.55 × $100) = -$5.50. Same match, same opinion, different price, opposite decision.
How to Read the Results
Breakeven win probability comes straight from the odds: 1 divided by the decimal price. Set it against your own estimate and the gap between the two is your edge, measured in probability points.
Expected value converts that gap into cash for the stake you entered. A 5-point edge is worth $12.50 on a $100 stake and $62.50 on $500, so the EV figure moves with your staking while the edge stays put.
Fair odds run the calculation in reverse. Given your probability, they show the lowest price worth taking, which is the number to hold in your head when the same market sits at 2.30 with one bookmaker and 2.50 with another.
A +EV Bet Still Loses 55 Times in 100
The worked example is a profitable bet, and it loses more often than it wins. Fifty-five losing Saturdays out of a hundred, with a price good enough to leave you in front across the full run. Bettors who stake as though a positive EV removes the risk tend to find out about the gap between average and outcome the expensive way.
A second caution: treat any EV above roughly +20% with suspicion. Bookmakers rarely misprice a football market by that much, and an optimistic probability estimate explains the number far more often than free money does. Re-check your input before you stake anything on the output.
The Hard Part Is the Probability
The formula takes seconds. Producing a win probability the market has not already priced in takes a model, a database, or years inside one league watching for the things odds compilers miss. Feed the calculator 60% on a bet whose true chance is 45% and it reports value on a loser, because it prices your opinion and nothing else.
Gecko Edge was built around that input problem. The platform generates a win probability for every market from Poisson goal models with a Dixon-Coles correction, blends in market and league priors, then runs the same EV test you see on this page across 130+ leagues. The maths runs first; the AI just puts it in plain English.
Further Reading
- All betting calculators: the full library of 34 free tools
- 7 mistakes you are making with EV betting calculations
- What does EV mean and how should you use it?
- We analysed 8,439 AI-generated +EV bets: what we found
What counts as a good expected value in betting?
Sustainable football betting edges tend to sit between +2% and +8% of stake. Single figures are the realistic hunting ground. A calculation showing +40% points to an input error far more often than a mispriced market, so re-check your probability estimate before acting on a number that size.
Does a positive EV bet guarantee a profit?
No. The worked example on this page carries +$12.50 of expected value and still loses 55% of the time. EV describes the average result across many bets at the same price, and profit arrives over a sample of decisions rather than on any given weekend. Staking discipline counts for as much as bet selection because of that gap.
How do I estimate my win probability?
Build a statistical model, derive probabilities from sharp market prices, or use a modelling platform. Gecko Edge produces one for every market through a Poisson and Dixon-Coles based pipeline before any recommendation appears. Whichever route you take, settle on your estimate before checking the current odds, or the price will anchor your judgement.
What is the difference between EV and edge?
Edge is the gap between your win probability and the breakeven probability in the odds, measured in percentage points. EV prices that gap in cash for a specific stake: a 5-point edge on $100 at decimal odds of 2.50 works out to +$12.50. Double the stake and the EV doubles while the edge stays at 5 points.