Expected Value (EV) is the cornerstone concept of rational betting. It is the average return you can expect per bet over an infinitely large sample, calculated by weighting each possible outcome by its probability. A positive EV (+EV) bet is one where you expect to profit in the long run; a negative EV (-EV) bet is one where you expect to lose.
The formula for a binary bet (win or lose) is: EV = (P(win) × Profit) - (P(lose) × Stake). For a bet with decimal odds of 3.0 where your true probability estimate is 40%: EV = (0.40 × £2) - (0.60 × £1) = £0.80 - £0.60 = +£0.20. This means you expect to earn 20p for every £1 staked, or a 20% return on investment.
Standard bookmaker bets are negative EV because the margin reduces the implied probability below 100%. At a 5% margin, if you placed every possible bet in equal measure you would lose 5% of turnover. Finding positive EV means finding bets where the odds are more generous than the true probability warrants.
EV vs short-term results is a crucial distinction. You can place 20 consecutive +EV bets and lose all of them due to variance. This is not evidence that your EV calculation was wrong — it is expected behaviour in any probabilistic system. EV only reveals itself reliably across hundreds of bets, which is why bankroll management and sample size are so important for evaluating betting performance.
Example
You believe a boxer has a 60% chance of winning a fight. The bookmaker prices them at 2.00 (50% implied probability). EV = (0.60 × £1) - (0.40 × £1) = +£0.20 per £1 staked. If you place 100 bets of this type with the same edge, your expected profit is £20 per £1 unit, regardless of which individual bets win or lose.