What is expected value (EV) in betting?

Expected value (EV) is the average monetary outcome of a bet if that exact wager were placed an infinite number of times under identical conditions. Calculated as (P_win × profit) − (P_lose × stake), EV tells you whether a bet makes mathematical sense before a single result is known. It is the single most important number in serious sports betting because it separates disciplined, probability-based decisions from gut-feel guessing.

The Formula in Plain English

EV reduces to one question: does the price the bookmaker offers reflect a probability lower than your own estimate? If your model says a team wins 55% of the time but the odds imply only 50%, you have found positive expected value (+EV). Negative expected value (−EV) means the bookmaker's implied probability exceeds yours — you are overpaying for the outcome.

The formula written out:

EV = (P_win × net profit) − (P_lose × stake)

Where P_lose = 1 − P_win for a two-outcome market.

Worked Numeric Example

A bookmaker offers +120 (decimal 2.20) on Team A to win. You risk €100.

• Net profit if correct: €120
• Loss if wrong: €100
• Your model estimates Team A's true win probability: 52%

EV = (0.52 × €120) − (0.48 × €100) EV = €62.40 − €48.00 = +€14.40

Every €100 wagered on this line generates an expected €14.40 in profit over a large sample. The bet is +EV. If the bookmaker had priced Team A at −130 instead, the implied probability would be ~56.5%, making the same 52% estimate a −EV position.

Where the Edge Actually Comes From

An edge requires a probability estimate sharper than what the odds imply. Three practical sources:

• Line-shopping: Different bookmakers post different prices. Comparing five books for the same event routinely surfaces 3–8% price differences.
• Sharper models: Public-facing odds are anchored partly to betting volume, not pure probability. Team-specific injury data, travel schedules, and weather inputs can move your estimate meaningfully.
• Prediction-market reference prices: Event contract markets — where participants trade positions on outcomes — often carry lower juice and reflect aggregated information quickly. They serve as a useful calibration benchmark for implied probabilities.

EV Is Not a Guarantee — Variance Matters

A +EV bet loses more than half the time if P_win is below 50%. Over 20 bets, a genuine +5% edge can still produce a losing run of 8 consecutive results. This is variance, not model failure. Bankroll management — typically flat-betting 1–3% of total bankroll per wager, or Kelly sizing — keeps a drawdown survivable long enough for EV to express itself statistically. Treating every single result as a referendum on your model is a fast route to abandoning a sound process.

How to Apply EV Before Every Bet

Use the WeeBet EV calculator to input your estimated probability and the available odds. The tool returns EV in both absolute (€) and percentage terms, flags whether the bet is +EV or −EV, and shows the breakeven probability — the implied probability embedded in the price. Build the habit of running this check before confirming any wager.

Responsible gambling note: Even systematic +EV betting carries real financial risk. Set deposit limits, never bet more than you can afford to lose, and use the responsible gambling tools available on your platform.

---

Frequently Asked Questions

Can a +EV bet still lose money over a whole season?

Yes. A genuine edge of +4% per bet across 200 wagers still produces a losing season roughly 10–15% of the time purely due to variance. Sample size and bankroll discipline determine whether the edge materialises in practice.

What is a good EV percentage for sports betting?

Professional bettors typically target +3% to +8% EV per bet. Anything above +10% on a recurring basis usually signals a mispriced market or a data error worth double-checking rather than celebrating uncritically.

Is EV the same as return on investment (ROI)?

Not exactly. EV is a pre-bet theoretical estimate based on your probability model; ROI is a post-bet realised figure across your actual results. Over a large enough sample, your realised ROI should converge toward your average EV if your probability estimates are accurate.

Does EV apply to casino games as well as sports betting?

Yes. Blackjack with basic strategy carries an EV of approximately −0.5%, while a European roulette straight-up bet sits at −2.7%. In casino games the probabilities are fixed, so EV is precise rather than estimated — and is almost always negative for the player.