Independent guide · Betting

How to calculate expected value (EV) in betting — and why EV is not a guaranteed profit

A practical guide to computing expected value (EV) for bets, reading results correctly, and understanding the difference between mathematical expectation and guaranteed profit.

Editorial illustration for: How to calculate expected value (EV) in betting — and why EV is not a guaranteed profit

Why this matters

Expected value is the long‑run average payoff given correct probabilities. A positive EV improves average results across many independent bets but does not guarantee profit on any single wager or over short samples.

What EV (mathematical expectation) means

Expected value (EV) is the long‑run average payoff of a random process: for each possible outcome multiply the payoff by its probability and sum the results. In betting this commonly appears as a per‑unit stake expression using decimal odds and your assessed win probability.

EV describes what happens across many independent repetitions (the law of large numbers). It is a probabilistic statement — it does not promise a profit on any single bet or over any particular finite sample.

  • Definition: EV = Σ(probability × payoff) across outcomes.
  • Per‑unit betting formula (decimal odds d): EV = p × d − 1.

How to calculate EV for a single bet

Use decimal odds for clarity. If the bookmaker offers decimal odds d and your assessed probability of winning is p (both expressed as decimals), compute EV = p × d − 1. That result is the expected net return per unit staked.

A positive EV (>0) indicates a favourable long‑run edge; a negative EV (<0) indicates an expected loss per unit stake. Convert fractional or American odds to decimal odds before applying the formula.

  • Example inputs: stake = $1, d = 3.00, p = 0.40.
  • Example compute: EV = 0.40×3.00 − 1 = +0.20 → +20% EV per $1 stake.

Why a positive EV is not the same as guaranteed profit

Mathematical expectation is asymptotic: the average converges to EV as the number of independent trials grows. Any finite sequence can show large deviations from EV because of variance — you can lose money despite a positive EV during typical, realistic sample sizes.

Guaranteed profit implies riskless arbitrage: zero variance and no exposure. True arbitrage opportunities are rare, short‑lived, and often consumed by transaction costs, limits or corrections. Most +EV situations are probabilistic edges, not riskless guarantees.

  • EV is an average — not a promise for a single bet.
  • Riskless guaranteed profit requires arbitrage and absence of costs/limits — uncommon in normal markets.

Limitations, risks and responsible practices

EV depends entirely on the accuracy of your assessed probability p. Model error, biased samples, or poor data make EV calculations unreliable. Always disclose the assumptions behind p (model, data window, devigging).

Manage bankroll and size stakes relative to risk. Use risk‑management frameworks (e.g., Kelly) only with careful estimation; keep expectations realistic and provide links to regulator guidance and help resources if readers have concerns about gambling harms.

  • Key limitations: estimation error, market margins (vig), transaction costs, betting limits and variance.
  • Practical controls: stake sizing, diversification, record keeping, and clear disclosure of assumptions.

Practical checklist

  • State the stake basis (per $1 or your bet size) and use decimal odds or convert other formats.
  • Show the probability p you used and explain how you estimated it (data/model/source).
  • Adjust for bookmaker margin if you want a devigged comparison; include fees and transaction costs.
  • Use bankroll rules and limit exposure; do not treat +EV as immediate guaranteed profit.

Sources and further reading

Fact-checked: . External sources open in a new tab.

  1. Expected value (basic)Khan Academy
  2. How can gaming machines meet their %RTP if they are random? (feature article)UK Gambling Commission
  3. House edge, hold and return to playerSacStat
  4. Probabilities and payoffs (insights)Morgan Stanley
← More guides