Expected Value (EV) in Sports Betting: Definition and Calculation

Updated on 2026-07-27 · 874 words

What it is

Expected value (EV) is a mathematical concept that measures the average amount you can expect to win or lose per bet over the long run. In sports betting, it’s calculated by comparing the true probability of an outcome (your estimate) with the odds offered by the sportsbook. A positive EV (+EV) means the bet has a statistical edge; a negative EV (-EV) means it does not. However, EV is not a prediction for any single event—it’s a long-term average that only converges over a large sample of bets. For example, a bet with +10% EV will not guarantee profit on the next play, but over 1,000 similar bets, you’d expect to be ahead by roughly 10% of total stakes.

In the US market, odds are typically expressed in American format (e.g., +150, -110). The calculation remains the same once converted to decimal. This page shows you how to compute EV using both formats.

The math

The formula for expected value per unit bet is:

EV = (Decimal Odds × Probability) − 1

If you prefer American odds, convert them to decimal first: for positive odds (+X), decimal = (X/100) + 1; for negative odds (-Y), decimal = (100/Y) + 1.

Let’s use a fixed decimal odd of 2.50 (American +150) and vary the probability to illustrate how EV changes with your estimate.

Positive EV example:
Suppose you estimate the outcome has a 48% chance (p = 0.48).
EV = (2.50 × 0.48) − 1 = 1.20 − 1 = +0.20 per dollar bet.
That means an expected profit of $0.20 for every $1 wagered. On a $20 bet, the expected value is $20 × 0.20 = $4.00.

Negative EV example:
Now suppose the same odds but your estimate drops to 35% (p = 0.35).
EV = (2.50 × 0.35) − 1 = 0.875 − 1 = −0.125 per dollar.
Expected loss of $0.125 per dollar, so on $20 that’s −$2.50.

Notice: the same odd of 2.50 can produce either positive or negative EV depending on your probability assessment. The odd alone does not determine value—your estimate does.

In practice, most sportsbook bets have a negative EV because the implied probability (after removing the vigorish) is higher than the true probability for the average bettor. Only when your estimate is more accurate than the market’s does EV become positive.

Worked example

Consider a real scenario: You are betting on an NBA game. The sportsbook offers the away team at +150 (decimal 2.50) on the moneyline. You have done your research and believe the away team has a 46% chance of winning.

  • Step 1: Write your estimated probability: p = 0.46.
  • Step 2: Use the formula: EV = (2.50 × 0.46) − 1 = 1.15 − 1 = 0.15.
  • Step 3: Interpret: For every dollar staked, you expect to win $0.15 in the long run. On a $20 bet, that’s $3.00 expected profit.

Now, suppose a friend thinks the chance is only 30%. For that friend, EV = (2.50 × 0.30) − 1 = 0.75 − 1 = −0.25 per dollar, a $5.00 expected loss on a $20 bet. The same odd yields opposite evaluations because of different probability estimates.

This example underscores that EV is a function of your personal model. Without a reliable probability estimate, the calculation is meaningless.

When not to use it

  • EV is a long-term average, not a short-term guarantee. A single bet with +EV can lose, and a string of +EV bets can still produce a losing streak. The EV only manifests over hundreds or thousands of independent bets. Expecting a quick profit from a few +EV wagers misunderstands the concept.
  • Your probability estimate is the weakest link. Overconfidence or poor data leads to systematically biased EV calculations. If your model overestimates win rates by 5%, you will see fake +EV everywhere. Always test your own predictions against actual outcomes over a large sample before relying on EV.
  • Sportsbooks limit winning accounts. Even if you have a true edge, licensed US bookmakers can reduce your stakes, limit the markets you can bet, or close your account entirely. The mathematical advantage does not guarantee access to unlimited betting volume.
  • Market efficiency varies across leagues and bet types. EV models that work in high-liquidity markets (e.g., NFL spreads) may fail in lower-liquidity ones (e.g., Korean baseball) where odds move slowly and contain more noise. Reusing a model across different sports without recalibration introduces false positives.
  • EV does not account for opportunity cost or bankroll management. Betting a +5% EV wager may be inferior to a +10% EV wager on the same card. Moreover, betting too large a fraction of your bankroll on a single +EV bet increases the risk of ruin—EV alone does not inform staking size.

For a quick EV calculation, use our bet analyzer tool.


Responsible gaming: Sports betting involves financial risk and is not a guaranteed way to make money. The long-term expected value for players is negative due to the built-in margin (vigorish). This content is for educational purposes only. You must be 21+ to bet in the United States. If you or someone you know has a gambling problem, call 1-800-GAMBLER.

Frequently asked questions

What does a positive EV mean in sports betting?
A positive EV means that, over a large number of similar bets, you would expect to make a profit. It does not guarantee the next bet will win, but indicates a statistical edge relative to your probability model.
How do I calculate EV with American odds?
First convert American odds to decimal: for positive odds (e.g., +150) decimal = (150/100)+1 = 2.50; for negative odds (e.g., -110) decimal = (100/110)+1 ≈ 1.909. Then apply EV = (decimal * probability) - 1.
Why is most sports betting negative EV?
Because sportsbooks add a margin (vigorish) to the odds, making the implied probability sum greater than 100%. The average bettor lacks an edge, so the EV is negative in the aggregate.
How many bets are needed to realize my EV?
Statistically, you need hundreds to thousands of independent bets for the actual results to converge to the EV. The more volatile the sport, the larger the sample needed.