What is expected value in sports betting
By George Boyle · Updated 2026-09-16 · The Sport Stack
Expected value (EV) is the average amount a bet wins or loses if you could repeat it thousands of times. It is your estimated win probability multiplied by the profit, minus your estimated loss probability multiplied by the stake. A bet is +EV when your probability estimate is higher than the probability the price implies — and over a long run, only +EV bets compound.
The calculation, with an actual number
EV = (probability of winning × profit if it wins) − (probability of losing × amount staked). Take a bet at +150 that you believe wins 45% of the time. A one-unit stake returns 1.5 units of profit, so the expected value is (0.45 × 1.5) − (0.55 × 1) = 0.675 − 0.55 = +0.125 units. That bet is worth an average of 0.125 units every time you can make it.
The same bet at a 38% win probability is (0.38 × 1.5) − (0.62 × 1) = −0.05 units: negative, and no amount of confidence changes that. The entire question is whether your probability estimate is better than the market’s, because the price is the market’s estimate.
Why the price is the hard part, not the pick
Most bettors focus on picking winners, but a winner at the wrong price loses money over time and a loser at the right price makes it. A 70%-likely outcome at -300 is roughly break-even before vig; the same outcome at -200 is strongly +EV. Nothing about the team changed between those two prices.
This is why serious bettors talk about numbers rather than opinions, and why the honest version of "I like this team" is "I think this team wins more often than 61% of the time, and I can get 61% implied at this book."
A +EV bet usually still loses
At +150 with a 45% true probability, you lose 55% of the time. That is the normal, expected experience of a good bet, and it is why results over any short stretch tell you almost nothing about whether your process is sound. A hundred +EV bets can easily show a loss; that is variance, not evidence.
The practical consequence is that you cannot evaluate a betting process on its profit and loss alone until the sample is large. Measuring whether you consistently beat the closing line is a faster and much harder-to-fake signal, because it tests the pricing judgement directly rather than waiting for outcomes to average out.
Common questions
How do you convert American odds to an implied probability?
For a negative price, implied probability = odds ÷ (odds + 100), using the absolute value: -150 gives 150 ÷ 250 = 60%. For a positive price, implied probability = 100 ÷ (odds + 100): +150 gives 100 ÷ 250 = 40%. These raw figures include the sportsbook’s margin, so they sum to more than 100% across a market and must be devigged before they can be compared to a real probability estimate.
Is a +EV bet guaranteed to be profitable?
No. Expected value is an average over many repetitions, and a +EV bet can lose the majority of the time — a bet at +150 with a 45% win probability is +EV and loses 55% of the time. It is also only as good as the probability estimate behind it: if your estimate is wrong, the calculated EV is wrong too, and confidence in the estimate is not evidence for it.
What is the difference between EV and edge?
Edge usually means the gap between your estimated probability and the probability the price implies, expressed in percentage points. Expected value converts that gap into an amount of money at the specific odds offered. The same edge is worth more EV on a longshot than on a heavy favourite, which is why edge alone does not tell you how much a bet is worth.
Written by George Boyle, who builds The Sport Stack — the models, the public ledger and these explainers. Corrections and questions: hello@thesportstack.io. Who runs this.