Positive Expected Value in Betting and How to Find It

A bet has positive expected value when its true chance of winning is higher than the chance its odds imply, so it returns a profit on average over many repetitions.

Negative expected value is the reverse: the price pays less than the true probability warrants, and the bet loses money over time by definition, according to The Rundown's EV calculator guide.

Expected profit is the dollar figure the calculation produces. The Rundown describes expected value as "the average profit you can expect from a bet over the long run."

The figure decides whether a wager, a live line or a cash-out offer is worth taking. It also bears on which accounts sportsbooks limit, a practice Massachusetts now requires operators to explain.

The formula published by Covers.com is: EV = (probability of winning × potential profit) − (probability of losing × amount staked).

The calculation needs two probabilities. One is implied by the odds. The other is the bettor's estimate of the true chance, and that estimate is the variable on which the result depends.

Implied probability comes from the American odds. For negative odds it is the absolute value of the odds divided by that value plus 100. For positive odds it is 100 divided by the odds plus 100, according to the PropsBot glossary.

By that method, -110 implies 52.38% and +150 implies 40%. A -110 price is therefore a 52.4% break-even rate: a bettor must win that often to avoid losing money, BoydsBets says.

Posted odds include the book's margin. A two-way market priced -110 on both sides sums to 104.8% implied probability, and the 4.8% above 100 is the vig, PropsBot says. Covers puts typical sportsbook vig at about 4.5% to 5%.

Removing the vig gives a no-vig fair probability. The Rundown recommends taking fair probabilities from no-vig odds at sharp, low-margin sportsbooks rather than from a single retail book's posted line.

A coin flip at standard juice. A $10 bet on a true 50% outcome at -110 pays $9.09. The expected value is −$0.45 per bet, or about −$45 over 100 bets, according to BoydsBets. That is negative expected value produced by the vig alone.

A price better than the market. The Rundown's example is a $100 bet on the Rams at +110 when the fair market price is +100, a 50% chance. The expected value is +$5, a 5% edge.

A break-even price. A bettor with a true 40% chance on a +150 line has exactly $0 expected value, BoydsBets says, because the price matches the probability.

A cash-out offer. DraftKings says its cash-out function redeems an unsettled bet "at a value specified by DraftKings at the time the Cash Out is offered," according to its support article. DraftKings does not publish the formula.

BettingTools says cash-out pricing builds in a margin for the book, as standard lines do, and trades expected value for certainty. DraftKings also says a request can be delayed and is rejected if the odds move or the offer is pulled during the delay.

Account limits. Massachusetts requires its seven licensed sportsbooks to notify a bettor within 48 hours of a limit and give a "specific and personalized" explanation, under a rule that took effect June 1, 2026, according to the Massachusetts Gaming Commission.

Operator data shows a direct correlation between above-average win rates and account restrictions, CBS Sports reported. Notices posted by bettors cite DraftKings reasons including "live market latency exploitation" and "structured wagering."

Expected value grades the price at the moment of the bet, not the result. A +EV bet can lose to normal variance, and a winning bet does not prove the decision was +EV, BoydsBets says.

Positive expected value is also not a high win rate. Sharp bettors often win only 53% to 55% of spread bets, or 45% to 50% of plus-money moneylines, and remain profitable because of the payouts, according to BoydsBets.

"A +EV bettor sometimes bets on a team they expect to lose, if that's where the value is," Jimmy Boyd of BoydsBets wrote.

Every price can be converted to an implied probability and compared with a no-vig fair probability before the bet is placed. A bet with no estimate of true probability has no measurable expected value.

The edges involved are small. Covers and BoydsBets both put most realistic +EV edges at 1% to 3%, and BoydsBets says larger edges often indicate a pricing error or a stale line. The Rundown says professional bettors typically target 2% to 5% or more.

A 3% edge yields about $3 of expected profit per $100 wagered over the long run, The Rundown says. Covers says sharp bettors often place 500 or more wagers a month, and more than 1,000 in busy seasons, so that the average can play out.

Losing streaks occur even with a real edge, as OddsShopper's bankroll simulations show; the mechanics are covered under how statistical variance affects results. The Rundown notes that the Kelly Criterion uses the size of the edge to set bet size as a share of bankroll.

A cash-out offer can be checked the same way: the offer is compared with what the live odds imply the bet is worth, and BettingTools says the book's margin applies to that offer as well.