Expected Value (EV) Calculator

Calculate the true expected value of any sports bet. Enter your estimated win probability, the odds, and your stake to see EV in dollars and percent, your edge over the market, and the break-even win rate.

What Is Expected Value in Sports Betting?

Expected value (EV) is the average amount you'd profit or lose per bet if you could place the exact same wager an infinite number of times. It's the single most important concept in quantitative betting — more important than win rate alone, because a bet can have a low win probability and still be profitable if the payout is large enough.

The formula is: EV = (p × profit) − ((1 − p) × stake), where p is your estimated win probability, profit is what you win if the bet hits (stake × (decimal odds − 1)), and stake is what you risk. When EV is positive, the bet is profitable long-run if your probability estimate is accurate. When EV is negative, the sportsbook's vig outweighs your edge.

The hard part isn't the math — it's getting p right. Read the full expected value guide for how to build reliable probability estimates, and the Kelly Criterion guide for how to size a bet once you know it's +EV.


Calculator

Formula Breakdown

$2.50
Expected Value ($)
2.50%
EV (% of Stake)
52.4%
Implied Probability
2.6 pts
Your Edge
52.4%
Break-Even Win Rate

Edge Detected


A Note on the Vig

The "implied probability" shown above is the raw implied probability from the odds — it has not been stripped of the sportsbook's built-in margin (the vig, or juice). On a standard -110/-110 two-way market, each side implies roughly 52.4%, and the two sides sum to about 104.8% instead of 100% — that extra 4.8% is the book's edge. True "no-vig" or "fair" probability divides each side's implied probability by the sum of both sides' implied probabilities to remove the overround. For a single-sided EV check like this calculator, comparing your probability against the raw implied probability is the more conservative (and more common) approach, since it already prices in the vig you're actually paying.


How Olympus Bets Uses EV

Every recommendation on Olympus Bets starts as an EV calculation, but the win probability that feeds it is never a guess:

  1. Monte Carlo simulation produces raw win/cover/total probabilities from 10,000+ iterations per game
  2. Bayesian calibration shrinks the raw probability toward 50% to correct for model overconfidence before it's ever used for EV
  3. EV and edge are computed against the actual offered odds, not a theoretical fair line
  4. Kelly sizing converts a positive EV into a unit recommendation, so bet size scales with edge strength
  5. Profitability zone gating blocks bets in historically unprofitable sub-niches even when raw EV looks positive

This is why a "positive EV" number alone is never enough at Olympus Bets — it has to survive calibration and zone gating before it becomes a recommendation.


See Calibrated Edges in Action

Every Olympus Bets pick shows the model probability, the market's implied probability, and the resulting edge — all calibrated through our full probability pipeline.

View Free Picks Go Premium

Further Reading