EV calculator
Work out what a bet is worth on average: from your own probability, or from a sharp book's line with the margin stripped out.
The bet
Percentages go in as numbers: 55 means 55%.
What the bet is worth
Positive expected value at this price.
You do not need to outguess the market — only to find a book slower than the sharpest one.
What expected value actually tells you
Expected value is the average result of a bet you could repeat forever. Two inputs, no opinions: the price offered and the probability you give the outcome. Think a team wins 55% of the time and someone offers 2.00, and the arithmetic is immediate — 55% of the time you gain a unit, 45% you lose one, so you make 0.10 units per unit staked. That is a 10% edge, and that is all +EV means.
The probability is where almost everyone quietly loses. A 5-point error turns a 4% edge into a 1% loss, and estimates made by eye miss by far more than 5 points. Hence the second mode of this calculator: instead of trusting your gut, take both sides at a sharp book, strip the margin, and use the fair probability that falls out. Read the result as a rate, not a promise — a 4% edge returns nothing on any single bet, it only appears across hundreds of them, and only if the probability was right. Which is why stake size matters as much as the edge itself.
Borrowing the probability from a sharp book
- Sharp book, both sides
- 1.95 / 1.95
- Fair probability, margin removed
- 50.00%
- Break-even price
- 2.00
- Your book offers
- 2.10
- EV on 100 staked
- +5.00 (+5%)
No opinion was required: the sharp line supplied the probability, and your book priced the same outcome at 2.10 where 2.00 was fair. Move the sharp prices to 1.90/1.90 and the fair probability stays 50% — that is the point of removing the margin, the answer does not move with the vig.
Questions
How do I get a probability I can trust?
Either from a model built on data, or from the market itself. The fair line at a sharp book — its prices with the margin removed — is a hard benchmark to beat with judgement alone, and it is free to look up.
Is a positive EV bet a good bet?
It is a necessary condition, not a sufficient one. A bet can carry positive expected value and still be a bad idea if the stake is too large for the bankroll, if the edge is inside your estimation error, or if the market moves against you before kick-off.
Why does the calculator ask for both sides of the sharp market?
Because you cannot remove a margin from one price alone. The margin is only visible when the implied probabilities of every outcome are added together, so the second price is what makes the fair probability computable.
What does EV per unit staked mean?
It is the same result expressed independently of your stake: 10% means you gain on average 0.10 for every 1 risked. Comparing bets by EV percentage rather than by currency stops a large stake on a thin edge from looking better than it is.
Other free tools
Odds converter
Betting margin calculator
Kelly criterion calculator
Probability calculator
Arbitrage calculator
Parlay calculator
ROI calculator
Yield calculator
Stake calculator
Bankroll calculatorThe model’s call of the day, free on Telegram
No account, no email. Every day we publish the match where our model disagrees most with the market, 15 minutes before it starts — and we settle every pick we publish, won or lost.
Open the channel →