EV Betting: A Plain-English Guide to Positive EV

EV betting (also called positive EV or +EV betting) means placing bets where your estimated win probability is higher than the price implies, so the bet has positive expected value over the long run. For example, a $100 bet at +150 odds with a true 45% win chance has an EV of 12.50% — an expected profit of $12.50 per bet — even though any single bet can still lose. EV betting is a math framework for identifying favorable prices, not a guarantee of winning any particular wager.

What “expected value” actually means

Expected value (EV) is the average result you’d get if you could repeat the exact same bet many times. It blends the size of the win, the size of the loss, and the true probability of each outcome into a single number. A bet with positive EV is profitable on average over many repetitions; a bet with negative EV loses money on average, even if it occasionally wins.

This is why EV betting is fundamentally different from picking winners. You’re not trying to predict a single outcome — you’re trying to find prices where the payout outweighs the real risk, based on your probability estimate.

Break-even probability: the number every price implies

Every betting price has a built-in “break-even probability” — the win rate required for that bet to have exactly zero EV. If your estimated win chance is above the break-even number, the bet is +EV; if it’s below, it’s −EV.

Price	Break-even probability
−200	66.67%
−150	60.00%
−110	52.38%
+100	50.00%
+150	40.00%
+200	33.33%

Notice that −110, a very common price (explained in our guide to how betting odds work), already requires 52.38% just to break even. On a true coin flip, that higher required win rate comes from the built-in margin covered in our guide to vig in betting: the bet is negative EV unless your real win chance clears the break-even number.

Worked example: +150 at a 45% win chance

Say you believe a $100 bet at +150 odds actually has a 45% chance of winning. At +150, the decimal odds are 2.50, and the break-even probability is only 40.00%. Since your 45% estimate is above that 40.00% threshold, this is a +EV bet with a 5.00 percentage point probability edge.

Here’s the math:

• Profit if it wins: $150.00
• Loss if it loses: $100
• EV: 12.50%
• Expected profit: $12.50 per $100 bet

The fair price implied by a true 45% win chance is exactly 11/9 fractional — about +122 in American odds or 2.22 in decimal, after rounding — so getting +150 means you’re being paid more than your estimated probability justifies, which is exactly the definition of a +EV opportunity. Over 100 such $100 bets, the expected total profit is $1,250.00 on average — though this is not guaranteed, since any individual bet can still lose.

Compare that to the same +150 price with a lower win chance. If your real win probability is only 35% instead of 45%, the EV flips to −12.50%, for an expected loss of −$12.50 per $100. The price didn’t change — only the accuracy of your probability estimate did. This is the central challenge of EV betting: the math is simple, but it only works if your win-probability estimate is accurate. Our EV calculator lets you plug in a price and your estimated probability to see the EV and break-even number instantly.

Using the no-vig market as a probability estimate

Since guessing true probabilities is hard, a common approach is to use a “sharp” (low-margin, heavily traded) market’s no-vig (vig-removed) price as a stand-in for the true probability, then compare it to a price elsewhere. For example, if a sharp market prices a game at −110/−110, the no-vig probability for each side is 50.00%. If you then find +105 for the same side, the EV works out to 2.50%, or $2.50 of expected profit per $100 risked.

The same idea applies to uneven markets. If a sharp market shows −150/+130, the no-vig win probability for the underdog side is 42.02%, with a fair price of +138. Finding that same underdog at +145 elsewhere gives an EV of 2.94%, or $2.94 of expected profit per $100. In both cases, the edge is the gap between the price you can get and the no-vig price — and it is only as real as the sharp market is accurate. A no-vig probability is the market’s estimate, not a certainty.

Staking: why bet size matters as much as price

Finding a +EV price is only half the equation — bet sizing determines how smoothly your bankroll grows (or survives a losing streak). The Kelly criterion is a formula for sizing bets based on your edge and the price. For the +150, 45% example above, full Kelly suggests staking 8.33% of your bankroll. Because win-probability estimates are never perfectly certain, a fraction of Kelly is a common alternative. A quarter-Kelly stake on that same bet, with a $1,000 bankroll, comes to $20.83 — just 2.08% of the bankroll. Smaller fractional stakes reduce variance at the cost of slower average growth.

Common mistakes

• Treating one win or loss as proof. EV is a long-run average; a single +EV bet can still lose, and a single −EV bet can still win.
• Using a hunch instead of a real probability estimate. Without a solid probability (often from a no-vig sharp market), the “EV” calculation is just a guess dressed up in math.
• Ignoring the vig. Your estimated win probability has to clear the price’s break-even probability — not just a coin flip — for the bet to be +EV by your estimate, since built-in margin like the −110 example above raises that bar.
• Oversizing stakes. Betting full Kelly or more amplifies swings; fractional Kelly staking, like the $20.83 example above, is a common way to manage that risk.
• Confusing EV betting with arbitrage betting, which tries to lock in a return across all outcomes rather than relying on an edge on one side — see our arbitrage betting guide for that distinction.

EV betting is a probability framework, not a prediction system, and it carries real financial risk on every individual wager. If you bet, do so only with money you can afford to lose, and gamble responsibly (18+/21+ where applicable).