Expected Value in Sports Betting: How to Calculate +EV

Most bettors ask one question before a bet: "Will it win?" Bettors who last ask a different one: "Is this price worth taking?" The tool for answering it is expected value, usually shortened to EV.

This guide explains what expected value means in sports betting, gives you the formula, and walks through worked examples you can reproduce on a calculator. It also covers the part most guides skip: why a bet with positive EV can still lose, and how that changes the way you should judge your results.

What expected value means in betting

Expected value is the average amount you would win or lose per bet if you could place the same bet, at the same price, many times over. It combines two things: how often the bet wins, and how much it pays when it does.

• A positive EV (+EV) bet pays more, on average, than its chance of winning requires. Repeated over a large sample, it makes money.
• A negative EV (-EV) bet pays less than its chance of winning requires. Repeated over a large sample, it loses money, even if it wins often.
• A zero EV bet is a fair price: you break even over time.

EV is not a prediction about the next game. It is a statement about the price. A bet can have strong EV and lose tonight, or have terrible EV and win tonight. Over many bets, the price is what decides the outcome.

The expected value formula

For any bet, the formula is:

EV = (probability of winning × profit if it wins) − (probability of losing × stake)

If you bet $100, "profit if it wins" is what you would collect on top of getting your stake back. Here is how to find the profit from American odds:

• Positive odds (+150): profit on a $100 stake is the odds number, so $150.
• Negative odds (-110): profit on a $100 stake is 100 ÷ 110 × 100 = $90.91.

The one input the sportsbook does not give you is the true probability of winning. That is your estimate, or a model's, and everything else flows from how good it is.

Worked example 1: an underdog

You think a team has a 45% chance to win, and the sportsbook offers +150.

• Break-even probability at +150 is 100 ÷ 250 = 40%.
• EV on a $100 bet = (0.45 × $150) − (0.55 × $100) = $67.50 − $55.00 = +$12.50

That is +12.5% of your stake per bet, on average, if your 45% estimate is right. The bet loses 55% of the time and is still a good price, because the payout more than covers the losses.

Worked example 2: a standard -110 bet

Most spreads and totals sit at -110, where a $100 bet wins $90.91. The break-even win rate is 110 ÷ 210 = 52.38%.

• If the true chance is 50%: EV = (0.50 × $90.91) − (0.50 × $100) = -$4.55
• If the true chance is 52%: EV = -$0.73
• If the true chance is 53%: EV = +$1.18
• If the true chance is 55%: EV = +$5.00

This is why a bettor who wins exactly half their bets at -110 slowly loses money. The gap between 50% and 52.38% is the sportsbook's margin, known as the vig. For the full breakdown, see how the vig works.

How to estimate the true probability

EV is only as good as the probability you feed it. There are three common starting points.

1. Start from the market, then remove the vig. The price itself implies a probability. Convert it, then strip out the sportsbook's margin so the two sides add to 100%. For example, with one side at -130 and the other at +110, the raw implied probabilities are 56.5% and 47.6%, which add to 104.1%. Divide each by 104.1% and you get fair probabilities of about 54.3% and 45.7%. Our guide to implied probability shows the full method for every odds format.

2. Build or use a model. A model estimates win probability from team strength, form, injuries, venue and other inputs. If its estimate is better than the market's, the difference between the two is your edge.

3. Compare against the sharpest price available. If one sportsbook is offering a number noticeably better than the rest of the market, that gap can signal value. Be careful here, because a stale line and a genuine edge look the same until the line moves.

Whatever method you use, treat the result as an estimate with error bars, not a fact. A 3-point edge built on a 5-point-wide guess is not an edge you can count on.

Edge versus EV

The two terms are related but not identical.

• Edge is the difference between your probability and the price's break-even probability, measured in percentage points. A 55% estimate against a 52.38% break-even is a 2.6-point edge.
• EV is what that edge is worth in money, which also depends on the payout. That same bet is worth +$5.00 per $100 staked.

Here is a second example that includes the vig adjustment. Suppose a model says a team wins 58% of the time and the sportsbook price is -130.

• Break-even probability at -130 is 130 ÷ 230 = 56.5%
• Profit on a $100 stake is 100 ÷ 130 × 100 = $76.92
• EV = (0.58 × $76.92) − (0.42 × $100) = $44.62 − $42.00 = +$2.62

That is a 1.5-point edge worth about 2.6% of the stake. It is positive, but small enough that a modest error in the 58% estimate would erase it.

Why a +EV bet can still lose

This is the part that trips people up. Take a bet with a real +5% EV: a 55% win probability at -110, worth +$5.00 per $100 bet. But each individual bet either wins $90.91 or loses $100, so the results swing hard around that small average.

Run the numbers on the spread of outcomes. The standard deviation of one such bet is about $95, nearly 19 times the $5 EV. Over many bets:

• After 100 bets, expected profit is $500, but the standard deviation is about $950. There is roughly a 30% chance you are still behind.
• After 500 bets, expected profit is $2,500 against a standard deviation of about $2,120. The chance of being behind falls to roughly 12%.
• After 1,000 bets, expected profit is $5,000 against about $3,000. The chance of being behind is roughly 5%.

These figures assume the 55% estimate is exactly right, which is the best case. In practice the estimate has error too, so the real picture is more uncertain than this.

Two takeaways follow. First, a short losing streak tells you almost nothing about whether your approach is sound. Second, a short winning streak tells you almost nothing either. Judge a method over hundreds of bets, and track whether you are consistently getting better prices than the closing line, rather than judging by the last week.

What expected value means for parlays

Parlays multiply payouts, and they also multiply the sportsbook's margin. Say you find three legs that are each a true 50% chance, priced at -110.

• A single leg has EV of -$4.55 per $100, or about -4.5%.
• A three-leg parlay at those prices pays about +596. With a true win chance of 0.5 × 0.5 × 0.5 = 12.5%, EV = (0.125 × $595.79) − (0.875 × $100) = -$13.03, or about -13%.

The margin did not stay the same; it roughly tripled. This is the core reason parlays are hard to beat: each added leg brings its own vig. A parlay only has positive EV if every leg is priced better than its true probability, or if the legs are positively correlated in a way the sportsbook underprices. If you want a deeper look, read are parlays worth it and use the parlay calculator to see the combined price before you bet.

How to use EV in practice

1. Convert the price to a break-even probability. Know what win rate the price requires before you look at anything else.
2. Form your own probability. Use a model, a fair-odds calculation from the market, or your own research. Be honest about how uncertain it is.
3. Compute the edge and EV. If the edge is zero or negative, there is no bet. Walking away is the correct result of the exercise.
4. Size the stake sensibly. Even a real edge can lose for a long stretch. Keep each bet to a small, fixed share of your bankroll. See bankroll management for how.
5. Record every bet and its price. Without a log you cannot tell skill from luck.

Where Parlitify fits

Parlitify is built around this exact comparison: a model's win probability set against a real sportsbook price. On Pro plans, the top pick for a game shows a model edge only when the model has a probability for that market and the sportsbook price is real. When no pick beats the price, the card says so rather than inventing a number. You can try the AI parlay builder to see how the model's probabilities line up against the posted odds.

Treat any edge, from any source, as an estimate. It narrows the odds in your favor; it does not remove risk.

Frequently asked questions

What does +EV mean in sports betting?
It means the bet's price pays more than its true chance of winning requires. If you could place it many times, you would expect to profit.

How do you calculate expected value in betting?
Multiply the probability of winning by the profit if it wins, then subtract the probability of losing multiplied by your stake. For a $100 bet at +150 with a 45% win chance, that is (0.45 × $150) − (0.55 × $100) = +$12.50.

Can a +EV bet lose?
Yes, and often. EV describes the average over many bets, not the result of one. Even a bet with a real edge can lose several times in a row.

What is the difference between edge and EV?
Edge is the gap between your estimated probability and the price's break-even probability, in percentage points. EV is what that gap is worth in money per bet.

How many bets does it take to know if I have an edge?
There is no fixed number, and small edges need very large samples. At a 5% edge on -110 bets, there is still roughly a 30% chance of being behind after 100 bets, and about a 5% chance after 1,000. Tracking whether you consistently beat the closing line is a useful extra signal.

Are parlays ever +EV?
They can be if every leg is priced better than its true probability, but the sportsbook's margin multiplies with each leg, so the bar rises quickly. Single bets are easier to price correctly.

Does Parlitify guarantee a profit?
No. No tool can. Parlitify shows probabilities and price comparisons to help you judge value, and every pick carries risk.

Betting involves risk and no price, model or estimate guarantees a result. Only bet what you can afford to lose, and only if you are of legal age where you live.