Sports Betting Value Bets Explained: Implied Probability, Removing the Margin, Expected Value and How to Actually Find an Edge

Updated 20 Oct 2026 · By the BetHubCA Editorial Team

If you hang around betting forums long enough, someone will tell you the secret is "only bet value." Great advice. Totally useless if nobody explains what it means. A value bet isn't a bet you feel good about, and it isn't a big underdog that "could easily win." It's a specific idea with simple maths behind it, and once it clicks, you'll never look at a betting slip the same way again. So let's slow down and go through sports betting value bets properly, from the very first step of reading odds to the habits that separate people who understand value from people who just say the word. Grab a pencil. There's a bit of arithmetic, but nothing you didn't do in high school.

A few ground rules before we start. The legal betting age is 19 in most provinces and 18 in Manitoba and Quebec, and this site isn't for residents of Ontario or Alberta. Cloudbet, the book linked on this page, holds a Curaçao licence and is not authorized by any Canadian province. Every price, probability and stake below is an illustrative example to teach the method. None of it is a live line or a tip.

What a value bet actually is

A value bet is a wager where the price you're getting pays more than the true chance of the outcome deserves. That's it. If something really happens 50% of the time, the fair decimal price is 2.00. If a book offers you 2.20 on it, that's a value bet. If it offers 1.90, it isn't, even if the team is your favourite and you're sure they'll win tonight.

Notice what's missing from that definition: whether the bet wins. A value bet can lose, and most underdog value bets do lose. A bad-value bet can win. Value is about the price versus the probability, measured over many bets. Think of it like a coin that lands heads 50% of the time, and someone pays you $2.20 back on a $1 bet every time it lands heads. You'll lose plenty of individual flips, but over 1,000 flips you'll come out ahead. That's value betting in one sentence.

Step 1: Turn odds into implied probability

Every price carries a probability hidden inside it. In decimal odds, the formula is just 1 divided by the price.

For American odds, the conversion is a little different. For a minus price, divide the number by itself plus 100: −150 becomes 150 ÷ 250 = 60%. For a plus price, divide 100 by the number plus 100: +150 becomes 100 ÷ 250 = 40%. Most Canadian bettors see both formats, so it's worth being comfortable with each. Our odds calculator converts them instantly if you don't want to do it by hand.

Step 2: Understand the margin

Here's where it gets interesting. Add up the implied probabilities of every outcome in a market and you'll get more than 100%. The extra is the book's margin, sometimes called the vig, juice or overround. It's how the book makes money regardless of who wins.

Take an illustrative hockey moneyline: Team A at 1.80 and Team B at 2.10.

• Team A: 1 ÷ 1.80 = 55.6%
• Team B: 1 ÷ 2.10 = 47.6%
• Total: 103.2%

That 3.2% is the margin. It means neither price is "fair." Both have been shaved a bit in the book's favour. Before you can tell whether a price is good value, you need to know roughly what the fair price would be without that margin.

Step 3: Remove the margin to find the "no-vig" price

The simplest way to strip out the margin is to scale each implied probability so they add to 100%. Divide each one by the total.

• Team A: 55.6 ÷ 103.2 = 53.8%, fair price about 1.86
• Team B: 47.6 ÷ 103.2 = 46.2%, fair price about 2.17

That's the market's best guess at the real probabilities once the book's cut is removed. It's not perfect. Books sometimes put more of their margin on the underdog or the popular side, and there are fancier methods to account for that. But for most bettors, the simple scaling method is plenty. Here's how margins look across some typical markets (illustrative):

The lesson: some markets are much more expensive than others. Props and futures often carry far more juice than main lines. That doesn't make them unbettable, but you need a bigger edge to overcome the cost.

Step 4: Expected value, the core formula

Expected value (EV) tells you what a bet is worth on average, over the long run. The formula is:

EV = (probability of winning × profit if you win) − (probability of losing × stake)

Let's run an illustrative example. You think a team has a 55% chance of winning, and the book offers 2.00. On a $10 bet:

• Win: 0.55 × $10 profit = $5.50
• Lose: 0.45 × $10 = $4.50
• EV = $5.50 − $4.50 = +$1.00

So on average, each $10 bet like this earns you a dollar. That's a 10% edge, which is very large by real-world standards. Here's the same idea across a few scenarios on a $10 stake:

A shortcut: EV as a percentage of your stake is simply (your probability × price) − 1. For 55% at 2.00, that's 0.55 × 2.00 − 1 = 0.10, or 10%. For 62% at 1.70, it's 1.054 − 1 = 5.4%. If the result is positive, you've got value. If it's negative, you don't.

Step 5: Where does "your probability" come from?

This is the hard part, and it's where most people fool themselves. Everything above is easy arithmetic. The challenge is producing a probability that's actually better than the market's. There are a few realistic ways bettors do it.

Method A: Compare books (market-based value)

The simplest approach doesn't need you to predict anything. You treat the sharpest, highest-volume prices as a good estimate of the truth, remove the margin, and look for other books offering a better price than that fair number. If the no-vig price on a team is 2.17 and another book offers 2.30, that's a value bet according to the market itself.

This works because books don't all move at the same speed. Some are slower to react to news or to big bets elsewhere. The catch: the edges are usually small, they disappear fast, and books may limit accounts that only bet on stale lines. It's a legitimate approach, but it's not a free lunch.

Method B: Build your own model

Some bettors build ratings or models for a sport. A very simple hockey model might combine each team's goals for and against at five-on-five, adjusted for home ice and the confirmed starting goalie. A simple soccer model might use expected goals and a Poisson distribution. You turn those numbers into probabilities and compare them with the market.

This can work, but be humble. Markets on big leagues like the NFL, NBA and NHL are efficient, meaning prices already reflect most public information. A home-made model needs to be genuinely better than thousands of sharp bettors combined. Most aren't, at least at first. Our NHL over/under strategy shows a simple model in action if you want to try one.

Method C: Specialist knowledge

The third route is knowing a niche better than the market. Smaller leagues, lower-profile sports, and some props get less attention, so prices can be softer. If you follow junior hockey, the CFL, a particular tennis tour or a niche esports scene closely, you might spot things the market misses, like a lineup change, a scheduling spot or a style mismatch.

The trade-off is that these markets often have higher margins and lower betting limits. You might find value, but you can't bet big on it. For most recreational bettors, that's fine.

Step 6: Closing line value, your best report card

Here's a problem with value betting: results are noisy. You could make 100 genuine value bets and still be down money through bad luck. So how do you know whether you're actually good at finding value? The best early answer is closing line value, often shortened to CLV.

The closing line is the final price just before the game starts, when the market has absorbed the most information. If you regularly get better prices than the closing line, it's a strong sign you're finding real value, because the market tends to move toward the right number.

An illustrative example: you bet a team at 2.20 on Monday. By game time, the price has dropped to 2.00. You beat the closing line by a good margin. Even if that bet loses, you made a good bet by this measure. If that keeps happening over dozens of bets, you're likely doing something right.

Track it in your betting log. It's a far more useful signal over 50 bets than your profit and loss, which can swing wildly from luck alone.

Why value bets still lose (and why that's fine)

This is the part people struggle with emotionally. A value bet at 3.00 that's really a 38% shot will lose 62% of the time. You could easily hit five or six losers in a row. Here's an illustrative look at how often losing streaks happen, even with a real edge:

Those numbers are for one specific run of bets. Over a long season with hundreds of bets, you're almost certain to see some long losing streaks somewhere, especially if you bet underdogs. That's why value betting needs patience and sensible staking. The edge shows up over volume, not over a weekend.

A related point is variance. Suppose you make 200 bets at 2.00 with a true 53% win rate, betting $10 each. Your expected profit is about $120. But the natural spread of results is wide: you could finish anywhere from a few hundred dollars up to a loss, just from luck. A small edge needs a large sample to show itself.

Staking value bets sensibly

Finding value is only half the job. If you bet too big, a normal losing streak can wipe you out before the edge has time to work.

Flat staking

Bet the same amount, say 1% or 2% of your bankroll, on every value bet. It's simple, it's hard to mess up, and it's what most recreational bettors should use.

The Kelly criterion

Kelly is a formula that sizes bets based on your edge. For decimal odds, the fraction of your bankroll to bet is: (probability × price − 1) ÷ (price − 1). An illustrative example: 55% at 2.00 gives (1.10 − 1) ÷ 1 = 10% of your bankroll. That's aggressive, and it assumes your probability is exactly right, which it never is.

Most people who use Kelly use a fraction of it, often a quarter or a half, to protect against overestimating their edge. If you're just starting out, flat staking is the safer choice.

Common value betting traps

Trap 1: Confusing "underdog" with "value"

A long shot at 6.00 isn't value because it pays a lot. It's only value if the true chance is above 16.7%. Plenty of underdogs are fairly priced or overpriced, especially popular ones that the public loves to back.

Trap 2: Overrating your own team

Fans are biased. If you're a Canucks fan, you probably think the Canucks are a bit better than they are. The market isn't a fan. Be extra careful with value bets on teams you love. Some bettors refuse to bet on their own team for exactly this reason.

Trap 3: Ignoring the margin on props and parlays

Props often carry 8–10% margins or more, and parlays multiply the margin with every leg. A four-leg parlay where each leg is a fair coin flip priced at 1.91 pays about 13.3, but the fair price would be 16.0. You're giving up a huge chunk of value. Our parlay calculator shows the true cost. If you like props, our guide what is a prop bet explains how they work.

Trap 4: Small samples

Ten winning bets in a row doesn't prove you've found value. Ten losers doesn't prove you haven't. Judge your process over at least 100 bets, and use closing line value as your early signal.

Trap 5: Chasing "sure things"

Anyone selling guaranteed value bets or "fixed" matches is selling a fantasy at best and running a scam at worst. Real value is a small, uncertain edge, never a guarantee.

Trap 6: Hedging away your edge

If you make a value bet and then hedge it every time it gets nervous, you're often trading away the expected value you worked to find. Hedging can make sense for risk reasons, but know the cost. Our hedging bets explained guide shows how to calculate it.

A worked value bet, start to finish

Let's walk through an illustrative CFL example from beginning to end.

1. The market: Team A at 1.65, Team B at 2.35. Implied: 60.6% + 42.6% = 103.2%, so a 3.2% margin.
2. No-vig probabilities: Team A 58.7% (fair 1.70), Team B 41.3% (fair 2.42).
3. Your view: You've noticed Team A's starting quarterback is questionable and the weather forecast is windy, which tends to hurt a passing offence. You estimate Team B at 45%.
4. Compare: At 2.35, break-even is 42.6%. Your 45% is above that.
5. EV: 0.45 × 2.35 − 1 = 0.0575, or about +5.8%. On a $20 bet, that's an expected +$1.15.
6. Shop: Another book offers 2.45 on Team B. EV there: 0.45 × 2.45 − 1 = 10.3%. Same bet, much better value, just from shopping.
7. Stake: Flat 1% of a $1,000 bankroll = $10, or quarter Kelly: (0.45 × 2.45 − 1) ÷ 1.45 × 0.25 = about 1.8% = $18.
8. Log it: Date, game, price, your probability, stake, closing price, result.

Whether Team B wins or loses tonight doesn't decide whether it was a good bet. If the price closes at 2.20, you beat the market. Our CFL moneyline guide covers more of the specifics for three-down football.

Where recreational bettors can realistically find value

• Line shopping. Simply getting the best price available across a few books adds value on every bet you'd make anyway. It's the easiest edge there is.
• Early news. Injury news, lineup confirmations and goalie announcements move prices. Being fast and accurate on news you follow closely can help.
• Niche markets. Smaller leagues and sports you know deeply may have softer prices, though often with lower limits.
• Avoiding bad bets. Cutting out high-margin parlays, novelty props and impulse live bets doesn't find value, but it stops you burning it.
• Patience. Most games don't offer value. Passing is a skill.

Keeping a value betting log

A simple spreadsheet is all you need. Columns to include:

After 100 bets, group them by your stated probability. If the bets you rated at 60% won about 60% of the time, your estimates are well calibrated. If they won 50%, you've been overconfident, and you'll want to shrink your edges.

Value betting and pro-level parlays

Some bettors argue that correlated parlays can create value, for example combining a team to win with the game total going over when those outcomes are linked. Sometimes books price that correlation properly, sometimes not, and many books restrict obvious correlated combinations. If you're curious, our article on professional sports betting and leg parlays digs into it. For most people, though, singles are where value is easiest to measure and protect.

Value in three-way and two-way markets

The maths changes a little when a market has three outcomes instead of two. In soccer, the standard 1X2 market includes the draw, and the draw is where many casual bettors misjudge value. People tend to bet on teams, not on ties, so draw prices can sometimes drift a bit long in evenly matched games. That doesn't mean "always bet the draw." It means check it.

An illustrative example: home 2.50, draw 3.20, away 3.00. Implied probabilities are 40.0% + 31.3% + 33.3% = 104.6%. Scale them and you get about 38.2%, 29.9% and 31.9%, so fair prices of roughly 2.62, 3.34 and 3.14. If you think a cagey match between two defensive sides has a 32% chance of a draw, the 3.20 price gives 0.32 × 3.20 − 1 = +2.4%. Small, but positive.

Two-way markets like hockey moneylines (with overtime included), tennis match winners and most totals are simpler, because there are only two outcomes to compare. That makes it easier to spot when one side is off. It also makes them more efficient, because everyone can do the same simple maths.

Totals deserve a special mention. On a total of 5.5 goals in hockey or 47.5 points in football, the half point means there's no push. On a whole number, like 6 or 48, a push is possible, and that changes the break-even. If a push happens 10% of the time, your win and loss probabilities need to be calculated on the remaining 90%. Ignore that and you can think you've found value when you haven't.

How value looks in different sports

These ratings are general impressions, not measured facts, and they change over time. The pattern holds, though: the more money and attention a market gets, the harder it is to find value in it. That's why many sharp bettors focus on one or two smaller areas they know deeply instead of betting a bit of everything.

A simple calibration exercise

Before you trust your own probabilities with real money, test them. Here's an exercise that costs nothing:

1. For two weeks, pick 30 to 50 games you'd normally bet. For each one, write down your probability for one side before the game, without looking at the odds first.
2. Then write down the market's no-vig probability for the same side.
3. After the games, compare. Did your numbers or the market's numbers do better? A simple way to score it: for each game, square the difference between the probability you gave and the result (1 if it happened, 0 if it didn't). Lower total is better. This is called a Brier score.
4. If the market beats you, which is very common at first, you've learned something valuable without losing a dollar.

Most people discover that their gut probabilities are more extreme than reality. They rate their favourites at 70% when the market says 58%, and the market is usually closer. Pulling your estimates toward the market price, sometimes called shrinking, is often the most honest adjustment you can make.

Do bonuses and promos create value?

Sometimes, yes. A free bet, a deposit match or an odds boost can create positive expected value on its own, because the book is effectively giving you something extra. The catch is always in the terms. Wagering requirements, minimum odds, expiry dates and maximum win caps can turn a generous-looking offer into something worth very little.

A rough way to value a promo is to ask: if I follow the terms exactly, what do I expect to get back on average, minus what I expect to lose meeting the requirements? An odds boost on a single bet with no strings is usually the cleanest. A big bonus that needs to be wagered many times on high-margin markets is often worth much less than the headline. Read the full terms before you count any promo as value, and never deposit more than you planned just to unlock one.

Tax and the law

For most recreational bettors in Canada, gambling winnings are generally not treated as taxable income, though people who bet in a business-like way can be treated differently. Our guide to tax on betting winnings explains the basics. Gambling rules are set province by province, so read is online betting legal in Canada before signing up anywhere.

Keep it fun

Understanding value makes betting more interesting, because every bet becomes a little puzzle instead of a guess. It doesn't make betting safe or guaranteed. Set a budget you're comfortable losing, stick to it, and never chase losses because you think you're "due." If betting stops feeling like entertainment, free and confidential help is available: Gambling Support BC at 1-888-795-6111, and Jeu : aide et référence in Quebec at 1-800-461-0140. You'll find more options on our responsible gambling page.

My take

Value betting comes down to one question you ask before every bet: does this price pay more than the real chance deserves? To answer it, convert odds to probability, strip out the margin, form your own estimate honestly, and do the EV maths. Shop for the best price, because that's the cheapest edge you'll ever find. Stake flat or with a small fraction of Kelly, and judge yourself by closing line value rather than last weekend's results. Most of the time the answer to the value question will be no, and that's the point. The few times it's yes are the only bets worth making.