Implied odds, more precisely implied probability, is the chance of winning that a betting price says an outcome has. A line of -150 implies 60%, +200 implies 33.3%, and decimal 2.50 implies 40%. Once you can make that conversion in your head, every line on a sportsbook becomes a claim you can test: is the real chance higher or lower than the price says?
This guide shows the formulas for all three odds formats, a conversion table you can bookmark, how the sportsbook's margin distorts the numbers, and how to use implied probability to find value instead of just reading it.
TL;DR
- Implied probability is the break-even win rate of a price. If you win more often than the number says, you profit over time. If you win less often, you lose.
- American odds: for a favorite, odds ÷ (odds + 100). For an underdog, 100 ÷ (odds + 100). Decimal odds: 1 ÷ decimal. Fractional odds: denominator ÷ (denominator + numerator).
- Both sides of a market usually add up to more than 100%. The extra is the sportsbook's margin (the vig), and you should remove it before comparing a line with your own estimate.
- A price is only worth betting when your estimate of the true chance beats the implied probability by a margin big enough to cover your error.
Table of Contents
- What implied probability actually means
- First, read the odds format
- The formulas, one format at a time
- A conversion table you can keep
- Why both sides add up to more than 100%
- Removing the vig to find the "fair" price
- How to use implied probability to find value
- Our perspective
- Where our platform fits
- Quick reference: reading a line in ten seconds
- FAQ
- Sources
What implied probability actually means
A betting price is a probability in disguise. Probability is a percentage chance, while odds can be fractional, decimal or moneyline, and each one implies a probability for the outcome. Sporting News puts it as the percentage chance an outcome would need to have, given the posted price, not a guarantee of what will happen.
That last part matters. Implied probability is what the market is charging you for, not what is true. The sportsbook sets a price, the price implies a number, and your job is to decide whether the number is too high, too low or about right.
There is a second reading that makes this practical: implied probability is the break-even rate. If a price implies 52.38%, you need to win 52.38% of such bets just to end up even. Every pick you make carries that hurdle, whether or not you ever calculate it.
First, read the odds format
The same price can be written three ways. Which one you see depends on where you bet. American odds are popular in the United States, fractional odds are traditional in Britain and Ireland, and decimal odds are common in continental Europe, Australia, New Zealand and Canada.
American (moneyline) odds
Negative numbers belong to favorites and show how much you must risk to win $100. Positive numbers belong to underdogs and show the profit on a $100 stake. You do not need to stake $100: as The Athletic explains, at -110 you risk $110 to profit $100, and at -200 a $10 stake produces $5 profit plus your stake back. The numbers scale in proportion.
Decimal odds
Decimal odds show the total return per $1 staked, stake included. Decimal 2.50 returns $2.50 for every $1, which is $1.50 of profit. Multiply by your stake and you have the payout, which is why many bettors find this format the easiest to work with.
Fractional odds
Fractional odds show profit relative to stake. Sporting News gives the example that 3/2 wins $3 for every $2 wagered. The same logic gives 7/4 on a $10 stake: $17.50 profit, plus your $10 back. So when someone asks what 7 to 2 odds mean, the answer is $7 of profit for every $2 risked, which works out to a 22.22% implied chance of winning.
The formulas, one format at a time
American odds to implied probability
The rule has two halves. For positive odds, implied probability is 100 ÷ (odds + 100). For negative odds, it is the absolute value of the odds ÷ (the absolute value + 100). Sporting News works the same examples: +200 gives 33.3% and -200 gives 66.7%.
Try a few more:
- -150: 150 ÷ 250 = 60.0%
- +150: 100 ÷ 250 = 40.0%
- -110: 110 ÷ 210 = 52.38%
- +250: 100 ÷ 350 = 28.57%
Notice that -150 and +150 are not mirror images. They imply 60% and 40%, which total exactly 100%, but only because those two prices happen to be mirror images with no margin. Real two-sided markets almost never look like that, as the next sections show.
Decimal odds to implied probability
Divide 1 by the decimal price and multiply by 100. Smarkets gives the example of decimal 5.50 implying about 18.1%. Likewise 2.00 is 50%, 1.50 is 66.7%, and 4.00 is 25%.
Fractional odds to implied probability
For odds of n/d, calculate d ÷ (d + n). The same Smarkets page lists this fractional formula alongside the decimal one. So 7/2 is 2 ÷ 9 = 22.22%, 3/2 is 2 ÷ 5 = 40%, and 7/4 is 4 ÷ 11 = 36.36%.
Moving between formats
If you prefer to work in one format, convert first. TNT Sports lists the general conversions: fractional to decimal is 1 plus the numerator divided by the denominator, and decimal to fractional is the decimal minus 1. For American odds, positive prices convert to decimal as odds ÷ 100 + 1, and negative prices as 100 ÷ the absolute odds + 1.
A conversion table you can keep
| American | Decimal | Fractional | Implied probability |
|---|---|---|---|
| -300 | 1.33 | 1/3 | 75.0% |
| -200 | 1.50 | 1/2 | 66.7% |
| -150 | 1.67 | 2/3 | 60.0% |
| -110 | 1.91 | 10/11 | 52.38% |
| +100 | 2.00 | 1/1 | 50.0% |
| +150 | 2.50 | 3/2 | 40.0% |
| +200 | 3.00 | 2/1 | 33.3% |
| +250 | 3.50 | 5/2 | 28.57% |
| +350 | 4.50 | 7/2 | 22.22% |
| +400 | 5.00 | 4/1 | 20.0% |
Decimal values are rounded to two places. The probabilities come straight from the formulas above.
Why both sides add up to more than 100%
Put -110 on each side of a point spread and the implied probabilities are 52.38% plus 52.38%, which is 104.76%. That cannot be true of a real event, since one team covers and the other does not. The excess is the sportsbook's margin.
Implied probabilities can exceed 100% because bookmakers price outcomes shorter than their true chances, building in a margin. TNT Sports calls the amount above 100% the margin, overround, vig or juice, and you find it by adding the probabilities and subtracting 100%. Here the overround is 4.76%.
This is also why -110 on both sides is the price of doing business. You need to win 52.38% of your bets to break even, not 50%. At 52 wins in 100 at -110 you finish a little behind, and at 53 you finish a little ahead.
Removing the vig to find the "fair" price
If you want to compare a sportsbook line with your own estimate, you should compare against a probability with the margin taken out. The simplest method, described in the CRAN documentation for the implied package, divides each outcome's inverted odds by the sum of all inverted odds.
A worked example, with a moneyline of -130 on the home team and +110 on the road team:
- Home: 130 ÷ 230 = 56.52%
- Road: 100 ÷ 210 = 47.62%
- Total: 104.14%, so the overround is 4.14%
- Home fair probability: 56.52 ÷ 104.14 = 54.27%
- Road fair probability: 47.62 ÷ 104.14 = 45.73%
The no-vig numbers now sum to 100%. For a -110/-110 spread the answer is a clean 50% each.
Treat this as an estimate, not the truth. The same documentation warns that margin removal can still leave bias, including favorite-longshot bias, and that this basic normalization is the simplest and most common method but can be the least accurate. Sportsbooks do not necessarily spread their margin evenly across both sides, so the fair number is a good reference point rather than a verdict.
How to use implied probability to find value
Reading the number is step one. Using it takes three more steps.
- Convert the price. Turn the line into an implied probability, then strip the vig if the market has two sides.
- Form your own estimate. Use something defensible: a model, matchup data, recent form, injury news. A gut feeling counts as an estimate, but it should be treated with suspicion.
- Compare. If your estimate is higher than the implied probability, the price may have value. If it is lower, there is nothing to bet.
Here is a hypothetical. A prop is priced at -110 on the over, so the break-even is 52.38%. Your analysis says the over hits 58% of the time. Over 100 bets of $110 risked to win $100, you would expect 58 wins ($5,800) and 42 losses ($4,620), a profit of $1,180, or about 10.7% of the money risked. That gap of 5.6 percentage points is your edge, and it is the reason to bet.
Now flip it. If your honest estimate is 53%, the edge is under one point. That is smaller than the error in most estimates, so a careful bettor passes. The edge you need should scale with how unsure you are.
Common mistakes
- Comparing against the raw number. If you compare 54% with 56.52% and ignore the vig, you will undercount your edge on one side and overcount it on the other.
- Treating the implied number as the real chance. It is a price, set by a business that also manages liability and public betting patterns.
- Ignoring price shopping. The same outcome can be -110 at one book and -105 at another. That is 52.38% versus 51.22%, and over hundreds of bets the difference is real money.
- Overconfidence. A model that says 58% when the truth is 53% turns a perceived edge into a loss.
If you bet baseball, the same price-first logic applies to totals, run lines and props, and we walk through it in our price-first workflow for analyzing MLB bets.
Our perspective
We build a sports analytics platform, so we spend a lot of time on the gap between a price and a probability. Here is what we have learned.
Implied probability is the easy half of the job. Anyone can compute 52.38% from -110. The hard half is the estimate you compare it with, and that is where most bettors fool themselves, because a pick that was right feels like proof of skill and a pick that was wrong feels like bad luck.
That is why we grade every published pick against the official box score and keep the results public. The only honest way to know whether your probability estimates beat the market is a long, checkable record. Ours currently stands at 57.6% on 2,413 graded picks, and we would rather you judge us on that page than on any claim we make. Notice what that number is measured against: the break-even rate of the prices involved. A hit rate means little until you set it next to the implied probability of the odds you took.
Our advice is simple. Convert every price, remove the vig, write your own number down before you look at the line, and track how often you are right. Wins and losses on single bets teach you very little. A graded record over hundreds of bets teaches you a lot.
— Brad
Where our platform fits
If you would rather not do the arithmetic by hand on every line, our platform pairs machine learning predictions with edge detection and line movement tracking, so you can see where a price and a model disagree. The Free plan has no published price, Pro is $9.99 per month and Premium is $19.99 per month. You can compare what each includes on the pricing page, and the results of every pick are public on our track record page.
Quick reference: reading a line in ten seconds
- Negative American odds: the favorite. Divide the number by itself plus 100.
- Positive American odds: the underdog. Divide 100 by the number plus 100.
- Decimal: divide 1 by the price.
- Fractional: divide the bottom by the top plus the bottom.
- Two-way market: add both sides. Anything above 100% is the vig.
- Break-even: the implied probability is the win rate you need.
A few anchors help mental math. Even money is 50%. Anything around -200 is about two-thirds. Anything around +300 is 25%. With those three in your head, most lines can be estimated within a few points without a calculator.
FAQ
How do you calculate implied probability from odds?
Convert the price using the formula for its format. For decimal odds, divide 1 by the price. For American favorites, divide the odds by the odds plus 100, and for underdogs divide 100 by the odds plus 100. Multiply by 100 for a percentage.
What does 7 to 2 odds mean?
Seven to two means you win $7 of profit for every $2 you stake, plus your stake back. Its implied probability is 2 ÷ 9, or 22.22%, which means the market gives that outcome roughly a one-in-four-and-a-half chance. In decimal that is 4.50, and in American odds it is +350.
Why do implied probabilities add up to more than 100%?
Because sportsbooks build a margin into every price. If both sides of a market are -110, each implies 52.38% and the total is 104.76%. The extra 4.76% is the overround, also called the vig or juice.
Is implied probability the same as the true chance of winning?
No. It is the chance the price assumes, including the sportsbook's margin and its view of how people will bet. Removing the vig gets you closer to a fair estimate, but biases such as favorite-longshot bias can remain.
What win rate do I need to break even at -110?
You need to win 52.38% of your bets. That is 110 ÷ 210. Anything above that rate is profit over a large sample, and anything below it is a loss.
Sources
- Understanding Betting Odds: Math, Probability, and Gambling (Investopedia)
- Sports Betting Odds: How They Work and How to Read Them (Investopedia)
- How to read betting odds: What + and - mean in sports (Sporting News)
- Understanding sports betting odds and how to read them (The Athletic)
- Betting odds explained: How to read and convert odds (TNT Sports)
- How to calculate implied probability in betting (Smarkets)
- Introduction to the implied package (CRAN)