How betting odds work

A price is a probability with the bookmaker's cut baked in. This guide shows how to read the three odds formats, take the cut back out, and price a parlay whose legs move together. Every number on LineSpy is built from these steps.

Three ways to write the same price

American odds are quoted against 100. A minus number is what you risk to win 100: at −150 you risk 150 to win 100. A plus number is what a 100 stake wins: at +130, 100 wins 130.

Decimal odds are the total return per unit staked, stake included. −150 is 1.667 and +130 is 2.30.

Fractional odds are profit over stake. +130 is 13/10 and −150 is 2/3.

To convert: a plus price A is decimal 1 + A/100; a minus price A is decimal 1 + 100/|A|. The odds converter does all of them at once.

Implied probability

The chance a price implies is one over its decimal odds. At −150 that's 1 / 1.667 = 60%. At +130 it's 1 / 2.30 = 43.5%. If you think something happens more often than its implied chance, the price is in your favour; if less, it isn't.

The vig

Add up the implied chances of every side of a market and you get more than 100%. A standard point spread at −110 on both sides implies 52.4% + 52.4% = 104.8%. That extra 4.8% is the book's margin, called the vig or juice. It's how the book makes money whichever side wins.

So a −110 price doesn't mean the book thinks the side is 52.4% likely. Its real opinion is closer to 50%, with the margin added on top. Taking the margin out is called devigging.

Devigging: four ways to remove the margin

There's no single correct way to split the margin between sides, so there are several methods. They agree on even markets and differ on lopsided ones.

Proportional (also called multiplicative) divides each implied chance by the total. In a −250 / +200 market the implied chances are 71.4% and 33.3%, totalling 104.8%. Dividing gives 68.2% and 31.8%.

Equal (additive) subtracts the same amount from each side: half of the 4.8% from each, giving 69.0% and 31.0%.

Power raises every implied chance to the same power, chosen so they sum to 100%. It takes more of the margin from the longshot, matching the fact that books tend to shade longshots more.

Shin models the book as protecting itself against a share of bettors with inside information. It also loads more of the margin onto the longshot, and works well on markets with three or more outcomes, such as a soccer match.

For close markets the methods land within a few tenths of a percent of each other. For a big favourite against a longshot, power and Shin give the favourite a higher fair chance than proportional does. You can compare all four side by side in the devigger.

Why LineSpy uses the median of many books

One book's devigged price is one opinion, and it can be stale or shaded toward the side its customers like. Devigging every book and taking the median gives a steadier read of where the market sits. Lines far from all the others, usually alternate spreads or totals, are flagged and left out so they don't drag the median.

Sharp books such as Pinnacle accept large bets from professionals and move quickly when they're wrong, so their prices are often treated as the market's best estimate. When a US book's price sits away from the sharp price, LineSpy shows the gap. A gap is a difference of opinion between books, not a guaranteed edge.

Parlays and correlation

A parlay pays only if every leg wins. If the legs are unrelated, the fair chance is the product of each leg's chance: two 50% legs make a 25% parlay, a fair price of +300.

Legs from the same game are rarely unrelated. A team's moneyline and its star's points over tend to hit together: when he scores a lot, his team usually wins. That positive correlation makes the parlay more likely than the product says. Two 50% legs with a moderate positive link might hit 30% of the time, a fair price of about +233 rather than +300. Negative correlation works the other way: a team's win and the opponent star's big night rarely go together.

Books know this, which is why same-game parlay prices are usually shorter than the product of the legs. The fair parlay tool lets you set how linked the legs are and see the fair price next to the book's.

How the simulated games work

The sim section on the board replays a single simulated game, possession by possession. Each player's shots come from his real season profile: how often he shoots at the rim, from midrange and from three, and how often those go in. Players ruled out on the injury report are removed and their minutes go to teammates. The simulation is then tuned so that, over 1,500 games, the average margin and total match the books' consensus spread and total.

The game you watch is one of those 1,500, chosen to land near the typical result. It shows how a game like tonight's could go, not how it will. When the injury report changes, the games are run again and the change is shown.

Closing line value

Lines move as news and money arrive, and the last price before tip-off, the closing line, is usually the most accurate the market gets. Bettors who consistently get better prices than the close tend to do well over time; those who don't, don't. It's a far better yardstick than a short run of wins and losses, which is mostly luck.

What none of this means

Fair prices, gaps and simulations describe the market. They don't say what to bet, and the books' margin means most bettors lose over time. LineSpy is for information and entertainment; read our responsible gambling page, and only bet where it's legal and with money you can afford to lose.