The fair price
Each book's own margin comes off its own market — never one book's favourite against another's underdog, which describes a market that never existed. The result is a probability you can argue with.
Pregame analysis · NBA · NFL · MLB · WNBA
Know the fair price before you bet
Every sportsbook price has the house margin baked into it. Antivig takes it out, shows you what the game is really worth, and puts every pick we publish on a record that cannot be edited afterwards.
See today's board Read the pick record
No wagering. No balances. No prizes. 21+.
A typical moneyline
3.6% margin
Favourite
−145
−133
Underdog
+125
+133
The book asks for 103.6% of the game. The fair price is what is left when you take the 3.6% back out — and it is the number every decision here is measured against.
The best number
The highest price on each side, which book is hanging it, and how long ago we saw it. A price nobody has confirmed in fifteen minutes is marked, not quietly shown.
The whole record
Every pick is timestamped before the game starts and chained to the one before it. Nothing can be deleted, back-dated or quietly improved — including by us.
Today's board
See every gameNever bet before? Start here
Three ways to bet a game
Moneyline. Pick which team wins, outright. If they win, you win.
Spread. The favourite has to win by more than the number; the underdog can lose by less and still win you the bet. "Chiefs -3.5" means the Chiefs must win by four or more.
Total. Bet on the combined score, whoever wins. "Over 47.5" wins if the two teams score 48 or more between them.
Books do not always agree on the number. Chiefs -3 and Chiefs -3.5 are different bets — about one NFL game in seven is decided by exactly three points — so prices are only ever compared at the same line, and the board tells you when a book is on a different one.
You will not find parlays or player props here, and that is deliberate. A parlay is several bets stapled together, and the book takes its cut on every leg — so a four-leg parlay carries roughly four times the house edge, which is why books advertise them. This site exists to help you pay less than a bet is worth. A parlay is the quickest way to pay more.
How to read a price
A minus number is the favourite and tells you what you must risk to win $100: at -110 you risk $110. A plus number is the underdog and tells you what you win from $100: at +130 you win $130. Nothing else about the notation matters.
What "fair" means here
A sportsbook prices both sides so the two add up to more than 100%. That extra is its cut — the vig — and it is why betting at random loses money slowly. We take each book's cut back out and show what is left: the price the bet would carry if nobody took a margin.
What you are looking for
One book pricing a side better than that fair number. That is the whole idea: not picking winners, but paying less than the outcome is worth. Most of the time no book does, and the board says so rather than inventing something.
Is there a safe bet?
No — and anyone selling one is not being straight with you. A heavy favourite wins most of the time but pays little, and the book's cut is built into that price, so backing favourites at the posted odds still loses money over time. This board shows two separate things: who the market expects to win, and where a price pays more than it should. They are often different teams.
Being right is not the same as winning
A price worth a 1% edge still loses close to half the time. The edge only shows up across hundreds of bets, which is why the stake we suggest is small and why anyone promising certainty is selling something.
This is analysis, not advice. Antivig takes no bets and holds no money. 21+.
19 games scheduled · page 1 of 4
Carolina Panthers at Green Bay Packers
- The likelier side of this line is Over 45.5, at about 50% once the bookmaker margin is taken out. Lines are set to be close to even, so neither side is a strong favourite.
- The market's view of Over 45.5 has held at about 50% since we first saw this game.
2 books · 4.76% margin
| Side | Line | Best price | Book | Chance | Value | Stake | Seen |
|---|---|---|---|---|---|---|---|
| Over | 45.5 | -110 | betus | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
| Under | 45.5 | -110 | betus | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
No side beats its fair price here. That is the usual answer, and it is the one worth trusting.
Charlotte Hornets at Washington Wizards
- The likelier side of this line is Over 235.5, at about 50% once the bookmaker margin is taken out. Lines are set to be close to even, so neither side is a strong favourite.
- The market's view of Over 235.5 has held at about 50% since we first saw this game.
1 book · 4.76% margin
| Side | Line | Best price | Book | Chance | Value | Stake | Seen |
|---|---|---|---|---|---|---|---|
| Over | 235.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
| Under | 235.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
No side beats its fair price here. That is the usual answer, and it is the one worth trusting.
Chicago Bulls at Boston Celtics
- The likelier side of this line is Over 232.5, at about 50% once the bookmaker margin is taken out. Lines are set to be close to even, so neither side is a strong favourite.
- The market's view of Over 232.5 has held at about 50% since we first saw this game.
1 book · 4.76% margin
| Side | Line | Best price | Book | Chance | Value | Stake | Seen |
|---|---|---|---|---|---|---|---|
| Over | 232.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
| Under | 232.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
No side beats its fair price here. That is the usual answer, and it is the one worth trusting.
Detroit Pistons at Brooklyn Nets
- The likelier side of this line is Over 225.5, at about 50% once the bookmaker margin is taken out. Lines are set to be close to even, so neither side is a strong favourite.
- The market's view of Over 225.5 has held at about 50% since we first saw this game.
1 book · 4.76% margin
| Side | Line | Best price | Book | Chance | Value | Stake | Seen |
|---|---|---|---|---|---|---|---|
| Over | 225.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
| Under | 225.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
No side beats its fair price here. That is the usual answer, and it is the one worth trusting.
New York Knicks at Philadelphia 76ers
- The likelier side of this line is Over 231.5, at about 50% once the bookmaker margin is taken out. Lines are set to be close to even, so neither side is a strong favourite.
- The market's view of Over 231.5 has held at about 50% since we first saw this game.
1 book · 4.76% margin
| Side | Line | Best price | Book | Chance | Value | Stake | Seen |
|---|---|---|---|---|---|---|---|
| Over | 231.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
| Under | 231.5 | -110 | draftkings | 50.0% -100 | -4.55% | No stake: this price does not beat its fair value. | 1096h ago |
No side beats its fair price here. That is the usual answer, and it is the one worth trusting.
How it works
-
1
We watch the market
Prices are read from licensed feeds and stored only when they move — so the line history is a record of what actually happened, not of when a job last ran.
-
2
We take the margin out
Four methods, because they disagree on lopsided games and that disagreement is the favourite-longshot bias. You can see which one is being used.
-
3
We publish, then we are judged
A pick goes on the record before the game, and is settled against the closing line afterwards. The record shows the losers at the same size as the winners.
The part most sites skip
A record you can check yourself
Every published pick carries a fingerprint of its own contents and of the pick before it. Change one old row and every row after it stops matching — which is what makes "we went 62% last month" something you can verify instead of something you have to believe.
It proves nothing was altered after the fact. It does not prove we published everything we thought — so we publish every pick we make, winners and losers, and show the open ones too.
Start with the free board
Fair prices and the public record cost nothing — they are the argument. Membership adds value alerts and closing-line value. Model probabilities come only once a model has earned them.