How Exact Score Bets Are Priced, Settled and Worth Taking

Why the margin on correct score is high

On the 1X2 market the book's overround is usually a few per cent. Take every price in a correct score market, divide one by each, and add the results: the total is often between 115 and 130%. The book is keeping somewhere between 13 and 23 pence of every pound staked across the whole market, before anyone has picked a winner.

Three things push it up. There are many outcomes, and a small shave on each one adds up. Bettors back favourite scorelines and long shots with little price sensitivity, so competition between books is weaker than on the match result. And the outcomes are rare enough that a punter rarely notices a price is short, because the bet loses most of the time whatever the price.

The margin is rarely spread evenly. Long scorelines such as 4–3 or 0–4 tend to carry more of it, while the common scores near the top of the list carry less. But “less” still means a lot by football standards.

Pricing a scoreline yourself: two Poissons multiplied

The standard way to estimate a score's probability is the Poisson distribution. You give each side an expected goals figure, called λ (lambda), treat the two teams' goals as independent, and multiply. The chance of a team scoring exactly k goals is e^−λ × λ^k / k!. The chance of a home score h and an away score a is the home probability for h multiplied by the away probability for a.

Take a home side expected to score 1.6 and an away side expected to score 1.1. The single-team chances are: home 0 goals 20.19%, 1 goal 32.30%, 2 goals 25.84%; away 0 goals 33.29%, 1 goal 36.62%. Multiply the pairs and you get the table below. Fair odds are one divided by the probability. The last column shows what a price might look like if a book applied a 125% overround evenly, which is an illustration, not any site's quote.

Two things stand out. The likeliest single score is a draw, 1–1, even though the home side wins this match about 49% of the time on the same model. A favourite can win by many different scores, while 1–1 is one point where both teams' distributions peak together. And the gap between fair and quoted odds is wide: a 10.00 stake on 1–1 at 6.76 has an expected value of about −2.00, while at the fair 8.45 it would be break-even.

The Poisson calculator runs the full grid from 0–0 up to 10–10. The model is only as good as the two λ figures you feed it, and it tends to underrate draws, especially 0–0 and 1–1 in tight, low-scoring matches. Many bettors nudge those two lines up by a point or so before comparing with the book.

The “any other score” line

The listed scores never cover every outcome, so the book adds a catch-all. With our 1.6 and 1.1 inputs, every score up to 3–3 adds up to 89.75%, leaving 10.25% for everything else, fair odds 9.75. That is close to the chance of a 1–0, which surprises most people the first time.

Read the label carefully. A single “any other score” line wins on any result not listed. Split versions, such as any other home win, any other draw and any other away win, each cover only their part. If the home side wins 4–2 and you hold “any other away win”, you lose. The catch-all is also where the margin can hide best, because few bettors know its fair price, so check it against the model before backing it.

Dutching several scores

One way to soften the all-or-nothing shape is to back several scores at once and size the stakes so that any of them returns the same amount. This is dutching. Say you like a narrow home win or a draw and back 1–0 at 7.50, 1–1 at 6.80 and 2–1 at 8.50, all illustrative prices, with a total stake of 30.00.

• Each stake is the total multiplied by that line's implied chance, divided by the sum of the implied chances. The implied chances are 13.33%, 14.71% and 11.76%, a sum of 39.80%.
• 1–0 gets 10.05, 1–1 gets 11.08 and 2–1 gets 8.87, which adds back to 30.00.
• Whichever of the three lands, the return is about 75.37, a profit of 45.37. The combined odds of the group are 2.51.

Now compare with the model. Our three scores add up to 32.04% on the Poisson figures, fair combined odds of 3.12. The group is priced at 2.51, so the expected result of the 30.00 is about −5.85. Dutching changed the shape of the bet, not its value. It pays only when your probability for the group is higher than the 39.80% the prices imply. The dutching calculator does the stake split for up to six lines.

Half-time correct score

Many sites that offer the full-time market also price the score at the interval. Fewer goals are expected in one half than in a full match, so the low scores dominate. Suppose, as an assumption for this example, the home side is expected to score 0.7 before the break and the away side 0.5. Then 0–0 at half-time is 30.12% (fair 3.32), 1–0 is 21.08% (fair 4.74), 0–1 is 15.06% (fair 6.64) and 1–1 is 10.54% (fair 9.49).

The half-time market suits bettors who have a firm view on how a match starts, for example a side that sits deep early. Its weakness is the same as the full-time version: lots of outcomes, a wide overround, and a model that needs a sensible first-half λ. Combined half-time and full-time score bets multiply two uncertain outcomes, and the margin compounds with them, so those lines are rarely good value.

When correct score is worth a bet

The market makes sense when you have a sharper view of the goal count than the result. If you expect a low-scoring match, a small group of scores such as 0–0, 1–0 and 1–1 can say that more precisely than the over/under 2.5 market. If you think both sides will score but neither will run away with it, 1–1 and 2–1 sit close to a both teams to score bet with a much bigger payout.

It makes less sense as a lottery ticket on 4–3 or 5–0, where the margin is widest, or as a habit on every match. Small stakes and a model to check the price against are the minimum. A bet builder that pairs a result with a goal line often says something similar with a lower margin.

Long odds and document checks

Sites that skip ID at sign-up still keep the right to ask for documents later, and a large withdrawal is a common point for that request. A correct score bet produces exactly the kind of return that can reach a withdrawal threshold in one go: a 20.00 stake on a 1–1 at 6.76 returns 135.20, and the same stake on a longer score returns several hundred. Before staking, know the site's withdrawal limits and its wording on verification. Our pages on fast withdrawals and on the thinnest margins are a starting point, and the methodology explains how we rate them.

Keep stakes within what you would accept losing, as most correct score bets do lose. Our responsible gambling page sets out the tools sites provide.

If you want to place these bets without sending documents first, start with our list of no verification betting sites.

For the markets with lower margins, see the rest of our betting guides.

Bitcasino.ioSportsbook

US$2,500When ID can be asked

Open