Correct score betting is a bet on the exact score of a soccer match when normal time ends, such as 2-1 to the home side or 0-0. Goals in stoppage time count, while extra time and penalty shootouts do not, so a cup tie that stands at 1-1 after 90 minutes settles as 1-1 whoever goes through.
Every score is a long shot, and the market as a whole carries far more margin than the match result: the illustrative market below has 17 prices adding to 126.1%, against 105.4% for the same match's 1X2 market. Soccer betting maps the other match markets.
How a correct score bet settles
NSW's betting glossary of terms lists this bet as match score exact: picking the goals scored by both sides. Bookmakers licensed in NSW and authorised by the NSW Minister to take bets on declared events settle it under the Bookmaker Declared Betting Events Betting Rules. Unless agreed otherwise, the score counts at the end of normal time, "in most circumstances, ninety minutes plus injury time" (clause 5.2.7.3). That includes the stoppage time added to both halves. Other bookmakers write their own terms, so check yours.
With $10 on a 2-1 home win at an illustrative $9.50:
| What happens | Your bet |
|---|---|
| 2-1 when normal time ends, the second home goal at 90+5 | Wins: $10 x 9.50 = $95.00 |
| 2-1 at 90:00, then an away equaliser at 90+6 | Loses: the score is 2-2 |
| 1-1 after normal time, 2-1 after extra time | Loses: it settles as 1-1 |
| An away defender's own goal makes it 2-1 | Wins: an own goal counts for the other side |
| A third home goal is ruled out on review | Wins: under Law 10 a goal counts only if the scoring team committed no offence |
| Called off at 2-1 in the 70th minute | Refunded, result declared or not (clauses 5.2.7.1 and 5.2.7.2) |
| Called off at 3-1, no official result declared | Loses: already decided against you (clause 5.2.7.1) |
| Called off at 3-1, official result declared | Refunded: bets on the number of goals are void (clause 5.2.7.2) |
Scoreline chances from a Poisson model
A price is good or bad only against the score's real chance. A Poisson model is a common way to estimate it: give each side an expected number of goals, work out its chance of scoring 0, 1, 2 or 3, and multiply the two sides' chances.
The Poisson distribution guide builds it in full. Here it prices one illustrative match: the home side is expected to score 1.55 goals and the away side 1.20. The formula is P(k goals) = e^-l x l^k / k!, where l is expected goals, e is the constant 2.718 and k! is 1 x 2 x ... x k.
| Goals | Home side, 1.55 expected | Away side, 1.20 expected |
|---|---|---|
| 0 | 21.22% | 30.12% |
| 1 | 32.90% | 36.14% |
| 2 | 25.50% | 21.69% |
| 3 | 13.17% | 8.67% |
| 4 or more | 7.21% | 3.38% |
A score's chance is the product of its two cells: 2-1 = 25.50% x 36.14% = 9.22%, a fair price of 1 / 0.0922 = $10.85. Both sides score three or fewer with a chance of 92.793% x 96.623% = 89.66%, so the 16 scores from 0-0 to 3-3 hold 89.66% and any other score the remaining 10.34%. The plain model undercounts 0-0 and 1-1, which the Poisson guide adjusts for.
Correct score odds and the margin across the market
An illustrative bookmaker market for the same match lists every score from 0-0 to 3-3 plus any other score. Implied chance = 1 / price, and the return per $1 is the model's chance x the price: what each dollar staked brings back on average if the model is right, worked from the unrounded model chances.
| Score | Model chance | Offered | Implied by the offer | Return per $1 |
|---|---|---|---|---|
| 1-1 | 11.89% | $7.00 | 14.29% | 83.2c |
| Any other score | 10.34% | $7.50 | 13.33% | 77.6c |
| 1-0 | 9.91% | $8.50 | 11.76% | 84.2c |
| 2-1 | 9.22% | $9.50 | 10.53% | 87.5c |
| 2-0 | 7.68% | $10.00 | 10.00% | 76.8c |
| 0-1 | 7.67% | $10.00 | 10.00% | 76.7c |
| 1-2 | 7.13% | $11.00 | 9.09% | 78.5c |
| 0-0 | 6.39% | $12.00 | 8.33% | 76.7c |
| 2-2 | 5.53% | $13.00 | 7.69% | 71.9c |
| 3-1 | 4.76% | $17.00 | 5.88% | 80.9c |
| 0-2 | 4.60% | $17.00 | 5.88% | 78.2c |
| 3-0 | 3.97% | $21.00 | 4.76% | 83.3c |
| 3-2 | 2.86% | $26.00 | 3.85% | 74.3c |
| 1-3 | 2.85% | $26.00 | 3.85% | 74.2c |
| 2-3 | 2.21% | $34.00 | 2.94% | 75.2c |
| 0-3 | 1.84% | $41.00 | 2.44% | 75.5c |
| 3-3 | 1.14% | $67.00 | 1.49% | 76.6c |
| Total | 126.12% |
Spread $1 across all 17 so that every score returns the same, and it comes back as 1 / 1.2612 = 79.3 cents: the bookmaker keeps about 20.7%. The same match's 1X2 at $2.05, $3.80 and $3.30 adds to 1 / 2.05 + 1 / 3.80 + 1 / 3.30 = 105.4% and keeps about 1 - 1 / 1.054 = 5.1%. Bookmaker margin explains both measures.
Why a correct score margin runs high
The cut is hard to see one price at a time. 1-1 at $7.00 against a fair 1 / 0.1189 = $8.41 reads like an ordinary short price, and only the sum of all 17 shows 126.1%; the bookmaker margin calculator adds up to 30 prices in one step. Here the three likeliest scores return 83.2c to 87.5c per $1, and most of the rest 71.9c to 78.5c.
A score is also harder to price than a result. Move the home side's expected goals by 0.2 and the home win shifts by about a tenth of its chance, while 3-1 and 0-0 shift by about a fifth, which gives a bookmaker more reason to leave room in every score's price:
| Home expected goals | Home win | 3-1 | 0-0 |
|---|---|---|---|
| 1.35 | 40.24% | 3.84% | 7.81% |
| 1.55 | 45.41% | 4.76% | 6.39% |
| 1.75 | 50.28% | 5.61% | 5.23% |
Each home win figure adds every score with the home side ahead, 4-0 and beyond included; 3-1 and 0-0 each multiply two cells.
The catch-all selection and how it settles
In the illustrative 17-price market, any other score collects every result in which a side reaches four goals. Like every selection, it is read when normal time ends, so 3-3 after 90 minutes and 4-3 after extra time is a loser. At 2.75 expected goals it holds 10.34%, second only to 1-1, and at $7.50 it returns 77.6c per $1 on the model. The any other score entry covers markets that list more scores.
Correct score in same game multis
A 2-1 home win already decides the result, the total and both teams to score, so adding those legs to a same game multi cannot make the ticket any less likely to win. Multiplied, 2-1 at $9.50, the home win at $2.05, over 2.5 goals at $1.85 and both teams to score at $1.75 (illustrative prices) come to 9.50 x 2.05 x 1.85 x 1.75 = $63.05. But the four land together exactly when 2-1 lands, 9.22% of the time in the model, so their fair price together is the fair price of 2-1 alone, $10.85. Judge any quote for the four against the 2-1 single.
Legs the score only partly decides are linked to it too. Say a home striker scores 35% of the side's goals (illustrative). In a 2-1 home win the striker scores with a chance of 1 - 0.65 x 0.65 = 57.75%, against 1 - e^-(0.35 x 1.55) = 41.87% across all matches, so the pair is 9.215% x 57.75% = 5.322% (a fair $18.79), not 9.215% x 41.87% = 3.858% ($25.92). Scorecast works a first goalscorer version, and how SGM odds are calculated covers the general case.
Risk: Betting involves risk. A same game multi loses if any one leg loses, and a 2-1 leg on its own loses about nine times in ten at these chances. See responsible gambling for limits and support.
Correct score strategy: what can and cannot work
Add up the market first: a score is worth a bet only where its price beats the fair price, and in the market above none does.
Compare one score across bookmakers. Say Bookmaker B offers 2-1 at $11.50 (illustrative): against the model's 9.22% that returns 0.0922 x 11.50 = $1.060 per $1, a 6.0% edge. If the home side's true figure is 1.35 rather than 1.55, 2-1 is 8.54% and the same price returns 0.0854 x 11.50 = 98.2 cents, so the edge is only as good as the inputs.
Covering several scores, known as dutching, does not dilute the margin. Splitting $100 across 1-0 at $8.50, 2-0 at $10.00 and 2-1 at $9.50 so that each returns the same takes stakes of $100 x (1 / price) / 0.32291, where 0.32291 = 1 / 8.50 + 1 / 10.00 + 1 / 9.50. Those stakes are $36.43, $30.97 and $32.60, each returning about $309.68. That is a combined $3.10 on a chance of 9.909% + 7.679% + 9.215% = 26.803%, so $100 brings back about 0.268 x $309.68 = $83 on average, against 0.4541 x $205 = $93.09 on the home win at $2.05.
Pick the market that asks your question: the 1X2 for the winner, a total for goals, and half time full time betting for how the match unfolds. Never use a correct score bet, or any other bet, to try to win back a loss: most correct score bets lose, and chasing losses only raises what is at risk.
Risk: Betting involves risk. A fair price from a model is only an estimate, a score priced above it still loses most of the time, and there is no guarantee of profit. See responsible gambling for limits and support.
Mistakes that cost money on correct score bets
| Mistake | What it costs, on the illustrative prices |
|---|---|
| Reading the score the wrong way round | A market may list scores by team name or home first, and a 2-1 home win loses a bet on 1-2 written home first |
| Pricing an SGM by multiplying legs | The four legs multiply to $63.05, but the ticket is exactly as likely as 2-1 alone, a fair $10.85 |
| Dutching scores to spread the risk | Three scores at a combined $3.10 return about $83 per $100 on average, against $93.09 on the home win |
Comparing the prices behind a scoreline
If you build your own scoreline model, it starts from the result and the goal total, the soccer markets B337's Terminal lines up: head to head, line and total prices from bookmakers side by side, with the best available price picked out and price history (flucs). Read across bookmakers, those prices let you work out the market's view of the two expected goals, to set beside your own.
Those prices refresh on a repeating cycle, so what the screen shows can lag the bookmaker's own price; confirm it in your account before you bet. A free account opens a limited view of the Terminal with live odds.
Risk: Betting involves risk. A correct score bet loses far more often than it wins, and the price you see may have moved by the time you bet. See responsible gambling for limits and support.