Correct Score Betting on Football: Odds, Probability and Practical Approaches
Available in US
The first correct score bet I ever landed paid 9.00 on a 1-0. I staked £5 and felt like I’d cracked a code. It took another 14 losing correct score bets before the next one came in, by which point I’d given back more than I’d won. That pattern — occasional thrilling payouts separated by long losing streaks — is the fundamental reality of correct score betting. The market rewards patience and punishes impulse, and understanding the probability structure behind the prices is the only way to make it work over time.
How Bookmakers Price Correct Score Markets
Every correct score price is derived from the bookmaker’s estimated probability of each team scoring a specific number of goals. The most common underlying model is Poisson-based: given expected goals for each side, the model generates a probability for every combination of scorelines from 0-0 up to, typically, 5-5 or beyond. Roughly 290 million online bets are placed each month on real events across the UK, and correct score markets absorb a meaningful share — partly because the high odds attract recreational money seeking big payouts from small stakes.
The margins on correct score are wider than on most football markets. Where a 1X2 market might carry a 104% overround, correct score overrounds frequently exceed 115-120%. The bookmaker builds in a substantial cushion because the per-outcome probabilities are small — a typical correct score has a 5-12% probability — and pricing errors on low-probability events are harder to detect and exploit.
The most commonly backed scoreline in the Premier League is 1-1, followed by 1-0, 2-1 and 0-0. These four results account for roughly 40-45% of all Premier League outcomes in a typical season. The remaining 55-60% of results are spread across dozens of other scorelines, each with a small individual probability. This concentration at the top means the shortest-priced correct scores are also the most likely — but “most likely” in correct score terms still means a sub-15% chance.
Building a Correct Score Model With Expected Goals
I built my first correct score model using nothing more than a spreadsheet and publicly available xG data. The logic follows directly from a Poisson-based approach: estimate the expected goals for each team, then use the Poisson distribution to calculate the probability of each team scoring 0, 1, 2, 3, 4 or 5+ goals. Multiply the individual goal probabilities together for each combination and you have a probability matrix for every possible scoreline.
For a match where the home side has an xG of 1.6 and the away side has an xG of 1.0, the Poisson model assigns the following approximate probabilities to the most common outcomes: 1-0 at 9.8%, 1-1 at 9.8%, 0-0 at 6.1%, 2-1 at 9.8%, 2-0 at 7.8%. Convert those to fair decimal odds: 1-0 at 10.20, 1-1 at 10.20, 0-0 at 16.39. If the bookmaker is offering 8.00 on 1-1, the price implies a probability of 12.5% — higher than the model’s 9.8%. That selection is negative expected value. If the bookmaker offers 12.00 on 0-0, the implied probability is 8.3% versus the model’s 6.1% — also negative EV. But if another bookmaker offers 11.00 on 1-0 against a model fair price of 10.20, the margin is tight but potentially positive.
The UK online sports betting market generated $8,284 million in revenue during 2025. The correct score segment contributes partly because the wide margins give bookmakers comfortable profits even when the occasional big payout lands. For bettors, the wide margins mean you need sharper probability estimates to find genuine value — and the Poisson model, while imperfect, provides a framework that outperforms guessing.
Practical Approaches to Correct Score Betting
Three approaches have worked for me across multiple seasons. First, focusing on low-scoring fixtures. Matches where both teams have low xG (below 1.2 per side) produce a narrower range of likely scorelines — 0-0, 1-0, 0-1 and 1-1 collectively account for 50%+ of outcomes. This concentration means a small number of correct score selections covers a meaningful share of the probability space, and the prices on these outcomes tend to be shorter but more frequently rewarded.
Second, combining correct scores into small cover bets. Rather than staking £10 on a single 1-0, I might stake £3 each on 1-0, 2-0 and 2-1 as three separate bets. If any of them lands, the payout exceeds the combined stake. This approach reduces variance significantly — you’re still betting on low-probability outcomes, but you’re covering a cluster of related results rather than a single precise scoreline.
Third, using correct score as an in-play market. After 60 minutes of a goalless match, the range of likely final scorelines narrows dramatically. 0-0, 1-0 and 0-1 dominate the probability space, and the live prices on these outcomes have already adjusted downward from their pre-match levels. If your pre-match analysis pointed to a low-scoring fixture and the scoreline at 60 minutes confirms that thesis, the in-play correct score market often offers better value than the pre-match price on the same selection.
One approach I have seen work for disciplined bettors is the “correct score portfolio” — treating each match as a mini portfolio of three to five scoreline bets that collectively cover 30-40% of the probability space. The combined stake is managed as a single position, and the target is a positive return across the portfolio rather than a single winning selection. This reframes correct score from a lottery to a probability exercise, and it demands the same xG-based modelling and stake discipline that any other systematic approach requires.
Correct score will never be a high-volume, high-frequency market for systematic bettors. The hit rate is too low and the variance too high. But as an occasional supplement to core markets — particularly in low-scoring fixtures where the probability space is concentrated — it offers payouts that no other standard market can match. The key is treating it as a calculated allocation within a broader strategy, not as a standalone approach. Stake small, model the probabilities, and accept that most correct score bets will lose. The ones that land make the maths work.
What is the most common correct score in the Premier League?
The most frequently occurring scoreline is 1-1, followed closely by 1-0, 2-1 and 0-0. These four results typically account for 40-45% of all Premier League match outcomes across a season. Despite being the most common, each individual scoreline still has a relatively low probability of under 15%.
Is correct score betting profitable long-term?
It can be, but the high variance and wide bookmaker margins make it difficult. The overround on correct score markets frequently exceeds 115-120%, meaning you need significantly sharper probability estimates than the bookmaker to find positive expected value. Most profitable correct score bettors use Poisson-based models and focus on a narrow range of low-scoring fixtures where the probability space is most concentrated.
Prepared by the Football Betting Sites editorial staff.
