Why Correct Score Bets Are More Expensive Than They Look

A 2–1 prediction at odds of 9.00 feels generous. You are not asking for the lottery numbers, only a plausible football score.

The difficulty is hidden in the alternatives. The match can finish 1–0, 1–1, 2–0, 2–2, 3–1, 0–0 or in dozens of less attractive ways. Correct-score markets spread the bookmaker margin across a wide field of outcomes, and the eye is naturally drawn to the one story that feels most likely.

Plausible is not the same as probable

Suppose you believe the home team will win and expect goals. A 2–1 scoreline fits the narrative perfectly.

But “home win with goals” contains many results. The exact score must defeat every neighbour. A late third goal ruins 2–1 even though the broader match analysis was correct. A missed penalty ruins it. One defensive error turns it into 2–2.

This is why correct-score bets pay more than ordinary match-result bets. The prediction is narrower.

The margin is harder to see

In a two-outcome market, you can convert both prices into implied probabilities and quickly see how much they exceed 100%.

A correct-score board may list twenty or thirty outcomes plus “any other score.” Calculating the overround requires converting every price and adding them together.

That inconvenience matters. A customer sees 9.00 and imagines the potential return. The operator sees a complete probability book with margin distributed across every scoreline.

Longshot bias can add another cost. Research has long documented that bettors may accept relatively poor value on unlikely outcomes with exciting payouts. [1]

A simple illustration

Imagine a deliberately simplified market with only five outcomes:

• 1–0 at 6.00: 16.7% implied
• 1–1 at 6.50: 15.4%
• 2–0 at 8.00: 12.5%
• 2–1 at 9.00: 11.1%
• Everything else at 1.80: 55.6%

The implied probabilities total 111.3%. The excess above 100% represents the built-in margin before any adjustment for how it is distributed.

A real market has more lines, but the principle is the same.

Scoreline models can still be useful

Correct-score probabilities are not impossible to estimate. Analysts use expected goals, attack and defence strength, line-ups and models such as Poisson distributions.

The model might estimate:

• 1–1 at 12%
• 2–1 at 10%
• 1–0 at 9%

If the available price implies a lower probability than the model, there may be value. The problem is uncertainty. Small changes in expected goals can redistribute probability across many scorelines.

A model that is broadly right about the match can still be wrong about the exact cell.

Why correct scores work so well socially

A successful 3–2 prediction looks prophetic. It is easy to post and difficult for viewers to forget.

The failed 0–0, 1–0 and 2–2 predictions from previous weeks vanish into the feed. This creates a false impression that exact-score insight is more common than it is.

A credible record must include every selection, not only the beautiful ones.

My view

Correct-score bets are not automatically bad. They are simply expensive places to be vague.

“Chelsea should win 2–1” is a match opinion. “The true probability of exactly 2–1 is higher than the 11.1% implied by odds of 9.00” is a betting claim. The second requires much more work.

If the analysis mainly supports a home win, betting the exact score adds precision the evidence may not deserve. The higher payout is compensation for all the ways a basically correct idea can still lose.

Before taking the number, calculate [implied probability](/articles/calculate-implied-probability-betting-odds) and remember that attractive long odds are not the same as generous odds.