Why compare odds across bookmakers?
Different bookmakers offer different odds for the same event. These differences arise because each bookmaker uses its own models, data sources, and market positioning to estimate probabilities and set margins. The differences are often small — a few hundredths in the decimal odd — but their cumulative effect on a player's long-term profitability is substantial.
A player who always bets at the first bookmaker he checks is systematically accepting odds that may be worse than what is available elsewhere. A player who compares odds across many bookmakers and places each bet at the best available odd is, by definition, maximizing his expected value for every bet.
The impact of small differences
Consider a bet with an estimated actual probability of 50%. Suppose bookmaker A offers a decimal odd of 1.95 and bookmaker B offers 2.05 for the same option.
At bookmaker A:
e = .50 · (1.95 − 1) − .50 = .50 · .95 − .50 = .475 − .50 = −.025 = −2.5%
At bookmaker B:
e = .50 · (2.05 − 1) − .50 = .50 · 1.05 − .50 = .525 − .50 = .025 = 2.5%
The difference of 0.10 in the decimal odd has transformed the bet from negative expected value (−2.5%) to positive expected value (+2.5%). Over 1000 bets of ¤10 each, this difference amounts to:
At bookmaker A: 1000 · ¤10 · (−.025) = −¤250
At bookmaker B: 1000 · ¤10 · .025 = +¤250
A total difference of ¤500, from a seemingly minor difference of 0.10 in the odd.
Best available odd
The best available odd for a given option is the highest decimal odd offered by any bookmaker for that option. Since a higher odd means a higher potential return per unit wagered, the best available odd always produces the highest expected value for any given probability estimate.
In practice, finding the best available odd requires access to the odds of multiple bookmakers. If 30 bookmakers are offering odds on the same event, the best odd for each option is the maximum across all 30. These maximum odds are the ones a rational player should use when evaluating and placing bets.
The difference between the best and worst available odds for the same option can be substantial. In less liquid markets (lower divisions, minor sports, proposition bets), the range tends to be wider, because bookmakers have less information and fewer market participants to correct their estimates. In highly liquid markets (major leagues, popular events), the range tends to be narrower, as competition between bookmakers drives the odds closer together.
Effective margin reduction
Each individual bookmaker embeds a margin in its odds. However, when a player selects the best odd for each option across different bookmakers, the effective margin — computed from the best odds across the full market — is lower than the margin of any individual bookmaker.
For example, in a two-option market:
Bookmaker A: option 1 at 1.85, option 2 at 1.95
Bookmaker B: option 1 at 1.95, option 2 at 1.85
Bookmaker A's margin: (1/1.85 + 1/1.95) − 1 = (.5405 + .5128) − 1 = .0533 = 5.33%
Bookmaker B's margin: (1/1.95 + 1/1.85) − 1 = (.5128 + .5405) − 1 = .0533 = 5.33%
Best available odds: option 1 at 1.95 (from B), option 2 at 1.95 (from A)
Effective margin: (1/1.95 + 1/1.95) − 1 = (.5128 + .5128) − 1 = .0256 = 2.56%
By comparing odds, the player has effectively halved the margin he is facing. In some cases, when the discrepancies between bookmakers are large enough, the effective margin can drop below zero, creating an arbitrage opportunity.
Consensus probability from odds comparison
Comparing odds across many bookmakers also serves an analytical purpose beyond finding the best price. The distribution of odds across bookmakers provides information about the market's collective estimate of the probabilities.
Statistical measures such as the mean, median, and weighted mean of the implied probabilities across all bookmakers produce a consensus probability — an aggregate estimate that tends to be more accurate than any individual bookmaker's estimate. This consensus can be used as a reference for evaluating whether a particular bookmaker's odd represents positive expected value.
For instance, if the consensus probability for an option is 45%, and one bookmaker is offering an odd that implies 38%, this 7-percentage-point discrepancy suggests that the bookmaker may be underestimating the probability of that option. A bet at that odd, if the consensus is accurate, would have positive expected value.
Practical considerations
Effective odds comparison requires maintaining accounts at multiple bookmakers and having funds distributed across them. This is a logistical constraint: the player cannot always bet at the best available odd if he does not have an account or sufficient balance at the bookmaker offering it.
Additionally, the best available odd may be offered by a bookmaker that imposes low maximum bet limits, or that is known to restrict the accounts of winning players. The practical best available odd is therefore the best odd at a bookmaker where the player can actually place the desired wager.
Despite these constraints, odds comparison is one of the simplest and most effective ways for a player to improve his long-term results. No amount of analytical sophistication in probability estimation can compensate for consistently betting at suboptimal odds.