Deriving it on a coin flip
#A fair coin priced fairly pays 2.00 on each side: implied probabilities of 50% and 50%, summing to exactly 100%, and nobody earns anything over time. No real bookmaker prices it that way. The standard price is 1.91 both sides, sometimes written as -110 and -110.
At 1.91, each side implies 1/1.91, or 52.4%. The market sums to 104.7%: the book has sold 104.7% of probability on an event holding only 100%. Betting both sides to guarantee every outcome costs 104.7 to receive 100, and that 4.5% gap is the margin, collected from the losing side whichever way the coin lands.
The same arithmetic on a real market
#A football match priced at 2.50 for the home side, 3.30 for the draw and 2.90 for the away side implies 40% plus 30.3% plus 34.5%: a total of 104.8%, so a margin of roughly 4.8% on the ordinary three-way market of a major league.
The number varies where players rarely look. Headline markets on big competitions run the thinnest margins, because they are the prices everyone compares. Obscure leagues, side markets and combination bets carry considerably more, precisely because nobody is checking. The formula above works on any of them, which is the point of knowing it.
Margin and the house edge, same role, different mechanics
#The margin plays the role the house edge plays in a casino: a structural percentage that makes the average bet lose. The mechanics differ in one respect: a roulette wheel carries a fixed edge set by the rules, while a bookmaker embeds its margin in prices it also uses to manage risk, which is why odds move and why the margin is not spread evenly across outcomes. Longshots typically carry more of it than favourites.
The practical reading is unchanged: the average bettor pays the margin on every stake, exactly as the average casino player pays the edge, and over volume the arithmetic converges the same way.
The one habit that provably pays
#Because margins differ between bookmakers and the calculation above takes seconds, comparing prices is the rare betting habit whose benefit is a theorem rather than a hope: the best available price on the same outcome is, by definition, the smallest margin paid. Over hundreds of bets the difference between taking 2.50 and 2.40 on the same selections is substantial.
It needs saying plainly: shopping for prices reduces the cost of betting, it does not make betting profitable. The best price on the market is still, on average, a losing one. It is the same relationship as choosing a French roulette table over an American one, better, and still a cost.
Questions, answered
How do I compute a market's margin myself?
Convert every outcome's decimal odds to implied probability, 1 divided by the odds, and sum them. Subtract 100%: what remains is the margin. On 1.91/1.91 it is 4.7%; on 2.50/3.30/2.90 it is 4.8%. Any market yields to the same three steps.
Why are margins lower on big matches?
Competition. Prominent markets are the ones bettors compare across sites, so books compress their margins there and recover them on side markets, obscure competitions and accumulators, where comparison is rarer and the arithmetic is checked less.
Is a low-margin bookmaker a profitable one to use?
It is a cheaper one, which is strictly better and still not profitable on average: a 2% margin is a smaller expected loss than a 6% one, not a gain. Profit requires your probabilities to beat the market's after the margin, which is a different and much harder claim.
Related