The Bookmaker Does Not Need to Know Who Will Win

The popular image of a bookmaker is a professional fortune-teller with better data.

Data matters, but the business does not require supernatural prediction. A bookmaker can offer slightly worse than fair prices across every outcome and earn from the gap.

Think of a currency exchange desk. It does not need to predict where the euro goes next to earn money from the spread between buying and selling. A sportsbook’s margin works in a similar spirit.

A fair coin

A fair coin has two outcomes, each with a 50% chance.

Fair decimal odds would be 2.00 on heads and 2.00 on tails.

A bookmaker might offer 1.91 on each side.

The implied probability of 1.91 is:

1 ÷ 1.91 = 52.36%

Add both sides:

52.36% + 52.36% = 104.72%

The market is priced to 104.72%, not 100%. The extra 4.72% is the overround.

That does not translate perfectly into guaranteed profit because betting volumes are uneven and outcomes create risk. It does show that customers are buying probabilities at a markup.

A football example

Imagine a match priced as:

• Home win: 2.20
• Draw: 3.40
• Away win: 3.30

The implied probabilities are:

• Home: 45.45%
• Draw: 29.41%
• Away: 30.30%

Total:

105.16%

The bookmaker has created a 5.16% overround.

To estimate the margin-free probabilities, divide each number by 105.16:

• Home: 43.22%
• Draw: 27.97%
• Away: 28.81%

The offered prices are shorter than those fair estimates.

Does the bookmaker balance every book?

Not always. Modern sportsbooks may deliberately carry uneven exposure when their models and risk limits support it.

The classic story says bookmakers adjust prices until equal money sits on every outcome, guaranteeing profit. Real sportsbooks may accept unbalanced exposure when their models, customer data and risk limits support it.

They may also move prices because sharp bettors reveal information, injuries change the event or competing bookmakers update their markets.

The margin remains central even when the liabilities are not perfectly balanced.

Where margins become expensive

Major football match markets tend to attract competition and large betting volumes, which can produce tighter prices.

Niche props, same-game combinations and novelty markets may carry much larger margins because customers have fewer comparison points and the probabilities are harder to assess.

Research on request-a-bet products found substantially higher bookmaker margins than on conventional markets. [1]

The convenience of choosing an amusing custom outcome can be expensive. The bet feels personal; the price is still industrial.

Why prices move even when the teams do not

Odds can change before any new injury, lineup or weather report appears.

A respected customer may place a large bet. Another bookmaker may move first. The operator may be carrying more liability than it wants on one outcome. A market-making model may update after thousands of smaller transactions.

The movement does not always mean a secret fact has entered the room. Sometimes the shop is changing the price because customers are queuing at one counter.

This is useful when reading steam moves. A shortening price says the market is willing to buy the outcome more aggressively. It does not tell you whether joining late is sensible. The information may already be fully charged into the new odds.

Why comparing odds matters

Suppose one bookmaker offers 1.85 and another offers 1.95 on the same selection.

On a €100 winning bet:

• 1.85 returns €185
• 1.95 returns €195

That €10 difference seems small once. Across hundreds of bets, it becomes the difference between paying a high commission and a lower one.

Bettors often spend hours predicting a match and seconds accepting the first price. That is like researching a car for a month and refusing to compare dealers.

Margin is not evenly distributed

A 105% market does not necessarily mean every outcome carries the same markup.

Longshots may be priced less generously than favourites. Academic work has documented favourite-longshot bias in multiple betting markets, meaning low-probability outcomes often generate worse average returns. [2]

This is one reason a simple proportional margin calculation is only an estimate.

My verdict

The bookmaker’s first advantage is not knowing the score. It is selling the probabilities for more than 100%.

Good prediction can help the operator manage risk. The margin pays the rent.

Before deciding whether a team wins, calculate what the market is charging. A correct prediction at a bad price can still be a bad bet. That sentence is less exciting than a tip. It is also more useful.