What are betting odds and how do bookmakers calculate them?
What are betting odds?
When a bettor asks “What are the odds?”, he is not asking what the probability is, he is asking how much he can win if he bets on it.
Betting odds are the way bookmakers let us know their “offer” or their “price” on an event happening. In statistics, odds reflect the likelihood that the event will take place. In sports betting, the odds are the ratio of payoff to the stake and do not necessarily reflect exactly the probabilities. Every time you place a bet with a bookmaker, you will be offered odds at the time, which means how much you can win. So betting odds allow you to calculate how much money you will win if you place a bet. The important thing to know is that betting odds do not reflect the real probability of something happening, just the bookmaker’s offer to you. Different bookies offer different odds. So you have to shop around for the best odds and you need to know if those odds are good value or not.
How to calculate odds from expected probability
In the simplest example, a coin toss, there are only 2 possible outcomes, heads or tails. So we can easily say the chances are 50-50, a 50% chance of heads and a 50% chance of tails. Of course, if we actually tossed a coin 10 times, we might get 7 heads and 3 tails, but this is just short term variance. If we tossed the coin 1 million times, we would reasonably expect nearly 500,000 of those tosses to be heads. The more times something happens, the closer to the theoretical probability we get.
In football, it is not as simple as a coin toss, but bookmakers create models to find patterns in the historical statistics and they get pretty good at predicting the probability of match outcomes. An extremely simplified way to create a model might be to look at recent form, season form, goals and league positions now and last season. We then create ratings for each of those statistics and apply weightings to them so that recent form affects the rating more than last season’s results. Finally we will end up with a rating that suggests how strong this team is now. We apply the same ratings system to the other team. We can then look at historical results between all teams in that league with similar ratings and get a probability for the results. That modelled probability might be a 33% chance of a home win, a 31% chance of a draw and a 36% chance of the away win. This model suggests that if time could be reset, and this match was played 100 times, that the home team would probably win 33 of those games, the away team would win 36 of those games and the other 31 games would be draws. But we don’t know which of those outcomes is going to come up in reality. This is a way that a model can calculate the probability of a football match’s result and the probability can be converted to odds.
Probability and American Odds, Fractional Odds or Decimal Odds
Bookmakers usually display odds in one of three formats, American odds, fractional and decimal odds. Online betting sites often have a settings option to allow switching between odds.
The odds of our coin toss and football game that we were imagining could be displayed like this table below.
|wdt_ID||Outcome||Probability||Fractional Odds||Decimal Odds||American Odds||Profit Potential|
The Bookmakers Profit Margin
Of course, bookmakers do not make models of football just for the sheer joy of seeing statistics in action, they do it to make money. If they let people bet on coin tosses with odds of 2.00, over the long term, they would make no money. They have to change things so that their payout is less than the payout would be at the true odds. In order to make money, they have to lower their odds to make sure that they make a profit regardless of the outcome. This is called the bookmaker margin. So a bookmaker might offer odds for the coin toss and football match looking more like this.
|wdt_ID||Outcome||Probability||Fractional Odds||Decimal Odds||American Odds||Profit Potential|
*For American odds under evens, they tend to show it as a negative number, which tells them how much they would have to wager to get 100 profit.
If £10 was bet on each outcome, over the long term, the bookmaker would make profit regardless of the outcome, because they had set the odds to add up to more than 100%. This is called the overround.
If the bookmaker runs this model on 100 events, they could reasonably expect this outcome.
Coin toss results from 100 events based on model probability, payouts based on bookmaker odds including margin:
Heads x 50 times: £1,000 in 100 bets taken, £900 paid out. £100 bookmaker profit.
Tails x 50 times: £1,000 in 100 bets taken, £900 paid out. £100 bookmaker profit.
£200 total bookmaker profit.
Football results from 100 games based on model probability, payouts based on bookmaker odds including margin:
Home win x 33 times: £1,000 in 100 bets taken, £957 paid out. £43 bookmaker profit.
Draw x 31 times: £1,000 in 100 bets taken, £930 paid out. £70 bookmaker profit.
Away win x 36 times: £1,000 in 100 bets taken, £947 paid out. £53 bookmaker profit.
£166 total bookmaker profit.
Reality versus Models: Why do the odds change up until the start of events?
Odds change because models are sometimes wrong, bookmakers want to balance their liabilities and sometimes team news can affect the perceived probabilities of outcomes.
In reality, bookmakers don’t get people nicely queuing up to bet equal amounts on all the outcomes and they tend to get more money coming in on some runners in an event rather than others. If they are getting near to the start of a match and they have lots of money coming in for the home win and nothing on the draw and away win, they are going to make a big loss if the game ends in a home win. Bookmakers may then choose to drop the odds for the home win to reduce payouts of any further bets taken, they may also increase the odds on the other outcomes to encourage more bets to come in and even out their liabilities. For this reason, bookmakers prices are different, because they are putting up odds to suit their profit targets.
Also, late news can affect perceived probability in the market, such as a star player being out or the manager suddenly putting out a team of youngsters because they have a more important game later in the week, then the odds will start to change as bettors seek to take advantage of the other team’s odds that seem more value.
And another reason for odds changing is that models are sometimes wrong. Bookmakers know that their models are not perfect, so they tend to lowball the market with their opening odds, and they slowly adjust them upwards later. By looking at the Betaminic statistics page we can see how most odds increase from their opening to closing values. They put out their opening odds and watch to see other bookmaker’s odds and also the market reaction to their odds. If they get people piling into one outcome in a way that is unusually higher than usual volumes, then they can guess that their odds are too high and then they will reduce them, but if no one is biting, and other bookmakers have higher odds, they can safely increase their odds without being caught out.
Soft and Sharp Bookmakers
Bookmakers that react quickly to the market are known as Sharp Bookmakers and those that are slower at changing their odds are known as Soft Bookmakers. Soft bookmakers tend to rely on larger margins which mean they do not need to be as worried about market changes because their odds are poor value from the beginning. They are not going to get dumped on since their odds are not the best to start with. Sharp bookmakers, such as Pinnacle, rely on a lower margin, but watch carefully for market changes and adjust their odds accordingly. Their odds are considered to have the lowest margins available, around 2%. This makes Pinnacle a good bookmaker to use, since we have less of a disadvantage to overcome from the start, a 2% margin rather 5-8% at others.
Finding Value in Model Mistakes
Bookmaker models are not perfect. If we can find which kinds of matches they are pricing incorrectly, we can take advantage of that and have a good chance to make profit. This requires the collection of masses of historical football data and corresponding bookmaker odds to go with it. Collecting and analysing all this data yourself is hard work. The Betamin Builder tool is an amazing data analysis tool from Betaminic.com that allows you to access and analyse over 100,000 matches from 50+ major leagues around the world going back to 2012. You can sign up and use it for FREE here.
The Betamin Builder allows you to filter football results by customizable parameters to see what the betting results would have been for games matching your selection criteria. In this way, we can apply our ideas about football, such as away favorites don’t win as much as we expect, and then see if backing the home/draw double chance bet would make a profit. In a matter of moments we can see if that was a profitable strategy.
Research your own strategies or follow existing winners
You can research your own strategies to follow or you can even see and follow strategies that others have found and decided to share. Betaminic also has strategies that have been up and running for over two years and still making profits.
Recommended Strategies Proven Over the Long Term
“Colossus 06 Scoring Away Overs”: 2 years running and 7.44% ROI
This strategy was publicly released by Betaminic over 2 years ago and is still going strong with 46 points profit from over 600 bets. It has over 100 followers. The ROI is nearly 8%. With the right staking plan, these profits can be multiplied even more. It bets on the over 2.5 goals market when selections match its criteria. It is highly recommended.
“DOUBE CHANCE” 1X Double Chance: 1.5 years running with a 15% yield
This strategy was publicly released in February 2019 and is also still going well. As suggested in the article, there are times when the away team is expected to win but does not get the 3 points they were expecting. This strategy finds those matches and has generated 50 points of profit with a near 15% ROI. You can also follow this by signing up for Betaminic.
“EPL Inversa”: Underdogs coming good
This is a great example of how big data backed strategies that focus on the statistics rather than the famous names can lead to profitable strategies. This strategy find teams that deserve to be respected more based on form, not pedigree, and a have a decent fighting chance against the favorites they face. Released in November 2018 it is still going well with 66 points profit and a 10% ROI.
“Pro-Home”: Odds on Favorites
For those who prefer lower odds bets, value can be found in the long term with this strategy. It also has been running publicly for 1.5 years and has a decent ROI of 4.87%. Lower odds stakes can be increased even more according to the Staking Plans book and profits can be multiplied more than higher odds strategies.
“Colossus 19 No Favourite Unders Draw”: Value in the draw market
The draw market is often overlooked and this strategy is not only one of the longest running but also has the most data to back its success up. After an amazing 1,282 bets it boasts a decent 3.82% ROI leading to 49 points of profit. It is a slow burner, but still going.
Betaminic allows you to research and create your own strategies for free in the Betamin Builder and also to easily follow other publicly shared strategies that have been running for a long period of time and have proved their patterns in the statistics are real inefficiencies in the bookmaker models, leading to long term opportunities for profitable strategies.
Join us here for free and experience big data betting at your fingertips.