True Odds Calculation Methods
One of the key requirements in matched betting (or any kind of value betting) is the ability to calculate the true odds of an outcome. If we know the true odds we can calculate the expected value of any bet we place.
In some cases it easy to find the true odds. For example, if there is an efficient Betfair market for the outcome we can safely assume that the Betfair odds are equal to the true odds. This is due to the wisdom of the crowd.
However, there are often times when we cannot use Betfair as a proxy and we need to look for alternatives.
Bookmaker odds can help us get in the right ballpark (especially if they are a 'sharp' bookie like Pinnacle) but - due to the way that bookies make money - we know that they can't be considered true odds. To make them more accurate we need to remove the bookies margin (a.k.a. overround, vig, or juice).
This is where things can get a bit tricky. It is easy for us to calculate how much margin a bookie has on a market, but it is impossible to know how they have distributed that margin between the different outcomes. As a result, different methods of calculating true odds will give different results.
Our True Odds Calculator uses four different methods so that you can see the range of possibilities.
Equal Margin (EM)
This is the most simplistic method of calculation. As the name suggests, the total margin for the market is split equally between each of the different outcomes.
This method works well on markets with evenly matched outcomes (e.g. a 'moneyline' market), but it is less reliable when the odds are skewed. Bookies tend to add more margin on the higher odds outcomes and less on the favourites. Which leads us to the next method...
Margin Proportional to Odds (MPTO)
This method distributes the margin using the formula:
TrueOdds = (n*BookieOdds)/(n-(margin*BookieOdds)
n represents the total number of outcomes in the market. So a selection with odds of 1.60 in a 3-way market with a total margin of 8% would have:
TrueOdds = (3*1.60)/(3-(0.08*1.60))
TrueOdds = 4.80/2.872
TrueOdds = 1.671
As a result of this distribution method, the True Odds of a favourite will be lower than if you used the EM method. This has been proven to be much more accurate.
Odds Ratio (OR) & Logarithmic (LOG) Methods
The final two methods are pretty complicated to calculate and, although their results vary slightly from MPTO, there is no strong evidence that they are more accurate over a large sample size.
For that reason (plus the fact that we're not fully qualified for that level of maths), we're not going to go into detail about them here.
However, you might want to check out this article by Joseph Buchdahl which goes into detail on page seven. The whole article is worth reading if you want to improve your understanding of odds and value betting.