Betting Strategy – ELO Ratings Guide
Why Ratings are so important to finding football betting profits
From Chess To Football: What is Elo?
Surprisingly in our world of endless acronyms (NFL, MIA… OMG), the origin of “Elo” is not based on an acronym at all. It is actually nothing more complicated than the surname of the Hungarian gentleman who derived the system; namely, Arpad Elo.
As a physics professor and chess master, Arpad was a pretty smart guy. It’s therefore not surprising that his system – initially intended for use in chess – has been adapted and applied to a number of global sports, including of course, football.
Much like any other ratings system, Elo is effectively a method of assigning a numerical value to the ability of a team or individual in a sports competition. The higher the rating, the greater the ability.
One benefit Elo ratings do have is that they are constantly evolving, with adjustments made to an individual team’s score to reflect each and every result of matches between rated teams.
This is typically achieved by each team putting 5% of its total Elo rating at risk for a given match. The winner takes the points from the other team, and in the event of a draw, the total number of points at risk is split evenly.
Elo Ratings: An Example
In February 2018, Chelsea played Newcastle in the F.A. Cup at Stamford Bridge. Currently sitting second in the Premier League, Chelsea had an Elo score of 1915, whereas struggling Newcastle’s total was 1617. Using the 5% method, Chelsea would therefore have 96 Elo points at risk in this fixture, and Newcastle 81. The table below shows the new Elo rating for each side resulting from the three possible outcomes.
|Chelsea New Elo||Newcastle New Elo|
|Chelsea Win||+81 move to 1986||-81 move to 1536|
|Newcastle Win||-96 move to 1819||+96 move to 1713|
|Draw||-8 move to 1907||+8 move to 1625|
Elo: Home Bias
One frequently mentioned drawback of the Elo system is that it does not accurately reflect all of the factors which may impact the result of a football match. The most obvious example one being home advantage.
Whilst home advantage has been decreasing over the years – partly due to a tightening up of the tactical side of the game resulting in significantly more draws – home teams in the Premier League, for example, still win on average around 40% of matches, as opposed to 30% for the away side. Should a team therefore be awarded the same Elo increase for winning a home game as an away game? Possibly not.
An example of a way around this which is utilised by certain punters, is by increasing the number of points a home team has at risk for a given game. A common way of applying this theory is through adding a predetermined total or percentage to the home team’s score.
Converting Elo ratings into betting odds
The use of Elo ratings as a tool for determining the superiority of one team over another is useful, but from a betting perspective, how does this help the punter identify value bets? The answer lies with the conversion of Elo ratings to betting odds.
Working it out:
There are a number of methods for doing this, but most are based around beginning with the base percentages of 40% home wins, 30% away wins and 30% draws. Adjustments are then made for the difference in Elo score between the two sides in question.
One commonly used approach is to increase the win % chance of the higher rated side by 1% for every 10 Elo points they are superior. Let’s use our Chelsea v Newcastle game as an example.
Chelsea are 298 points superior to Newcastle, meaning we need to add 30% to their 40% winning chance as the home team. (298/10 = 29.8, rounded up to 30). This gives Chelsea a total winning chance of 70%.
A popular draw chance adjustment is to remove 1% from the 30% average draw chance for each 20 points one team is superior over the other, but ignoring the first 100 points of superiority. In our example this would therefore be:
(Chelsea’s total superiority of 298 -100)/20 = 9.9.
Rounding this up to 10% and subtracting from 30% leaves us with a draw percentage chance of 20%.
Newcastle’s win percentage chance is simply the remaining total once the Chelsea win and draw totals have been subtracted from 100. 100-70-20 = 10% for the Newcastle win.
To convert these percentage chances into odds, simply first convert into decimal form and then divide into 1. This gives us the following odds.
Chelsea win: 1/0.7 = 1.43
Draw: 1/0.2 = 5.00
Newcastle win: 1/0.1 = 10.00
The above outlines the basic method, but of course individuals may choose to make adjustments for examples such as home bias.
Calculating odds from Elo ratings gives you a solid scientific point of reference when assessing the odds on offer by the bookmakers, and thus identifying any potential value.
Elo: Other Variables
Clearly there are many other elements at play for any individual game, e.g. is it a cup or a league game? Injuries, transfers, new managers, etc, etc.
It is important to weigh all these factors when making decisions, but this does not negate the value of having the empirically based ability assessment tool offered by Elo. In combination with other factors, Elo provides an excellent foundation for shrewd bet selection.
So there you have it, a beginners guide to Elo. As with most things in life, it's probably best to take it slow and err on the side of caution to begin with. Learn what you're doing with small betting pots and build from there as your knowledge and confidence grows. Good luck!
Weekly Best Bets:
It is designed to be a really easy and enjoyable service to follow along with just a couple of emails a week and a handful of well researched and supported bets recommended to you each week. If this sounds like something you'd enjoy why not join us for the full European season for just £9.95 a month or £77 for the whole season.
Step by step guiding to building a successful and profitable betting portfolio. Download for free here
Jon is the Founder and Chief Tipster at Football Advisor and Predictoloy. He started life as data analyst in the digital marketing field before find his true calling in the world of Football and Horse Racing Betting.
Jon has been sharing his professional expertise since 2009 and specialises in using objective data analysis and subjective experience of betting built up over more than a decade of professional betting.
In 2014, Jon also launched (and continues to run) the trusted Football Advisor service service which provides a variety of football and horse racing betting models and portfolios. A few years later, Jon launched the Predictology platform which is the worlds first betting system builder and analyser covering a wealth of football betting related statistics covering more than 200,000 matches.
Jon has also lent his knowledge and expertise to several of the trusted Premium Services offered by the respected Secret Betting Club, including Football Lay Profits. Racing Bet Profits and, most recently, Racing Lay Profits.