About Us
The idea for creating this type of website came from a deep love for sports and a passion for following football results. To help others and myself access the desired results more easily on sports betting platforms, I created a site that exclusively deals with football statistics related to goals and match outcomes.
I have also integrated an ELO rating system for football teams, which tracks their results in league and cup matches. I wish you all the best in using my website and reaching your desired outcomes. Good luck!
About ELO Rating System
Elo rating (aka. ELO score) is originating from chess, it was invented by a Hungarian mathematician called Arpad Elo to rank chess players. It has many features which make it possible to be used in other fields like sports, and video games to rank people who are competing against each other.
In a nutshell what it does is assign a score to each club, based on their strength. Using this single score we not only can rank players but we can tell the win and lose probability of a given match.
How it does it? That is what you will see in this article. In short, there is a calibration period, and at the start of this period, every club starts from the same ELO score, and then when they compete, based on the outcome of their match their scores get updated. The winner gain points the loser loses points. The points that are at stake for one match are dependent on the point difference between the two competing players. After some matches, players scores will converge to their true value.
Wikipedia provides a great summary if you would like to read more about the topic: ELO rating system
Calculating ELO
Writing the ELO calculation function
To be able to calculate the ELO score, we need two things:
1Calculate the expected probability of an outcome
Calculate the expected probability of an outcome, based on two input scores from the two opponents. For calculating this we need to use the following formula:
If club A has a rating of RA and club B a rating of RB, the exact formula for the expected score of club A is:
EA = 1 / (1 + 10((RB — RA) / weights))
Similarly, the expected score for club B is:
EB = 1 / (1 + 10((RA — RB) / weights))
(usually, the default weight is set to 400 but we can parametrize this too)
2Update the score with new values
Based on the actual result update the score with new values for each team. When a club's actual tournament scores exceed their expected scores, the Elo system takes this as evidence that the club's rating is too low, and needs to be adjusted upward. Similarly, when a club's actual tournament scores fall short of their expected scores, that club's rating is adjusted downward.
Elo's original suggestion, which is still widely used, was a simple linear adjustment proportional to the amount by which a club over-performed or under-performed their expected score.
The maximum possible adjustment per game is called the K-factor. This K-factor can be set to different values in chess it is set to 16 or 32 based on the level of the players, but we can set it to an arbitrary value that is close to this.
Suppose club A (with rating RA) was expected to score EA points but actually scored SA points. The formula for updating that club's rating is:
R'A = RA + K × (SA — EA)
This update can be performed after each game or each tournament, or after any suitable rating period.
Implementation in Our System
In our football statistics platform, we've implemented the ELO rating system to track team performance over time. This allows us to provide accurate predictions and rankings based on historical match data. You can explore team ELO ratings in the ELO Ratings section.