xGtoday provides an objective rating system for European football clubs based on Expected Goals (xG), by condensing their performance into a single, universal rating on a 100 point scale.

These ratings are extremely useful tools for predicting future matches. You will see predictions throughout the site for upcoming matches and - to keep us honest - the historical predictions from before the match took place as well.

When teams play matches where xG data is available, xGtoday feeds the information into a modified Elo-based formula to compute the ratings that drive all the predictions on the site

xGtoday offers something for all sorts of football fans. Using xG today allows you to:

• Use xG analysis to identify potential value in betting markets
• Have some evidence to back up your claims when you're debating which team is better and why
• Compare different teams and leagues who might not face each other often with a simple rating that tells you which team would be more likely to win
• Dive deep into the xG stats by team, competition and country for your own analysis

Expected Goals, or xG for short, is a method of quantifying the quality of a football team's attempts at goal.

xG models use an enormous amount of historical data in order to estimate the probability of a given shot resulting in a goal. The models take into account the shot distance and angle, the positioning of the other players, the height of ball, the body part used (head, foot, etc.) and other factors.

Using xG provides a clearer insight into the quality of a team than match results alone. Since goals are relatively rare events, the result of a given match or run of matches may not reflect how well each team played. Using xG can help cut through the noise and more accurately assess the quality of a given team's chances.

All shots are not created equal. A shot from 40 yards has a very low probability of scoring. A tap-in near the goal line when the keeper is out of position is nearly certain to be a goal. Penalties are converted approximately 79% of the time. xG captures these differences while merely counting shots (or even shots on target) does not.

Summing the xG across a match or a season provides a better assessment of the quality of a team than the score of a match or even the league table. As a result, xG is more predictive of future results than actual goals.

Please note: Our analysis is for informational purposes. We do not guarantee outcomes, and any betting decisions are made at your own risk. Always gamble responsibly, do not bet more than you can afford to lose, and follow the law in your locality. If you need support, please visit gambleaware.org

xG is a highly predictive factor for future matches. xGtoday provides unbiased ratings and predictions based on xG performance which can be useful when assessing matches. With that said, it is important to note what xGtoday does and does not do.

xGtoday predictions and ratings do:

• Accurately portray the strength and form of the teams using xG.
• Account for home field advantage and how it varies in different competitions.
• Distill a match into home/draw/away probabilities and assess them against the current odds on offer.

xGtoday predictions and ratings do not:

• Adjust for injured players - if a team is missing a key player or players they may be expected to perform worse than their current rating might suggest.
• Account for "being clinical". Obviously not every player/team is going to score exactly as many goals as their xG would suggest. Given enough attempts, however, most players, including the greats do come very, very close. Famously, Lionel Messi does not.
• Guarantee successful bets. Anything can happen in a football match.

Elo describes a rating system developed by Arpad Elo for the purposes of evaluating chess players.

You may have noticed it depicted in the film The Social Network.

xGtoday has taken this system and modified it to apply to football matches using xG as the primary input.

xGtoday uses a modified Elo-based system to determine the ratings for each team and then adjusts them to a 100 point scale.

The exact methodology for calculating the ratings is proprietary.

This is a challenging aspect of devising a rating system! xGtoday relies on a combination of a few basic rules to determine the starting ratings.

If we have no rating for a team and we do have ratings for other teams in the competition they are joining, they get the same rating as the worst team in the competition. This makes intuitive sense - we generally expect promoted sides to be in and amongst those who barely stayed up last season.

When we didn't have any ratings at all back at the beginning, we used a concept called bootstrapping to assign the ratings for the teams so they would reach their "true" rating faster.

A team's rating is considered "provisional" until they have reached 30 matches of xG data.

The rating of a given team is a numerical representation of their performance based on xG, adjusted for home field and strength of opposition.

As a result, xGtoday ratings are completely free of bias and accurately represent xG performance.

We're not a data provider - we're an analytics platform. There are plenty of places to find an xG table or the match-by-match xG. We add value by using xG and our proprietary rating system to give you actionable insights.