Advanced Metrics or Analytics are making their marks into the football mainstream. Teams are looking for strategies to get the better of their opponents. Although football can sometimes be viewed as a game of luck, metrics have played important roles in helping coaches and executives devise strategies of winning behind the scenes. Though critics of these analytics in sports are saying they will affect the flow of the game, thereby ridiculing its prominence. My opinion still remain that the tool is as good as the handyman. So let’s dive into the subject matter. One of such metrics is the expected goals (xG)
What is Expected goals?
Expected goals is a statistic used to work out how many goals should be scored in a match, with every shot taken given an ‘expected goal’ value based on the difficulty of the attempt.
It assigns value to the chances of a shot resulting to a goal. It crunches data of thousands of historical shots taken, filtering them against factors such as distance, type of shots, type of pass and number of defenders between the shooter and the goal.
Based on these number of factors including distance from goal, type of shot and number of defenders, the ‘expected goal’ value reveals the likelihood a specific shot will end in a goal; the higher the value, the more likely a goal should be scored from that shot. This creates a percentage chance, on average of a particular shot going in with all these factors being put in consideration.
The ‘expected goal’ value of every shot in a game is used to calculate the ‘expected goals’ (xG) of a match. It takes into consideration the position of a shot, the body part used to take it and the proximity of defenders. It could be used to assess the team and individual player’s performance
It is usually expressed as a number between 0 and 1, with 1 being the maximum and representing a particular goal. A chance of 0.5 expected goals means a goal can result on one out of every two occasions.
So how does xG practically works?
Expected Goals focuses more on the “quality of shots” rather than on “the number of shots”. Let’s take an example with Dortmund. Dortmund played badly last month as they lost points frequently. But one can be tempted to presume that they didn’t play that bad as shown, just that they were unlucky in front of goals. Data showed that they have the same expected goals with Bayern Munich. We can deduce that Dortmund has been very unlucky in front of goals or they have problem in finishing not in creating chance. So one can predict that Dortmund will rise again when they have luck and confidence. And it happened as evident against Hoffenheim when Christian Pulisic won the match at the 89th minutes.
We can also use this statistics to predict the outcome of a match between Team A and Team B. For example if A has xG of 0.3 per match and B has 0.83 per match. Even if A has higher number of shots, has better possessions or not with other teams, team B will have more likelihood to win in an actual game.
However, football is an unpredictable game. We can all remember the champions league game between Barcelona and Chelsea in 2012. Barça had penalties, tons of shots but couldn’t score that most important single goal. Suddenly, Torres became a savior, running through a solo effort to round the keeper and scored. You cannot predict that. By using xG, you can know when a team is over achieving or under achieving, lucky or unlucky and predict their result in the long run.
In our subsequent articles, we will look at more interesting topics regarding expected goals such as its usefulness to the football gambler, how you can annex it to win more bets, helping you avoid scams online like fixed matches, its controversial stats and shortcomings.