All you need to know about xG data


Data-driven analytics is changing the game in the world of football. One of the hottest topics when it comes to assessing team performance is expected goals (xG). The concept was devised to help teams and analysts quantify the calibre of goalscoring chances created. Every chance is graded, with the most likely opportunities given the highest scores. At the end of a match, a team’s xG denotes how creative and how dominant they have been in the final third.

xG is increasingly used to pinpoint teams that may not be getting the rub of the green at present. They might not be taking their golden chances in games and eventually losing by a narrow margin. These teams are typically backed to improve over time, providing their xG data continues to demonstrate that goalscoring chances are flowing.

The variables used to rate goalscoring chances in xG


The quality of a goalscoring chance created is based on a string of variables. First and foremost, it looks at the type of assist. Was it a through ball from a teammate or was it a long throw that the defence simply misjudged? It also looks at the angle of a shot on goal. The narrower the angle, the less likely a chance is to be scored. xG also looks at the distance from which an attempt on goal is made. If a shot is fired at goal from six yards, it is likely to draw a much higher xG score than a speculative long-range effort from 30 yards. Interestingly, penalty kicks tend to incur very high xG scores, since approximately four-fifths of penalties are converted in the top leagues.

xG is therefore an increasingly popular metric used by sports bettors to help find teams that are creating plenty of chances, particularly those seeking matches with a higher number of goals and trading the Over/Under 2.5 and 3.5 goals markets. Sports bettors that prefer to take a longer term view in the ante-post markets can also use xG data to try and predict the future performance of a team. Of course, the main limitation of using past xG data to anticipate future results is that xG does not take into consideration aspects like injuries, transfers in and out, and managerial sackings and replacements.

It can also be useful to utilize xG data to pinpoint potential strikers for fantasy football team selections. The premise of fantasy sports betting sites is that players accrue points for every goal or assist their strikers rack up. xG can highlight teams that are creating plenty of chances, even if they may not be scoring them at present. It may be that a striker is simply waiting to find form and could therefore be an ideal fantasy team pick for future game weeks.

Ultimately, teams that start a season by regularly outperforming their xG data are more likely to revert to the mean over time. When we say outperforming their xG data, we mean winning games 2-0 despite only having an xG of 0.80. This may mean their strikers are in a rich vein of form and are supremely clinical with even the merest sniff of a goalscoring chance. A team may also have a set piece specialist capable of scoring free kicks out of nothing, which barely register on the xG scoring system.

Some teams are simply designed to outperform the xG data models. In the English Premier League, Burnley have consistently achieved more points than their xG data suggests they should have. That’s largely because Sean Dyche’s team are built on solid foundations and are prepared to soak up plenty of pressure from the opposition and hit teams on the counter-attack, without creating a glut of chances.

Ultimately, there is never going to be a 100% fool-proof way of recording team data in the world of football. However, xG goes some way to articulating the fluidity and cohesion of teams in the final third.

All you need to know about xG data

Comments 0 Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.