Leeds United's xG deficit explained - what does data tell us about Whites' Arsenal chances?

Here, we try and shed some light on the burning questions surrounding the metric, which has crept into our consciousness over the past few seasons.

We’ve included an interactive graph – made with Flourish – which notes every team’s expected goals for and against.

We’ll also take a look at what expected goals means for Marcelo Bielsa and Leeds United’s season so far.

WHAT ARE ‘EXPECTED GOALS’?

For a statistic that can seem complicated at first glance, the explanation is actually rather simple.

Expected goals - or xG as it appears on your television screens - measures the quality of a shot based on different variables.

These include assist type, distance from goal, shot angle and whether it was defined as a ‘big chance’.

In short, a player having an xG of 0.3 means he or she would score 30 per cent of the time.

When looking at the xG of a Premier League team, an xG of 10 would equate to the number of goals a team has been expected to score up until that point in the season.

For example, Chelsea boasts an xG of 13.6 but have netted 20 goals in the Premier League so far this campaign, meaning they are over-performing.

WHAT CAN EXPECTED GOALS TELL US?

Measuring expected goals can give fans an indication of the number of times that a team or individual player should have scored on average, given the type and number of shots they have taken.

Of course, this isn’t an exact science and the expected goals statistic is prone to fluctuate during a season.

WHAT DO EXPECTED GOALS TELL US ABOUT LEEDS UNITED?

Leeds United’s xG this season so far is 14.27, the fourth highest in the Premier League. In reality, the Whites have found the net 14 times this campaign.

That means Bielsa’s side are ever-so-slightly under-performing compared to the number of goals they are expected to score using the xG metric, but not by much.

The expected goals Leeds are predicted to concede sits at 16.59, the highest in the top-flight and ahead of West Brom and Newcastle United, which isn’t a good sign.

In their fixtures so far, Leeds have shipped 17 goals, the most in the division and tied with the Baggies.

That means Bielsa’s side are performing just about as expected using the xG metric.

That, however, throws up a big cause for concern given Leeds’ xG deficit when subtracting expected goals for from against is -2.32.

By looking at Leeds’ xG, the data predicts the Whites probably won’t turn a corner against Arsenal on Sunday, but football does not always dance to the data’s tune.

CAN WE TRUST EXPECTED GOALS?

You’ll find many traditionalists that don’t want academics crossing over into the beautiful game, and xG is certainly a product of academia.

Indeed, it is unlikely that anyone could predict every team and their player’s real live goal-scoring potential an exact science.

Nevertheless, xG remains valuable as a predictor and indicator but should definitely be balanced alongside watching matches with your own eyes and drawing your own conclusions.