18+ | Commercial Content | T&Cs apply | Begambleaware.org
Christmas is nearly upon us and with that arrives not just presents, but reviews of every single Premier League club.
Their campaigns to date have assessed along with how their summer signings are getting on before each side is given a grade for the first third of their respective top flight seasons.
This time around we've decided to do something a little different. We're looking at expected goals (xG), expected goals against (xGA) and expected points (xPTS) in an attempt to see which clubs in the English top flight are over-performing and which of them are under-performing.
Expected goals, for those unfamiliar with the term, is simply a method of assessing chance quality. The better the chances you have in a match, the higher your xG will be at the end of it, and, generally speaking, the more likely you are to win. It’s a better predictor of future goals than goals themselves, and is a strong indicator of team performance.
For example, if a team is on a poor run of attacking form in terms of goals scored, but is getting good chances, we can expect them to start scoring more over the long term. If they aren’t scoring, and also aren’t getting good chances, their attack is probably just legitimately bad.
Expected goals against is the reverse of xG. It looks at the quality of chances a team is conceding. Just as xG is a good indicator of the quality of a team’s attack, xG conceded is a good indicator of their defence.
Expected points is a prediction of how many points a team would typically have based on the quality of chances they have had and conceded.
Is your team’s form sustainable, or can fans expect the rest of the 2017/18 season to differ from the first? We break down the findings. All stats can be found on understat.com.
What the xG numbers say
Crystal Palace fans may be scared to look at the table, but xG numbers may provide them some solace. The Eagles are bottom of the top flight and are five points adrift of managerless West Brom, who sit just outside the relegation zone in 17th.
They've scored just the six goals in their opening 12 matches but you would imagine the return of Christian Benteke will see this number rise. Especially when you consider they're underperforming their xG for the season by 9.63 goals already.
What that shows is Roy Hogdson's team are creating quality chances but, for differing reasons, they're unable to put the ball into the back of the net. It could be in part down to fantastic goalkeeping, or a run of poor finishing that is likely to be really unlucky.
Whatever the case, Benteke knows where the goal is and he should be able to right a number of wrongs. Last season the big Belgian finished with 15 goals in the league but his xG was closer to 18. So he will get chances.
Palace aren't the only team under performing with regards to their xG. Southampton are 5.5 goals worse off, Swansea City are 3.31 under what their expected return is and Newcastle United come in 2.31 down.
Interestingly, Liverpool are 2.31 under their xG for the season, too. They're the only team in the top ten to be under performing their xG model by more than two goals.
It's surprising given Mohamed Salah is scoring 0.87 goals per 90 minutes in the league this season and they're already the third highest scorers behind only the two Manchester clubs.
When looking at the Premier League table, the obvious choices for clubs over performing would be Burnley and Brighton. Both make the top five in terms of over performing their xG but they're still way off those in first and second position.
Chelsea have scored 7.56 more goals than they've been expected to in just 12 matches. It seems like a lot and may appear to be unsustainable, but elite players can often over-achieve expected goals totals for extended periods.
In Antonio Conte‘s first season, for example, their xG was 63 yet they finished the campaign with 85.
The Premier League champions aren't just fortunate, they're clinical in front of goal. This ruthlessness will keep them in the top four hunt for the remainder of the campaign.
Following closely behind the Blues is Manchester City. The league leaders are rampant and they're currently outscoring their xG model by 5.86 goals this season. Last season their xG was just 0.61 out. If it's accurate again then City should score over 100 goals – 104 if we're being exact.
Burnley (3.92) and Brighton (3.54) are going along nicely, and that's unsurprising with both clubs sitting in the top half of the table despite being tipped to be in a relegation battle.
What is unexpected, however, is it's 15th placed Stoke City who are next on the over performing list having scored 2.12 more goals than they've been expected to and they're closely followed by Watford (2.06).
No other team has outscored their xG by more than two goals.
What the xGA numbers say
Everton spent close to £50million on bringing in defensive upgrades this summer in the shape of Jordan Pickford and Michael Keane. Despite this it's the Toffees who top the list of under performers for expected goals against.
Is it sustainable? Well last season their xGA was pretty spot on with the Toffees conceding 44 and their xGA being 46. You'd expect them to improve over the rest of the season. If not, they could find themselves in trouble.
Stoke, who over performed in the xG model, are doing so in the xGA, too. They've conceded 4.52 more goals than they've been expected to. It negates their positive xG numbers and means they're basically two goals worse off overall.
West Ham United have also conceded more than four goals over what they should have. If the Hammers were performing as expected then they wouldn't be in the relegation zone and Slaven Bilić might still be in a job. Should be Croat feel hard done by? Perhaps.
Next up is Palace who are seemingly having an unfortunate season. They're way under their xG and are also underperforming in the world of xGA by 3.4 goals. That's a 13 goal swing over the two models for the Eagles and explains why they're at the foot of the table and battling for survival.
West Brom (2.82), Leicester City (2.44) and Liverpool (2.11) are the only other teams to have conceded two or more goals than their expected model suggests they should have.
Only Crystal Palace and Liverpool appear in the under performing category for both xG and xGA.
Three clubs are over performing in xGA, with Burnley, again, posting strong numbers.
The Clarets have conceded 6.64 goals less than they should have. You can put that down to some superb goalkeeping by both Tom Heaton and Nick Pope as well as the team being brilliantly switched on when it comes to making vital blocks in dangerous areas.
Manchester United aren't far behind having conceded 6.53 goals less than expected. Only one man is to thank for this and it's David De Gea who has yet again kept the Red Devils in games single-handedly at times.
Bournemouth come in third on the list with the model suggesting they should have conceded 4.37 more goals than they should have this season. They're sitting in 13th position now but had these chances been buried they could well be in a relegation scrap.
What the xPTS numbers say
It's as you'd expect, really. If teams are conceding more goals than they're supposed to be and failing to convert as many chances as they'd expect, then they're going to be worse off points wise.
Crystal Palace have 11 fewer points than the xPTS model suggests they should have. For even more context, in the expected points table, they'd find themselves in eighth position. It demonstrates just how small the margins in football can be.
Leicester City are five points worse off than they should be and are sitting in seventh in the xPTS table. West Brom (2.73) would be in 14th and Southampton (2.57) would find themselves in ninth position.
The only top ten team to be under performing by two points or more is Liverpool (2.28). The Reds are second in the xPTS table, just six points shy of Man City at the top.
Just as with the under performers, it's obvious after looking at xG and xGA which of the teams are over performing. Burnley have ten points more than their xPTS says they should. In the expected table they're in 16th position and struggling to stay up.
Chelsea are 5.26 points better off, but would sit in fifth if all of the matches had finished as expected. It was similar for Conte's men last season after they collected 17 more points than the xPTS model.
Brighton (3.63), Huddersfield (3.55), Man Utd (3.30), Watford (3.14) and Man City (3.14) are the only other teams with three or more points than what they have on the expected points table.
The sample size isn't the largest but it does help validate a number of theories.
Firstly, David de Gea is extremely important to Manchester United. The Spanish No.1 is the best goalkeeper in the Premier League and he's, to some extent, the difference between the Red Devils battling for the title and being in a fight for fourth.
Chelsea, meanwhile, continue to be clinical. They managed it last season and they're repeating it again this time around.
Their quality of chances might not be that high – Crystal Palace and Leicester City have a better xG than them – but they make their chances count.
Spurs and Arsenal don't get a mention in these lists simply because they're performing as expected so far this season. It's a good sign for both clubs in terms of sustainability with the xPTS model having them in third and fourth place respectively.
Further down the table, could the Burnley and Brighton bubble burst at some stage and will they go into free fall? If they maintain this for the season then they've massively over performed. Or will their numbers balance out as they improve?
It will be interesting to review this at the end of the season.