Five games into the Premier League season, the table can still tell a misleading story. At the same stage last season, Liverpool were top with five wins from five and Tottenham were third; they went on to finish fifth and 17th respectively. The sample size is small and some teams have had harder starts than others, so any statistical reading needs caution. Expected goals, or xG, offers a more useful picture of how much teams have actually dominated games, based on the number and quality of shots they have taken and faced. It measures general performance before finishing, shot-stopping and luck come into play. Premier League history shows that the more a team dominates, the more points it tends to win over the long run. Penalties are best excluded from this analysis because refereeing decisions and defensive clumsiness can give a false impression of how good teams are at attacking and defending.
On overall xG difference, champions Arsenal remain the best team despite suffering their biggest defeat in three years at Brighton last weekend. Leaders Manchester City are not far behind them in terms of dominance. Unbeaten Brentford’s strong start has not been fortunate. Manchester United continue to swing between feast and famine, yet only three teams have a higher xG difference than them. Nottingham Forest and Bournemouth have also performed better than their early table position suggests. At the other end, Aston Villa, Crystal Palace and Newcastle have been the least impressive overall, with Newcastle perhaps the most surprising given their top-half position. Villa showed last season how quickly a slow start can be turned around, finishing fourth after a difficult beginning. The promoted sides may not make life easy for established clubs either, with Hull and Ipswich both making genuinely solid starts.
Going forward, Brighton have had the best attack so far based on the number and quality of chances they have created, followed by Manchester City, Brentford and Manchester United. Arsenal are more admired for their defence, but it is still surprising that eight teams have a higher xG than the champions, while Liverpool and Chelsea rank in mid-table. Brighton have been ruthless in front of goal, scoring 15 non-penalty goals from 10.7 xG. The only team more clinical is Newcastle, who have the lowest xG in the league so far. Newcastle have scored 9 goals from 4.5 xG, twice as many as a team would historically be expected to score from those chances. Head coach Matthias Jaissle will want his side to create more rather than rely on such sharp finishing. Coventry, Sunderland, Nottingham Forest and Tottenham have all scored three goals fewer than expected from their first five games, leaving the question of whether they have a finishing problem or whether their luck will change.
Defensively, Arsenal still have by far the best backline in the league, even taking the heavy defeat at Brighton into account. Nottingham Forest head coach Oliver Glasner has work to do to get his side scoring, but he has already instilled a solid foundation at the back, as has Andoni Iraola at Liverpool. There is a lot of construction work needed in the capital, with Crystal Palace, Fulham and Chelsea all far too open at the back, although Xabi Alonso’s side are not quite as terrible as the 12 goals they have conceded suggests. Brighton’s brilliance going forward has come at the cost of defensive solidity, and Fabian Hurzeler’s team may have been a little fortunate not to concede more often. Everton and Leeds have also conceded three goals fewer than expected given the chances they have faced, and both have been among the strongest defensively overall.
With only five games played, it is too early to make concrete claims about how the season will unfold. Even so, the underlying numbers have given a glimpse of what might be in store.