Premier League 2026-27: high non-penalty xG, subdued finishing and unusually many draws
By Oct 9, 2026, Premier League matches in the 2026-27 season were producing more non-penalty expected goals than actual non-penalty goals, with 2.88 non-penalty xG per game compared with 2.70 non-penalty goals per game, and the proportion of draws higher than in any previous Premier League season.
Matches this season have averaged 2.70 non-penalty goals per game, a figure higher than last season and higher than in two of the four previous campaigns, while non-penalty expected goals have averaged 2.88 per match, the second most in the past six seasons, creating a shortfall that would equate to almost two extra goals per week if teams converted at par.
The distribution of underperformance is broad: 14 clubs have scored fewer goals than expected (whether penalties are included or not), up from 10 last season; eight teams sit within one goal of their underlying numbers and the other 12 are split 8-4 towards underachieving, and there were 24 instances across the first five match weeks of a team scoring at least one goal fewer than their xG implied, twice the number recorded across the opening 50 games last season.
Chance quality does not appear to account for the shortfall in full, with the expected goals per non-penalty shot average only dropping once more decimal places are applied than feels healthy, and touches in the opposition penalty area have changed only marginally at league level, averaging 24.8 touches per game, down 0.6 on last season.
The team-level breakdown is more revealing: four of last season’s top five for box touches are seeing fewer touches per match, with Aston Villa and Liverpool experiencing around a third of their previous averages disappear, while Manchester United are the only team from that quintet to increase their box touches but have faced just one club that finished in the top 10 last season and saw five of their eight goals come against Championship runners-up Ipswich.
Manchester United lead the Premier League for non-penalty shots yet are joint-second bottom for average expected goal value, and their leading player for non-penalty xG, Bryan Mbeumo, is one of 10 players who have scored at least one goal fewer than underlying numbers imply; Mbeumo underachieved by 1.19 last season and United are described as reliant on the 27-year-old.
Top players are taking a larger share of each club’s chances than last season: the leading player for expected goals at almost every Premier League club has a bigger share of the xG pie, and the top players collectively have scored 6.8 non-penalty goals fewer than expected, almost 40% of the division’s shortfall coming from just 20 players; Thierno Barry has taken 49.4% of Everton’s non-penalty xG this season, up from 17.7% last term, and has scored twice from chances valued at 3.5 xG.
There are further examples of misalignment between player profiles and expectations: Alexander Isak has seen a greater leap in his share of team expected goals but did not feature often last season because of injuries and remains relatively new to minutes at Liverpool; Isak is playing under a new man in the dugout, and Isak’s replacement at Newcastle, Yoanne Wissa, underperformed against xG by more than two goals last season, as did Marcus Tavernier at Bournemouth, while Tottenham’s leading non-penalty xG player is centre-back Jan Paul van Hecke; Fulham’s Gonzalo García and Ipswich’s Emersonn are 22 years old and new to English football, Hull’s Oli McBurnie has never scored more than six goals in a Premier League campaign, and Erling Haaland remains a reliable xG-heavy frontman though the situation at Manchester City may affect his form.
Until the reliance on a small group of players or the small-sample issues driving these gaps change, do not bank on the goal rate reaching the expected-goal level in the Premier League.
