Matchday two of the Premier League season brought excitement and drama, with significant wins for Chelsea and Manchester City. Noni Madueke and Erling Haaland both scored hat-tricks, while Everton's poor start continued with a 4-0 defeat at Tottenham. Using Opta data, we analyse the weekend's winners and losers based on underlying metrics.
Chelsea's 6-2 triumph over Wolves was the standout result of the weekend. Despite scoring six goals, their expected goals (xG) was only 1.68, while they allowed chances worth 1.96 xG. Noni Madueke, Nicolas Jackson, Cole Palmer, and Joao Felix all found the net. Enzo Maresca became the first manager to see his team score six in his debut away game.

Arsenal also had a fortunate win, beating Aston Villa 2-0 despite an xG of just 0.87 compared to Villa's 1.28. Goals from Leandro Trossard and Thomas Partey kept Arsenal's perfect start intact. This marks only the third time Arsenal have won their opening two matches without conceding, previously achieving this in 1924-25 and 1971-72.
Wolves' heavy loss to Chelsea might seem harsh given their performance. They managed 12 shots to Chelsea's 14 and had more touches in the area (25 to Chelsea's 21). Chelsea's xG overperformance of 4.31 is the highest this season, surpassing Brighton and Spurs' overperformances of 1.57 against Everton.
Aston Villa were unlucky not to replicate their matchday one success against West Ham when they narrowly won 2-1. Against Arsenal, Villa had three shots on target from eleven attempts but failed to score. Ollie Watkins missed two chances worth a combined xG of 0.87, making him one of the most wasteful players this season alongside Nottingham Forest’s Nikola Milenkovic.
Everton’s dismal start persisted with a crushing defeat at Tottenham. The Toffees were outclassed throughout the match, failing to find any rhythm or create significant chances.
The weekend saw justice largely served across other fixtures except for Chelsea’s and Arsenal’s games where they won despite lower xG figures than their opponents.
This analysis highlights how underlying metrics like xG can provide deeper insights into match outcomes beyond just the final scoreline.