Michael Carrick will arrive at Stamford Bridge on Saturday with the air of a dead man walking. Yet despite Manchester United being fifth-bottom of the Premier League and Chelsea fifth-top, the Blues have performed worse this season according to Opta’s expected points model.

Is there a glitch in the matrix? Has the fabled supercomputer clicked on a suspicious link? Even though the margin between their data is small, it feels unthinkable that Chelsea could be below United’s level.

That is until you review a few statistics. Xabi Alonso’s side held healthy expected goal margins in their wins over Fulham, Brighton and Everton, but were heavily outgunned against Arsenal and Brentford. The xG count went slightly against them versus Hull, Bournemouth and Tottenham, which is concerning when all three were at home in matches they would reasonably expect to win.

Opta’s data suggests Chelsea are the 12th-best team in the Premier League. They have risen to fifth by virtue of a hot finishing streak. Chelsea have scored 6.7 goals more than the quality of their chances imply they deserve, second only to Brighton (6.8), or above them if penalties are excluded. Chelsea scored five times from 10 shots against Bournemouth; while not without precedent, managing so many goals from so few efforts happens less than once a season in the Premier League.

Jordan Henderson has scored twice from five shots outside the box, after finding the net once for every 20 long-range efforts he attempted between 2014 and last season. Pedro Neto’s total of three league goals is two shy of his career best for a Premier League campaign. Morgan Rogers outscoring his expected goals figure is not unusual but he is ahead by 1.87 when he was up 3.15 across the whole of last season. This trio are unlikely to maintain such hot form across another 30 league games.

Adding in João Pedro gives Chelsea four of the 12 players who are above their non-penalty xG by at least 1.5 in the Premier League this season. The two league games the Brazilian missed are the two in which Chelsea failed to score, mustering a combined 1.3 expected goals across those matches (against Brentford and Tottenham). João Pedro’s return from injury cannot come soon enough for the club.

This is not least because he provides a good focal point for the frontline, even when he is not scoring. Data from Gradient Sports shows João Pedro has had possession in the opposition penalty area 5.7 times per 90 minutes this season, the most among centre-forwards with at least 400 minutes in the Premier League. Despite playing in wide areas, Estêvão Willian’s average for this metric is 5.8 since joining Chelsea. No single stat can justify a player’s inclusion but the teenager is being omitted when he offers something the team lack in João Pedro’s absence.

Keeping possession in the penalty area has become increasingly difficult for Chelsea, who have dwindled after a positive start. Third in the division for opposition box touches after four matches, they now sit eighth. While Premier League home teams are averaging 27 touches per match in the box, Chelsea had 17 in their previous two games at Stamford Bridge. Given their possession average is markedly lower than in any recent campaign, it could prove difficult to increase their penalty area activity.

Alonso has overseen three league matches in which Chelsea had a maximum 38% possession; he had four such games in almost three years at Leverkusen. This is not as game-state driven as you might assume. Chelsea had 40.8% possession at the point they went 3-1 up at Fulham, 34.6% thereafter. They raced into a 3-0 lead against Brighton despite attempting 28.4% of the total passes to that point.

This cannot be by design when Alonso’s Bundesliga-winning Leverkusen team topped the German possession standings with 61.8%. He acknowledged as much when quizzed on the issue before Wednesday’s Carabao Cup defeat at Anfield. “You want to have possession,” Alonso said. “Knowing the context, knowing what we are building, I think that we have room for improvement in that part. We need to play more confidently with the ball.”

History shows Champions League qualification is unlikely with Chelsea’s 46.5% possession. The miracle of Leicester’s title win provides the only instance from the past 12 seasons of a team finishing in the top four with a lower possession average. Beyond that, West Ham finishing sixth in 2020-21 is the best showing.

Things may improve once the tactical plan becomes embedded and the lineup settles. Alonso has deployed 3-4-3 and 4-2-3-1, neither for more than three consecutive league games. Only Tottenham have averaged more changes to their starting XI from game to game in the Premier League (and this with neither side having to rotate owing to European commitments). Chelsea have started three goalkeepers, summoning distant memories of 1996-97 when Craig Forrest, Kevin Hitchcock, Nick Colgan, Frode Grodås and Dmitri Kharine got starts between the sticks. It is little surprise the Blues have the second-worst rate in the division for conceding over their expected goal level.

Much is up in the air at Stamford Bridge, yet there is a weird sense of certainty about Manchester United. They are a poor team and Carrick will surely leave soon. Alonso is a long way from the exit door but many questions need answering about his side.