Analysis: Manchester sides swap personas as potential race for title shapes up

The pre-international break fortunes of both Manchester sides swapped on Saturday – a heavy scoreline which could be deceiving fell in City’s favour, while United surprised in looking like mere mortals.

Jose Mourinho’s United were never going to keep up their record from those first three games – scoring ten goals and conceding zero – as they drew 2-2 at Stoke.

Both West Ham and Swansea crumbled at the end of their matches, and taking the final ten minutes of each put together, United scored five goals. That’s more than 11 sides in the league have scored all season.

Similarly to his team, Anthony Martial couldn’t keep up his record of scoring in each game he was substituted onto the pitch, having done so in the first two games of the season (he then started the third game, against Leicester).

This will not necessarily be any surprise to the select few privy to a model from the Sports Analytics Machine (SAM), developed by academics from the universities of Salford (Dr Tarak Kharrat), Liverpool (Professor Ian McHale), and University College London (Dr Javier López Peña).

The model quantifies the quality of chances created and produces the number of Expected Points that would usually be won based on how each side performs. They then compare how a team performs when an individual is playing versus when they are not.

Martial was one of the players who performed the worst for Manchester United in 2016/17, according to the model, and his performances this season appear to match what SAM suggested about him last term.

In United’s first two games, the team were winning when Martial came onto the pitch (and scored). They didn’t open the scoring against Leicester until he had been replaced by Marcus Rashford, and they failed to score again after Martial had been introduced on Saturday.

A similar version of this model was featured on BBC Sport last week, but Professor Ian McHale says that the Expected Points model has a small difference in how it can be used to that one.

“[The Expected Points model] deals with players who play well when the score is 3-0, when it’s 4 to 5-0,” McHale told MM.

“Whereas there are some players who do well when it’s 0-0 and 0-1, and turn a result around.

“Actually that’s arguably what most managers are looking for, the players who perform well when it really matters.”

Some of the players who are highly rated by this model for United are Paul Pogba and Eric Bailly, and for City David Silva and Raheem Sterling, based on their performances at those clubs last season.

It was Raheem Sterling who scored an equaliser against Everton earlier this season, and who got the winning goal against Bournemouth just before the international break.

City on fire

Struggling, somewhat, was the story of Manchester City’s season pre-break, but that flipped around completely in a heavy 5-0 victory over Liverpool at the weekend.

However, before Sadio Mané’s red card in the 37th minute, the two sides were very even. City were edging it, leading things 1-0 by the score-line and 5-3 in terms of shots on target.

Liverpool, down to ten men against a Pep Guardiola side who had already been in good form before the dismissal, couldn’t keep up and didn’t muster a single shot for the rest of the game.

For both Manchester sides, their true value probably lies somewhere between their performances before and after the international break.

With them both sharing the top of the table – ten points each with United only ahead on goal difference – a fascinating battle in the city could be due to take place. 

Image courtesy of BBC Sport via YouTube, with thanks.

Related Articles