This Week in Stats: Manchester is red after surprise United win

Manchester United turned it around against Manchester City, Southampton deserved more at Arsenal and top quality chances wreaked havoc elsewhere.

In March 2016, the Etihad Stadium hosted a derby. By the 65th minute, Manchester City had taken a total of 13 attempts at goal. In the whole of that match, plus the two derbies at City since, Manchester United have had 13 shots.

City, by contrast, have taken a total of 65 shots across these three games. Yet United have now won two and drawn one of them. This was certainly the biggest upset of the three, as City were 2-0 up at halftime. It’s almost a decade since the Sky Blues last lost after going two goals up, and few defeats in that period will have hurt as much as this one.

Pep Guardiola’s side were so dominant in the first half that a two-goal deficit flattered United. The shot tally was 9-0, with five on target, and four of those categorized clear-cut chances. In the 86 big six matches since the start of 2015-16, a side has had four-or-more top quality chances 19 times. To have that many in a half in a big match is incredibly rare.

City will regret only taking one of the four great chances, but there was little inkling of what lay in store.

After Paul Pogba, who had been wholly ineffectual in the first half, scored twice, Chris Smalling got the winner. City had only conceded one set-piece goal in the league all season, and none since early December. Yet United’s third goal came via a crossed free-kick, which added to the unlikely nature of the turn around.

There’s no need for anyone at City to panic. Bookmakers claim to have taken new bets on United winning the title, but there’s no way that will happen. It is, however, fair to say City are wobbling at the point a great season could’ve become legendary. The match with Liverpool on Tuesday will play a huge part in how this season’s City side are remembered.

Southampton improve but lose again

While Southampton will never expect to go to Arsenal and win, they will have few better chances than they did on Sunday. The Gunners fielded a weakened lineup, as their focus is now entirely on the Europa League.

In fairness to the Saints, it was arguably their best attacking performance of the season, and it certainly was in terms of away matches.

Their first 14 away league fixtures in 2017-18 saw them amass 14 clear-cut chances, yet they had four on Sunday. James Ward-Prowse had already missed one before Shane Long gave Southampton the lead with their second. Charlie Austin scored their third to make it 2-2, but he then missed their fourth, which proved costly.

Not that Southampton can ask much more of Austin. Despite only making six starts in the league he’s their top scorer with seven goals. Had he been available more this season, they might not be in the relegation trouble they are.

For the second week in a row, they conceded three goals in London. Southampton were far better here than they were at West Ham, but with Chelsea at home next, their London troubles may not be over.

Still, Mark Hughes can take heart from the attacking verve which the Saints have too scarcely displayed this season. Their average expected goals total on the road has been 0.9, but this was comfortably their best effort, at 2.6.

It’s also the first time Southampton have been at least one expected goal better than their hosts, and their eight shots on target marked easily their best performance in that area too.

Finishing can be everything, though. Arsenal won the day despite having fewer shots of far lower average quality. Their home record against the teams below them is phenomenal. This was their 13th win out of 13. It certainly proved to be unlucky for Southampton.

Next: Liverpool hold Everton in scrappy derby

Big chances can make all the difference

This Week in Stats usually focuses on three matches, but aside from the two five-goal thrillers there were interesting statistical occurrences in several other games. It’s well known in analytics circles that a couple of very good chances far outweighs having far more low quality ones, even if the expected goal totals are similar. There were multiple examples of how really big chances can determine outcomes this weekend.

Take Bournemouth vs. Crystal Palace, for instance. It finished 2-2, yet the expected goals figures suggest the home side deserved to win. However, their late equalizer accounted for 27 percent of their xG tally. It wasn’t close to the most extreme example, though.

Brighton had taken the lead in a relegation six pointer, before Shane Duffy’s error presented Huddersfield with 56 percent of their total expected goals on the day. Far more importantly, it presented them with a goal and a valuable point. Around a half of both Everton and Liverpool’s expected goals came from one of their shots in their derby draw, too.

Watford were in reasonable control at Vicarage Road, until 58 percent of Burnley’s xG on the day fell to Sam Vokes, who equalized. But the real difference maker was Jamie Vardy’s goal against Newcastle.

His late strike didn’t just account for 69 percent of Leicester’s attacking value, it was 51 percent of the match’s total expected goals. As Vardy’s effort was more than Newcastle’s entire attacking output on the day, perhaps it should’ve been more than a mere consolation.