23rd June 2019
With the Group Stage of the 2019 Women’s World Cup done, I thought I’d look forward to the Round of Sixteen and check in on some of the numbers.
I was going to do a roundup of the groups, but thought I can cover similar points here as well as looking to see how teams match up. I’ve decided to just reel off some numbers (mostly xG, pressure and progression) about teams, then give a quick mention to something that seems like it might be fun to look out for based on the data, before giving a short prediction.
Like I mentioned in my previous posts, this World Cup is my first real venture into women’s football and I’m terrible with previews/predictions in general so don’t expect any predictions to be accurate, it’s just a fun thing to do.
I also made heavy use of the soccermatics (@JoeDGallagher) package and a few FC_rStats (@FC_rstats) packages (particularly the pitch) for this, which I’d highly recommend if you’re looking to use StatsBomb’s free data.
Taking maximum points in the World Cup Group Stage for the first time since the men’s side in Spain ’82, coupled with some strong underlying numbers, England have given themselves a strong base to build from.
Wins against Scotland and Argentina were expected, with the game against Japan being seen as the real test. Scotland were better in the second half and England had a debatable penalty, but England did have the better xG numbers, while a great performance from ‘keeper Vanina Correa kept the score down against Argentina.
As expected, England didn’t create quite as much against Japan but didn’t concede many chances until late on when Japan started to apply more pressure and England seemed happy to sit and look to break. This can be seen in the xG timeline below.
I’m getting sick of mentioning this in these previews, but a big part of England’s attack is their right hand side. Excluding Karen Carney, who’s attempted a lot of progression in her substitute cameos, three of England’s top four players for attempted progression come from the right hand side – Nikita Parris and Rachel Daly are the top two, before Beth Mead and right-back Lucy Bronze.
Looking at England’s key passes into the danger zone, there’s only been three in open play that haven’t come from the right hand side, which can be seen below.
With Phil Neville making several changes each game, it’ll be interesting to see who will lineup against Cameroon, with some question marks over a few positions.
The main positions where there seems some debate is the #6, with Keira Walsh getting caught in possession a few times vs Japan, it’ll be interesting to see whether Neville will stick with her or go for the more experienced Jade Moore.
Personally, I’d stick with Walsh, I am a fan of hers and have seen more of her than Moore, but it’d feel harsh to drop her after a couple of possession losses against one of the best sides in the world, who also have a high proportion of pressures in the final third.
Watching a video of Neville talk about his tactics, he talks a lot about gaining superiority in midfield, pushing the full-backs wide and mentions specifics about Toni Duggan dropping into midfield and trying to get the midfielders on the ball.
The clip below shows a turnover following a bad touch by Walsh where this tactic can be seen, but also how Japan pressure and are there to capitalize on the poor touch. If Walsh could have poked the ball through to Stanway as she turned, England would be in a great position to run at the Japanese back line. Instead she looks to roll the ball back, open her body and probably play it to Jill Scott, but the Japanese midfielder reads this.
It’s walking a fine line, as something like this could be costly down the line, but it also feels like these mistakes aren’t going to be eradicated unless she’s given a consistent run of games, rather than being constantly changed. It feels like England can use Walsh, Lucy Bronze, Jill Scott and Nikita Parris/Rachel Daly to progress well on the right hand side, like in the clip below.
It’s something England did a couple of times. Going out to Bronze in the build-up and getting Jun Endo to press opened up space between Endo and Hina Sugita, creating an opening in the right half-space kind of area, allowing Bronze to knock the ball back to Walsh to progress it to Scott or Daly with a chance to run at the Japan defence. The same happened in the build up to the Toni Duggan chance in the second half.
Elsewhere in midfield, it’ll be interesting to see whether Neville goes back to Fran Kirby in the more attacking role, or rewards Georgia Stanway after her assist vs Japan – or does something else entirely and plays Karen Carney there.
Stanway got the assist but didn’t get on the ball too often against Japan, particularly in advanced areas, which Neville placed a big emphasis on when talking about his tactics. It is against Japan, but the passes received by Stanway against Japan shows how little she received it in advanced areas.
With her long range shooting, Stanway feels as though she offers a more direct threat, while Kirby will be better suited in and around the box, dribbling at the defence and looking to slide it through. I think I would lean to go back to Kirby, but it isn’t an easy decision to drop Stanway after how she was productive vs Japan.
Finally, there’s also question marks over the wing positions, with Toni Duggan making her first appearance at the World Cup and Rachel Daly making her first start, after Nikita Parris and Beth Mead lined up on the wings before.
I’d lean towards Parris and Mead coming back into the side, but Daly and Duggan put forward strong numbers against better opposition in the game against Japan. Mead has been England’s most productive player for xG + xA p90, while her set-piece ability is also a big bonus.
Despite Parris looking strong and having good progression numbers, she hasn’t turned that into great xG + xA numbers in her two games, while Daly has some strong numbers from her substitute appearances plus start vs Japan. It’d be harsh on Daly after a decent performance against Japan, but I would go back to Parris to start I think.
Putting that together means I’d go with the line-up that started against Scotland, although I’d be surprised if Neville does opt for that.
Getting there in the most dramatic way possible, Cameroon are the lowest ranked sides to make the knock-out rounds but with some strong numbers from a tough group, England shouldn’t be taking them lightly.
Cameroon’s big defeat, the 3-1 to the Netherlands, happened due to some good finishing by Vivianne Miedema, rather than poor defending from Cameroon, as can be seen on the timeline below.
Then their win against New Zealand, a team who beat England in their final warm-up game, was a convincing one, marred only by an own goal.
Given the tough group they had, Cameroon coming away with the 11th least xG conceded is a respectable record.
It’s a lazy comparison to make, given they’re the only two African sides to have made it out the group stage, but it’s a similar situation as Nigeria. They’ve been pretty solid in a decent group, both conceding around the same xG, but Cameroon managed to create double the xG that Nigeria did and even created slightly more than France did.
Putting the above together, Cameroon ended up leaving the group with a positive xGD, having a better xGD than Brazil and even Japan – who had two games against teams they were expected to have routine wins against.
Looking at pressure events, Cameroon have the 5th highest proportion coming in the middle third and 8th most in the defensive third, while also being 2nd for the most opposition passes in the midfield area (middle third and width of the 18 yard box) made under pressure.
What’s interesting is that this doesn’t lead to great numbers for pass completion, their numbers for pass completion of passes in the midfield area and progressive passes are around average.
Cameroon’s numbers without the ball are a lot stronger than their ones with it, attempting the 5th least progression with the 2nd worst completion rate in attack, but it does mean they’re efficient in terms of progression to chances, getting good xG numbers from poor progression numbers.
Looking at individuals, while Ajara Nchout scored the goals to take them through, Gabrielle Onguene has been their most productive player for xG + xA p90, mostly thanks to a hugely impressive 0.55 xG p90. Looking at players who have played more than 100 minutes, she has the 2nd highest xG p90 for those who don’t play for a top ten ranked side.
Both teams have a high proportion of pressures in midfield, as can be seen in the plot below.
It’ll be interesting to see how the pressure in midfield effects the game, particularly after talking about how Keira Walsh had a couple lapses in possession against Japan – who also have strong midfield numbers above.
Lucy Bronze vs Gabrielle Onguene
If the midfield is tough to play through, there’s a good chance England will turn to the flanks to build-up and look to create overloads out wide like the clips from earlier.
However, with the full-backs pushing forward and wide, it could leave space for the likes of Gabrielle Onguene to run into. Against New Zealand Ajara Nchout played in the #10 role and looked to find Onguene and Michaela Abam out wide.
Cameroon have lined up slightly differently in each game, but if they play the same against England, you’d think there’d be room to counter with Onguene and Abam having space in behind. Then, given Bronze’s importance to England’s attack and build-up and Onguene’s strong xG numbers, that seems England’s right side could be a big factor on the result.
England should progress to the next round, but it doesn’t look like it’ll be an easy game. Cameroon coming out of a tough group with a positive xGD is hugely impressive, they’ve proved themselves to be tough to play against and have an effective counter attack that should look to take advantage of the space left behind England’s attacking full-backs.
Hosts France did as was expected, qualifying from the group with three wins out of three, relying on a strong defence in doing so.
Not including penalties, the only game where they managed more than 1 xG was the 2-1 win over Norway, with 1.1. Their four goals in the opening game against South Korea came from an xG of just 0.8, while their win against Nigeria came from an xG of 0.9 and thanks to a controversial penalty retake.
While they weren’t great going forward, at the other end of the pitch they were incredibly strong. The only goal they conceded was an own goal, while their total xG conceded from the three games was a measly 0.55.
None of the teams in their group put up great attacking numbers, so it may be a case that their defence just hasn’t been properly tested yet, but an xG conceded of 0.55 is hugely impressive regardless.
Looking at standout players, centre-back Wendie Renard would be the first name worth mentioning, playing a key part in almost every goal in France’s games. Her retaken penalty, own goal and set-piece exploits are obviously the main headlines, but she was also an efficient ball progressor for France too – albeit not with a huge amount of progression.
Her pass map against Nigeria shows this, with just two passes, long towards the flank, being unsuccessful.
Elsewhere, midfielder Amandine Henry and left-back Amel Majri were their two players with the most progression p90. After having a look at Majri’s pass maps, it seems a lot of her progression comes from crosses rather than passes upfield, while most of Henry’s longer passes tend to be towards the flanks.
Ignore me being lazy and neglecting the key, but it seems a lot of France’s play revolves around the wings. Possessing wingers like Eugenie Le Sommer, Kadidiatou Diani, Delphine Cascarino and Vivane Asseyi it shouldn’t be too surprising, but the pass map from the Nigeria came showing passes into the final third shows a lot of passes out wide.
Then it was also similar against Norway.
Brazil had some big moments during the Group Stage of the World Cup, like Cristiane’s hat-trick and Marta’s record breaking penalty, but as a team their numbers were pretty middle of the road.
Their opening win against Jamaica was routine enough, with the xG being 1.5 to 0.4, and they were unlucky to conceded two flukes in the second half to Australia, but there wasn’t much attacking threat from Brazil in their two games against Australia and Italy. Against Australia they managed just 0.4 xG and 0.7 against Italy.
Defensively their xG numbers are pretty average too, conceding slightly more than the aforementioned Cameroon, although this isn’t too bad considering their group.
What’s interesting about Brazil (and France for that matter) comes when looking at their pressures. On the graph showing midfield pressure from above they have a high proportion of their pressure in the middle third, yet only a small portion of passes in the midfield area are made under pressure – with France in a similar predicament. It makes it unsurprising to hear they made the 5th lowest pressure events in the group stage.
This doesn’t have to be a bad thing, U.S.A. made the lowest and France the 4th lowest, but this is likely down to them playing weaker opposition who see a lot less of the ball. With Brazil playing evenly matched teams like Italy and Australia, it does seem a lot more odd.
Despite the lack of pressures, Brazil still maintain an average completion rate on progressive passes conceded, slightly better than England who have the 2nd most pressure events so far, which can be seen in the chart below.
Looking at where they conceded their progressive passes, it doesn’t bode well for Brazil. I imagine a lot of progression comes from the flanks for everyone, with teams putting more focus into defending the centre, but there are a grouping of passes going towards the left flank, as well as some crosses coming from the right flank.
With France playing a lot down the flanks, having Majri play a big part in the attack on the left and Renard progressing from left sided centre-back, they could find some joy on Sunday. Then, if Diani starts on the right, there could be opportunity for her to try and beat her marker and look to cross.
Going forward tfor Brazil the importance of Marta and Cristiane can’t be understated. Her numbers aren’t huge, but Marta is the player with the most progression, while Cristiane is their player with the highest xG p90.
Looking beyond those two, Ludmila has strong xG + xA p90 numbers, but only started in the last game against Italy, while left-back Tamires has the second most progression after Marta.
If Brazil want to pull of an upset they’ll need the likes of Marta and Cristiane to be on top form.
Amandine Henry To Dictate The Game
Henry has put up strong numbers for progression, completing the most progression for France. With Brazil not applying much pressure to passes in the midfield area it feels like a great opportunity to shine for Henry and really control the game.
The same argument could be made for Marta on the other side, but it feels like the collective strength of France will allow Henry to be the one who makes the bigger impact.
The method I’m using at the minute seems to award crosses a fair bit (I should probably filter them out to look at players for the next piece), but both left-backs Majri and Tamires have strong numbers for their respective sides. With a big part of France’s game being their wide play and Brazil allowing a few progressive passes down their right side, she could be important on Sunday.
France haven’t been hugely productive going forward and their strong defensive numbers are likely influenced by the opposition they’ve faced just as much as their defensive ability, but it feels like they’re the stronger side and should go and progress, possibly to a huge quarter-final tie against the United States.