15th September 2019
To try and have some more regular content this season I thought I’d try and give quick round-ups for the FAWSL after each matchday. I quite enjoyed doing a similar thing with the knockout round of the World Cup in the summer, giving just a few things that stood out and trying to see how/if it’ll impact future games.
As always, thanks to StatsBomb for giving this data out for free and also sorry to StatsBomb, because it was only this week I realised I’ve been using the old logo from GitHub on everything rather than the newer one with the radar.
On the surface, the top three – who were hugely dominant last season when looking at xG – starting the season with wins doesn’t feel particularly interesting, especially given two of them played newly promoted sides, but there were still some interesting points from the games.
The Manchester derby, played at the Etihad, was the main attraction of the weekend, both before the season started and also because it broke the attendance record for the FAWSL with over 30,000 people attending. Focusing on the action on the pitch however, it was incredibly close with few chances between the two sides.
This is pretty good from a Manchester United perspective, allowing just 0.4 xG away to the side who finished 2nd last season, while creating a strong chance at the other end that was well saved by Ellie Roebuck.
From a City perspective, it’s not so good. They moved the ball out of defence well, having the 2nd highest progression via passes as a team and Steph Houghton having the most of all players this weekend, but struggled once they got into more advanced areas.
If I worked this out properly, which is a big if, they had only the 7th most possessions reaching the final third and 6th most reaching the box while also having the lowest percentage of possessions reaching the box to end in a shot. Looking at the shot map, they managed only one shot inside the box, which was also of low quality.
It feels easy to point and say that City missed a striker but it did seem to be the case on Saturday. Janine Beckie led the league in progression via carries this weekend and got the ball into some good areas, but rarely had anyone to pick out once she got there.
It feels as though the four of Weir, Wullaert, Stanway and Beckie all want to do a bit of the same thing, it was rare one of them would be making runs in behind to stretch the defence or into the box should the ball be moved out wide in position for a cut back.
It’s only one game so I don’t think there’s any reason for City to worry just yet, especially as they scored seven in the week, but I would hope to see some improvement from the front four against Reading.
It feels like they either need more attacking full-backs, particularly on the left if Weir and Beckie are going to be more central, or go for a more conventional 4-3-3. Beckie and Wullaert can play wide while Scott and Weir can be attacking #8s who support Stanway down the middle and give City more presence in the box.
Full-backs Aoife Mannion and Demi Stokes can be more defensive/inverted, which feels as though it could suit them, particularly with Mannion playing centre-back rather than full-back for Birmingham last season. With the exception of Beckie’s position, their pass map wasn’t exactly far off this.
City got the result they wanted but you’d expect their performances to improve over the course of the season.
If they’re going to stick with the same front four until Ellen White is back they need to make sure they get bodies into goalscoring positions, with it feeling as though they’re more likely to hang around the box than get into strong goalscoring positons at the moment.
Quickly coming back to Manchester United, they gave a good account of themselves and will be disappointed to lose after restricting City to shots they did. It’ll be interesting to see whether the low xG for City came thanks to some good defending and organisation from them or some teething issues in this current City front four, although it likely lies somewhere in the middle.
Individually, Jackie Groenen had a strong game, having the most pressures this weekend as well as some decent numbers for progression via carries.
On the surface there are quite a few parallels between the Manchester derby and this game, both were played at the club’s main stadiums, both were derbies, both were a top team against a newly promoted team and both ended 1-0 thanks to a great left-footed goal from outside the area. However, looking at the numbers, Chelsea’s win over Spurs was fairly comfortable, thanks to lots of low to mid quality shots.
It’s worth pointing out that Chelsea still had the 6th highest xG per shot this weekend, but their first chance of the game looks to be their best one (which is annoyingly cut off on the FA Player replay).
While having fewer high quality chances is deemed better than multiple lower quality chances, this was a hugely encouraging performance from Chelsea – who still have players like Fran Kirby and Erin Cuthbert to come back into the starting fold.
Emma Hayes lined up with a diamond midfield, with Sophie Ingle at the base, Ji So-Yun and Drew Spence in the middle and Bethany England the tip. The pass map below shows some pretty good links with the back two and three in midfield, creating a square with Ingle in the middle
Emma Hayes lined up with a diamond midfield, with Sophie Ingle at the base, Ji So-Yun and Drew Spence in the middle and Bethany England the tip. The pass map below shows some pretty good links with the back two and three in midfield, creating a square with Ingle in the middle.
It worked quite well for Chelsea, it meant they had a gathering of technically proficient players in central areas to combine, while Hannah Blundell and Maren Mjelde provided the width.
In build-up the two full-backs could push forward and the likes of Ji and Spence could drop into space to the side of the centre-backs to pick up and then progress the ball. It helped Ji start the season with some great progression numbers, following on from last season, as she was the central midfielder with the most progression via both passes and carries, with most of her progressive passses coming from a deeper left position.
The full-backs also put up some strong numbers, with Mjelde having the best progression via carries per 90 for defenders and Blundell having the 3rd best.
Defensively they can be narrow, forcing the opposition out wide where they can press them. I’m not great at making these charts and it’s not compared to league average or anything, but below you can see a lot of Chelsea’s presures came out wide on Sunday, particularly on their left.
If the opposition do play the ball inwards, Chelsea still have numbers in the middle to crowd them out.
An example early in the game against Spurs can be seen below. Spence drops deep to pick up the ball from Magdalena Eriksson, who misplaces her pass but when Tottenham play the ball inside, Rachel Furness is quickly surrounded. Sophie Ingle stops her moving the ball forward as Beth England and Guro Reiten track back to stop her going backwards too. Reiten will be disappointed with her final ball, which should have really put Adelina Engman in on goal. (A GFY link if Streamable is geoblocked)
Chelsea look an exciting prospect with the diamond, even more so considering players like Kirby and Cuthbert didn’t start on Sunday, who feel as though they’d be well suited to this system.
While it wasn’t the result Tottenham wanted and their xG numbers aren’t great viewing either, Gemma Davison’s performance was a positive for the North London club.
She had the 4th highest progression via carries at the weekend as well as being progressive with her passing when she passed. I’m trying to find a way of measuring how progressive a player is when they have the ball and to try and weigh up who often gives the ball away when progressing it, as I currently only look at completed passes.
To try and do this I divided the number of attempted passes by progression via passes so it’s how many passes a player attempts per 1 unit of progression. It’s not a great way and I need a way to type it quicker but, looking at this, Davison had the 2nd best numbers for players this weekend, behind only teammate Kit Graham.
Away to Chelsea on the opening day you’d expect a newly promoted Tottenham side, with a huge number of new signings, to struggle but, while they didn’t create anything significant, they didn’t really look out of place either and caused a few problems in the second half.
They play Liverpool at home next week which should provide a good game to see what this Tottenham side are really like, particularly going forward.
I was quite surprised when I saw the xG figures for the Arsenal game. While Arsenal had some good chances, with Danielle van de Donk missing pretty much an open goal, some defensive lapses in the second half felt as though they gave West Ham a route back into the game and that Arsenal were quite fortunate not to throw the game away.
The xG timeline below shows West Ham had a few chances in the second half, but they almost flatlined after actually pulling one back – with the indirect free kick likely having a low xG due to the fact half of North London was standing on the goal line as Manuela Zinsberger charged the shot down.
The argument could be made that on another day Martha Thomas scores the chance where Zinsberger is caught in possession or Adriana Leon scores her back post header, but also on another day Danielle van de Donk would have scored her open goal to make it 3-0 and Arsenal cruise for the second half.
Arsenal may have had some lapses at the back but I don’t think it was anything more than a bit of opening day sloppiness due to a lack of match sharpness or new centre-back pairings getting used to each other but, with plenty of games coming up this month, you’d imagine these problems won’t last much longer.
Leah Williamson and Arsenal’s Midfield
With Leah Williamson being one of the first players to stand out when I first looked at the FAWSL data, she’s one of the players I’m most looking forward to watching properly.
She’s played an interesting role in the opening two games for Arsenal – including the Champions League game against Fiorentina. I did actually want to write a separate piece about how she lined up against Fiorentina but couldn’t get my hands on a download for the match so I’ll quickly mention it here instead.
Against West Ham she lined up as an orthodox central midfielder, as the pass map below highlights, but against Fiorentina she played a hybrid centre-back and midifeld role. It was almost a back three of Leonie Maier, Jennifer Beattie and Katie McCabe with Leah Williamson stationed in front of them forming a midfield with Jill Roord, Kim Little and Danielle van de Donk.
I’m intrigued as to whether both Williamson in midfield and the hybrid role she played against Fiorentina will be a regular feature of Arsenal’s play this year or whether it’s just being used to cover for the absence of Lia Wälti – although it’s an odd time to try it considering she had a long-term injury and is now back in training.
If it is the latter though, while Williamson and Beattie will be a good orthodox partnership, it will be slightly disappointing from a neutral point of view.
It’s a big risk to play her in the hybrid role and you can argue it limits the attacking influence of the full-backs, but it allows her to have more influence and use her passing ability slightly higher up the pitch.
In addition to this, it allows Arsenal to squeeze an extra midfielder in, without having to play one out wide, which is probably where they’re strongest. If they play a traditional 4-3-3/4-2-3-1 with Wälti holding they’ve got Van de Donk, Little, Roord and Jordan Nobbs up for just two spots in the team.
Having these players find space between the lines and combine for Arsenal is huge in their play.
Last weekend Danille van de Donk and Kim Little were first and second for passes received in the central area of the final third not under pressure. Then, looking at progression via passes for midfielders and forwards, Kim Little was 2nd and Leah Williamson 4th, while Little was also 2nd when looking at progression via carries for central midfielders, behind only Ji So-Yun.
To show some this in action and what Williamson brings slightly higher up the field I’ve included a few clips. The first shows her receieve the ball, she has a quick look forward but nothing is on so she spreads it wide, before getting it back and sliding it through for Kim Little, who then plays Lisa Evans in behind for a cross. (A GFY link)
I like the next clip because you see her point to where the player on the ball should go a couple of times, then when she gets it she shapes her body for a wide pass before going inside to Evans on the edge of the area, who lays it off to Roord for a shot. (A GFY link)
For the last pass, I’ve watched it a few times and I can’t decide if she meant for it to turn out how it did. If she did, it’s a great pass, first time with the outside of the boot and breaking the West Ham line. (A GFY link)
From what I’ve seen of this Arsenal side, they seem to a good job of showing @NathanAClark’s ‘players aren’t positions’.
They have players play different positions or roles, but I’m yet to see it feel like a player is playing out of position. It feels as though they’re just using that player’s attributes in a different area of the pitch, rather than asking the player to do something completely different.
It sounds a bit vague, but Williamson hasn’t looked any different in the three different roles I’ve seen her play, while Danielle van de Donk didn’t seem odd out wide against West Ham. It’s something I really like from what I’ve seen of them so far.
A Strong Debut For Martha Thomas
West Ham lost Jane Ross, their player with the highest xG last season, to Manchester United in the summer, but Martha Thomas gave an encouraging first game that may have them forgetting about Ross pretty quickly.
23-years-old and signed from French 2nd division side Le Havre, Thomas showed herself to be a focal point, having received the 7th most progression at the weekend, a hard worker off the ball, making the 2nd most pressures, as well as grabbing a goal for her side.
Like with the previous two teams, seeing how West Ham get on next week playing Birmingham will likely give a better indicator as to how they’ll be this season than an away game against Arsenal will.
The three games I’ve talked about are the three that I watched and I didn’t get around to watching the others on FA Player, although I did watch bits of the Bristol City vs Brighton and Birmingham vs Everton games.
With this in mind, I’ll just give a quick run down of anything that stood out in the numbers from these games. However, I haven’t really got anything to mention from Liverpool vs Reading, nothing jumped out in the data and I didn’t watch it, so I’ve left that out.
Ellie Brazil Impresses As Brighton Are Unlucky Not To Win
Looking at the numbers, Brighton seemed to do everything right against Bristol City except for score. They created the 3rd highest xG this weekend, not including the saved penalty, behind just Arsenal and Chelsea, while also conceding the 2nd lowest. Not only this, they also allowed the least progression with the worst completion rate on progressive passes for each third of the pitch.
Looking at individuals, the stand out player seemed to be Ellie Brazil – which, on a selfish note, was annoying because I was going to include her in the players to watch piece but took her out.
Brazil had the 9th highest progression via passes (highest for wingers and forwards), 6th highest progression via carries, 12th highest progression received and joint 3rd most pressures.
Birmingham Struggle To Turn Progression Into Chances
Birmingham seemed to have the same game as Manchester City at the weekend, except they didn’t get the result they wanted.
The West Midlands side completed the 3rd most progression in the league and had 17 shots, but only Manchester City had a lower xG per shot. This doesn’t really seem surprising when seeing just how far back some of their shots came from though.
You could chalk it up to Birmingham having a lot of change in the summer and it’ll take a while for the attackers to develop an understanding, but they had similar issues last season.
Looking at the xG per 1 unit of progression (again I hate when the units get to feeling vague like this) they had the 4th worst numbers in the division, suggesting they progress the ball pretty well, but struggle turning that progression into good quality chances on a regular basis.
One bright spark for Birmingham was 22-year-old Sarah Mayling. Last season she appeared at right-back and in midfield, but playing left-back against Everton she had the 3rd highest progression via passes in the league. New signing Brianna Visalli also had the 3rd highest progression via carries for central midfielders.
Stand Out Players
To finish off, I thought I’d just throw up the top ten for each of progression via passes, progression via carries and progression received from the weekend.
Progression via Passes
|Steph Houghton||Man City|
|Sarah Mayling||Birmingham City|
|Léa Le Garrec||Brighton & Hove Albion|
|Kerys Harrop||Birmingham City|
|Ellie Brazil||Brighton & Hove Albion|
|Keira Walsh||Man City|
Progression via Carries
|Janine Beckie||Man City|
|Ellie Brazil||Brighton & Hove Albion|
|Brianna Visalli||Birmingham City|
|Abbi Grant||Birmingham City|
|Georgia Stanway||Manchester City|
|Martha Thomas||West Ham United|
|Kayleigh Green||Brighton & Hove Albion|
|Janine Beckie||Manchester City|