Fixing Felipe: how to get the best out of Felipe Anderson
Felipe Anderson had a poor 19/20 season by anyone’s standards and, as a result, West Ham have apparently been looking to sell him to fund any moves they might make in the transfer market. That hasn’t materialised and, with a week left until the start of the new season, he’ll presumably have a part to play.
The point of this piece is to ask three main questions:
What’s gone wrong?
Well, that depends on your perspective. In terms of direct goal contribution, his first season in the Premier League stacked up pretty well:
Felipe Anderson’s first Premier League season ranks 13th in the last 20 years; only 3 players in the last decade have directly contributed to more goals in a season for West Ham than Anderson did in 18/19. Dimitri Payet, Trevor Sinclair, and Stewart Downing are the only non-strikers who are above him during this time period.
That was last season though and there’s undeniably a touch of “but what have you done for me lately?” about some of the criticisms levelled against Anderson. Even so, his performance in the 19/20 season represented a significant decline in these headline figures: just one goal and four assists.
But there’s more to football than scoring goals and the consensus with West Ham fans is that Felipe Anderson hasn’t lived up to his price tag. A pair of performances immediately following lockdown — a largely anonymous display against Wolves where he was given the hook at half-time and a hapless 20 minute cameo off the bench against Spurs — appear to have broken the camel’s back both inside and outside the club.
The Numbers
Felipe Anderson technically had 5 seasons at Lazio, but only managed ~500 minutes in that first year, so we’ll be generous and write that off.
In general, Anderson has been less productive in the Premier League than he was in Serie A. It’s worth taking Anderson’s last season at Lazio with a pinch of salt, as he was limited to just 9 starts:
His xG per 90 has been fairly stable despite a considerable reduction in Anderson’s shot volume (suggesting the shots he’s cutting out are low quality ones), whereas his hefty xA per 90 drop since moving to England has come in conjunction with a concurrent fall in key pass volume. (We’ll get onto the 15/16 drop in xA and key passes in a bit).
There’s a spike in Anderson’s passes into the final 3rd in his first season at West Ham (a product of a deeper playing role in the Premier League) but that dipped again this year. All of the other passing metrics have tailed off season-on-season. That reduction in high passes and in passes that were offside is intriguing and points to the different type of movement he’s had ahead of him at West Ham compared to Lazio.
On the dribbling front, the main thing to note is that he’s attempted and completed a fair amount fewer for West Ham than he did at Lazio and that in Serie A his dribbles started higher up the pitch.
There are, of course, some caveats here: the variation in Anderson’s output across his four seasons in Rome is a reflection of the various different roles he played for Lazio; his 17/18 and 19/20 numbers are based on significantly fewer minutes than his other seasons; those 17/18 numbers in particular are juiced a bit by the fact that he made most of his appearances off the bench; plus there’s potential league effects.
In short though, he’s shooting less, creating less, running with the ball at his feet less often at West Ham.
So let’s dig in to why that might be.
Why did we sign him in the first place?
If we go back to the goals and shots numbers for a second, we can see that there were basically two different iterations of Anderson at Lazio:
Under Pioli, he was more of a volume shooter, while under Inzaghi he was more of a pure creative type. Either way, he was someone who regularly made things happen in the final third and West Ham fans feel like they’ve not seen that this season from their second most expensive signing ever.
There’s not much we can get out of the publicly available performance data, so looking at something really basic like where on the pitch Anderson played his football helps shed some light on what’s changed over the years and helps us identify where things have gone wrong for Anderson at West Ham.
So let’s go through his Lazio seasons to get some context on the stylistic and tactical shifts in Anderson’s role.
2014/15
Stefano Pioli is in charge and plays a possession-oriented game, swapping between 4–3–3 and 4–2–3–1 shapes. 36-year old Miroslav Klose plays 1835 minutes upfront, scores 13 goals and gets 6 assists. Felipe Anderson, meanwhile, largely operates from the left or as a 10, with Antonio Candreva playing from the right.
I can’t find a heatmap for this whole season, so I’ve picked a few examples from a range of opponents to hopefully get a fair spread. This isn’t meant to be comprehensive — the purpose is simply to illustrate a general trend with a snapshot.
3–0 win vs Atalanta:
Playing as a 10, Anderson has over 100 touches (almost 20 more touches than any other player on the pitch) and is involved right across the attacking third of the pitch.
0–1 loss vs Genoa:
A few deeper touches this time, but they’re either tackles or passes from the first 20 minutes as Lazio feel their way into the game.
1–2 loss vs Roma:
In a derby game between two quality sides with a lot at stake (Roma finished 2nd, Lazio 3rd), Anderson crops up all over the place.
Here’s an overview of his role in the final third during this season:
He mostly creates chances by fizzing in low crosses from the right for Klose to poach. Alongside that, as the quickest player in this frontline, Anderson is often the one stretching the game by making runs in behind and regularly gets into the box himself.
2015/16
Pioli’s at the helm for most of the season so the back four is still in place. Anderson sees a drop in his game time thanks to the emergence of Keita Baldé, who plays a 1000 more minutes than the previous season, and Sergej Milinković-Savić, who arrives from Genk and plays an advanced central midfield role.
As this heatmap from SofaScore indicates, Anderson was still predominantly tasked with affecting play in the opposition third and was an occasional penalty area threat, but he didn’t get to the byline as often as he did in the 15/16 season.
Although he still had license to roam, Anderson’s withdrawn positioning on the right is mostly a product of the personnel changes: the inclusion of Keita and Milinković-Savić meant Anderson was no longer the most mobile player in this Lazio attack.
With more dynamic runners in the team around him, Anderson’s role shifted slightly:
He finds himself bursting beyond the last man less frequently and instead threads passes into the box for players making off-the-ball runs from deep. That goes some way to explaining the drop off in xA and key pass numbers — Anderson’s playing the pass directly before the shot less regularly in this role. There’s still a few elements of traditional wide play in the mix though.
2016/17
Towards the end of 15/16, Lazio got pasted 4–1 by Roma in the derby and sacked Pioli, putting former player and youth team coach Simone Inzaghi in charge on an interim basis for the rest of the season. Over the summer, Lazio appointed Marcelo Bielsa as head coach, but Inzaghi was given the job on a full-time basis when Bielsa resigned after less than a week in charge.
The other major personnel change is the signing of Ciro Immobile from Sevilla (for less than €10 million!), who goes on to score 23 goals. Inzaghi steadies the ship and guides his club to a 5th placed finish, mostly playing a 4–3–3 along the way with Anderson on the right, Keita on the left, and Immobile through the middle. Inzaghi’s eventual 3–5–2 shape does creep in towards the end of the season, especially against tougher opponents, and Anderson mostly gets used as a right wing-back (with surprisingly good results) in that system.
Anderson’s heatmap for this season is like an orthodox winger — predominantly gets on the ball close to the touchline, lots of touches down near the corner flag, touches in wide areas of the box. That little island of red in the classic number 10 spot shows what a varied attacking threat he was for Inzaghi this season, though.
At this point it’s worth mentioning that these heatmaps include defensive actions, which helps explain why Anderson has more touches in his own half here — at least 8 of his 33 starts came in that RWB role.
Defensive shape also has an influence. Look at Lazio’s defensive line in these two games vs Inter:
Same fixture, similar scenario in possession, 7 months apart, two very different approaches. Inzaghi’s midfield line defends at the point Pioli had his defensive line playing.
Under Inzaghi there’s also a significant shift in chance creation — less from passing combinations/dribbling into the box from wide areas and more from counter attacks/floated crosses:
Anderson still slips those passes through to players moving ahead of him when Lazio have sustained possession in the attacking third, but he’s carrying the ball into the box and shooting less himself. His relationship with Baldé Keita was especially productive.
It’s important to note how high up the pitch he receives the ball though, even in a team that defends in a low block and plays on the break. When he does come short, he lays it off quickly before getting up the pitch again.
2017/18
A few big things happen over the summer that result in less game time for Anderson:
- Lucas Leiva comes in to replace Biglia, who is poached by AC Milan, and Adam Marušić is signed to play as a wing-back, prompting Inzaghi to move to a back 3 permanently.
- Anderson picked up an injury over the summer so he misses the start of the season.
- The form of his replacement, Luis Alberto, coupled with the move to a wingerless formation, meant that Anderson was ultimately limited to 9 league starts in the league.
Playing as part of a strike partnership in a 3–5–2 sees Anderson spearheading counter attacks and acting as an outlet pass in advanced central areas:
That said, he had the freedom to roam out wide, usually opting to pick out Milinković-Savić with those lofted crosses when operating in those positions.
All of that is a long-winded way of establishing that Felipe Anderson consistently had an overwhelming majority of his touches the opposition half during his time at Lazio, even against strong opposition, and even when Inzaghi took over and implemented a more direct, reactive style of football.
Now contrast that general trend with where he’s taken his touches for West Ham:
2018/19
Pellegrini brings Anderson to Stratford and stations him on the left of a 4–3–3 shape for the majority of the season.
The main things to pull out of this are:
- the increase of touches deeper in his own half — the majority of these are passes/dribble attempts.
- the increased clustering of touches in the middle of the pitch either side of the halfway line
- few touches in the number 10 zone; many touches in the left halfspace/left corner of the box.
It’s worth looking at this season’s map alongside the 16/17 (his last as a wide player at Lazio) to get a better feel for how things changed:
This Pellegrini team tried to defend fairly high up the pitch, so our defensive structure shouldn’t have dragged Anderson deeper than Inzaghi’s low block Lazio did:
Despite Pellegrini’s possession-focused style, I think it’s significant how many of Anderson’s best moments in this season came from counter attacking situations:
It’s also noteworthy how deep he’s collecting the ball, even in these moments of transition, and how much better he looks when he has Arnautović and/or Antonio making runs in front of him.
2019/20
Pellegrini
Same story at the start of the 19/20 season.
This SofaScore heatmap covers the season as a whole and it’s a more extreme version of the 19/20 one: fewer touches overall thanks to injury and form but in general he’s much less involved in advanced areas and much more active closer to his own goal compared to his time at Lazio.
It probably says a lot about Anderson’s performance level and the team’s general dysfunction that the best examples of his play came in games against the 3 sides that were eventually relegated:
But, broadly speaking, he’s creating chances by clipping those crosses in from the left to Haller at the back post and again looks most dangerous when playing on the counter, which is somewhat antithetical to the style of football Pellegrini tried to instil.
Moyes
It’s difficult to highlight how his role altered after the managerial change; Anderson only started 4 league games under Moyes and only one of those was in a wide position (against Liverpool):
The Bournemouth touch map was reminiscent of the way Anderson played under Pioli, but the following week Moyes changed the system to match up better with Sheffield United and Anderson reverted to dropping off to get on the ball.
So how do we fix it?
To answer that, we need to figure out why he’s played so deep at West Ham.
It’s partly systemic — Pellegrini’s idea of playing out from the back was all about building through the flanks, with the full-backs moving in advance of wide midfielders who drop off to get in possession.
It’s also partly a personnel issue — West Ham don’t have anyone in the back half of the pitch who reliably moves the ball forwards with their passing, so Anderson has to do it.
That’s fine but it does mean that your most creative player ends up essentially playing left-back a lot of the time. You can see the effects of that here:
Even when it works and we manage to play through the press, Anderson has to cover loads of ground with his dribble to even get into the final third. On the one occasion Lanzini manages to break the lines with a pass here, Anderson is back in our own half so he can’t affect the game in a creative sense.
Contrast that with Lazio’s build-up play:
If we want to continue playing a more direct style of football, as seems likely under Moyes, Anderson absolutely has a part to play in that as a forward but he needs a Biglia-type deep midfielder (or a CB with passing range) behind him, so he can push on without having to worry about the ball getting to him. Rice showed hints of development on that side of his game in the post-lockdown fixtures, but it remains to be seen whether that will continue. He’s unlikely to be around much longer anyway, so another solution is required.
One option might be to re-purpose Lanzini into that role, although maybe his slight physique would limit his effectiveness there. If we’re looking at signings for that job, that’s an entire article in itself, but off the top of my head, some cheap-ish players like that would be:
- Expensive: Lewis Cook
- Cheaper: Youssef Aït Bennasser
- Ultra budget: Jonjo Shelvey
One of Felipe Anderson’s biggest issues at West Ham was being fundamentally miscast by Manuel Pellegrini: he’s a winger, not a wide playmaker.
That might seem like splitting hairs or some semantic bullshit, but I think it hits on an important distinction. Perhaps Pellegrini’s choice to get Anderson progressing the ball was one made out of necessity, but watching him play for Lazio makes it clear that Anderson’s at his most effective when he’s receiving the ball in the final third, especially when he operates on the right wing.
I don’t think Anderson’s actually a particularly great passer in deep areas; he sees the big picture and his decision making is generally quite good, but his ball-striking technique on his passing is erratic and he’s better when he’s passing on the move or when he has players making runs into the box around him. Factor in his own intelligent off-the-ball movement and a lack of press resistance on top of that and you’ve got someone who doesn’t have a skillset suited to orchestrating play from the half-way line.
But if we are going to have him do that, the least we can do is to put players in front of him who complement Anderson’s game.
The only match last season that Anderson, Antonio, and Haller started together was the opening day defeat against Manchester City. Aside from that (+ a pair of injury time sub apps for Anderson vs Arsenal and Southampton), the only other time all three were on the pitch at the same time was a 23-minute spell against Watford. Haller scored twice in that period.
Injuries have played their part and there’s presumably a reason neither manager has felt able to play all of them at the same time without unbalancing the side. But at Lazio, Anderson created his best chances with Baldé Keita and Immobile making darting runs in behind the opposition backline and with Milinković-Savić providing a big target in the box to cross the ball to; Antonio and Haller provide a close approximation of those passing options for Anderson. Getting all three of our best attackers on the pitch at the same time seems like a no brainer.
Anderson only started 4 games in total alongside Antonio in 19/20 (1–0 win vs Chelsea, losses against Wolves, Liverpool, and Arsenal) and his only appearances of any kind with Antonio against non-top 8 sides were the last 15 mins in a loss vs Palace and the aforementioned 23 min spell in the Watford win. Anderson’s had very little opportunity to link up with Antonio playing as a striker and it’s a combo that would benefit both of them.
Jarrod Bowen, whose game in the Championship was predicated on clever off-the-ball movement, only played with Anderson in the last 20 mins vs Spurs and the first half vs Wolves post-lockdown. Tomáš Souček has played more than 10mins with Anderson twice and he’s basically Milinković-Savić if you stripped away any semblance of technical quality; giving Anderson the opportunity to supply him from out wide could prove fruitful.
Conclusion
At Lazio, Felipe Anderson proved himself to be a different, more versatile attacker than the one we’ve seen at West Ham, one who played both as a wide forward with direct goal threat under Pioli and also as a final third creator on the right in a transition-based attacking system under Inzaghi.
And I think that gives us an insight into Felipe Anderson’s major limitation as a player: he’s a facilitator who can adapt to help his team’s attack function but he’s not a one-man attacking system by himself. He’s not Dimitri Payet.
That’s fine. Given our status as a club, we have to accept that players we sign are going to have significant flaws. But we’ve amplified those flaws by asking Anderson to do a different job — one he’s not particularly suited to — compared to the one he excelled at with Lazio.
I can understand why other West Ham fans feel disappointed. When you spend that kind of money on a player, you expect them to be a game changer by themselves but that’s not the reality of Felipe Anderson. He needs to right pieces around him to be effective.
The club have seemingly given up on him, which is a shame, because if we can get Felipe Anderson playing higher up the pitch (preferably on the right hand side), play him with complementary attacking players, and find someone else to take responsibility for progressing the ball during our build-up play, we can get the best out of a talented but misused player.
