Window Shopping: Where Do Arsenal Go From Here?

Taking over after Arsene Wenger’s 22 year long reign, it was always going to be a tough season for Unai Emery. Despite this, on the surface the Spaniard’s first season in charge doesn’t look too bad – they finished with 7 more points than last season and still have a route into the Champions League with the Europa League final. However, with some poor underlying numbers, a squad in need of change and a fairly tight budget, it seems Arsenal, and Emery, have a lot of work still to do to get Arsenal back on the right track.

Before continuing, I’d reccomend reading @MoeSquare’s season review for Arsenal over on StatsBomb, which does a much better job of analysing Arsenal’s season and particularly their underlying numbers.

Poor Underlying Numbers

Arsenal’s poor underlying numbers have been a big talking point this season, as they were overperforming at both ends of the pitch. They scored more than expected and conceded less than expected.

The Gunners ended the season with the 3rd most goals in the league, scoring 73, but had the 6th highest xG in the league with 58.7. This overperformance of 24.3% was the highest in the Premier League last season.

At the other end of the pitch, things weren’t quite as drastic, with Arsenal conceding 6.2% less than expected, the 6th highest overperformance in the division. However, it’s worth pointing out that @MoeSquare’s review is a better read for the expected goals numbers, particularly with the goalkeeping numbers for Bernd Leno showing just how he overperformed this season.

Now, you sometimes see an argument that overperforming isn’t that worrying or that the best players tend to overperform their expected numbers which is why they’re the best players. This could be backed up by the fact that both Liverpool and Manchester City also ran hot at both ends of the pitch this season. However, the difference when comparing them with Arsenal is that their underlying numbers were good and then their actual numbers built on top of them. Arsenal’s actual numbers seem to cover up their underlying numbers rather than building on them.

Arsenal conceded more shots than they took this season, which seems alarming for a top six side. They averaged 11.63 shots per game, while their opponents averaged 12.18. They did manage a positive xGD, but with just +4.34 it was only the 9th highest in the division, placing them firmly in mid-table. Wolves, Leicester and Everton all finished the season with a better xGD than the Gunners.

Splitting these xG numbers into home and away makes for an interesting contrast. At home Arsenal averaged an xGD of +0.57 per game, but away from home their xGD was -0.34 per game. It feels almost impossible that a top six side could be putting up these numbers. It’s not particularly surprising, given they only kept one clean sheet away from home all season, but it’s definitely a huge area of concern.

To top all of this off, despite the extra seven points, Arsenal have worse underlying numbers than last season. Their xG production has gone from 66.03 to 58.73, while their xG against went from 47.94 to 54.39.

Arsenal have the underlying numbers of a midtable club and it feels as though serious change is needed if they wish to compete with the rest of the top six.

Results vs Process

Many Arsenal fans may hear about the poor underlying numbers and shrug it off. If Arsenal win the Europa League final and are back in the Champions League come next season, who cares about their xG from this season? Their goal was to get back into the Champions League and they’ve done it.

To an extent, this is understandable. You could argue that getting into the Champions League, having the financial benefits that come with it and being able to attract more players means they can start to take bigger strides forward and this season can be wrote off, but it makes it feel as though short-term thinking would be firmly at the wheel. It feels as though Arsenal would be building on the fly and doing all they can to avoid dropping out of the Champions League once more, neglecting any kind of long-term planning just to make sure they’re in the top four come May.

This is where I think there’s a legitimate criticism of Emery: He hasn’t seemed to bring a distinct style of play to the club. Many people have compared Emery’s first season points tally to the first season of Pep Guardiola or Jurgen Klopp, but these comparisons don’t mention either the underlying numbers or the process of the managers.

Pep Guardiola’s first season in charge of Manchester City was disappointing, but they still had the best xGD in the league. So Pep’s process, even in the first season, was working. They produced the most xG and conceded the 2nd least xG in the league. The process from the disappointing first season to the record breaking second season didn’t change (for the most part), but they signed players that were better suited to Pep’s demands, refreshing an old squad in the process.

Arsenal don’t have the spending power of City, so the process won’t be as quick, with City overhauling their squad in the space of two seasons, but that makes Liverpool an interesting comparison.

Jurgen Klopp at Liverpool is similar to Guardiola at City, in the sense that he’s always had good underlying numbers in his time there. It hasn’t been a surprise to see them go from strength to strength as they’re able to build on these good underlying numbers, finding players that compliment and enhance the team. However, the investment seems to have been mostly funded by raising money through player sales, rather than having the oil money of Manchester City.

It’s easy to get bogged down with words like process (after writing the above the word has lost all meaning to me) and turning it into some kind of ideological debate, but at the end of the day the idea of a good football team is simple. A good team is one who creates good goalscoring opportunities and restricts the opponents good goalscoring opportunities. Teams and managers will all have different ways of going about this, but that’s all it comes down to. xG is just a nice way to quantify this and look at patterns over time. Even before they were as strong as they are today, City and Liverpool were doing this. They had problems they needed to address, but in general they were creating more xG than their opponent per game and it seemed only a matter of time before this translated into actual goals and then points.

Emery has the reverse of this situation. Whatever Arsenal were doing last season didn’t work. Their underlying numbers were poor. It’s not just a case of fitting better or better suited players into their current set-up, especially because it still feels like there isn’t really a current set-up. They’ve tinkered with styles and formations throughout the season, making it hard to say exactly what they need in the window.

When Emery was hired it seemed as though he would bring more organisation to Arsenal, both with and without the ball. It seemed that they’d become more structured in possession, helping them progress the ball and break down defences more efficiently and also look to win the ball back quicker upon losing it.

This doesn’t really seem to be the case though. Nothing about them profiles as vastly different to 2017/18. Their attacking production is down, they take slightly less passes per possession, while the opposition takes marginally more passes per opposition and Arsenal register a slightly higher PPDA compared to last season.

There is an argument that Arsenal didn’t necessarily want a huge stylistic shift, which would be fine if Emery’s first season had better underlying numbers. The bad underlying numbers just makes it feel as though something about what Arsenal are doing isn’t working and therefore needs fixing. I can be quite stubborn in thinking individual players aren’t going to make a big difference to a sides fortunes (I mean, I was one of the doubters of the Van Dijk deal) but I struggle to see how signing a few players this summer is going to make Arsenal significantly better, when it feels a lot of the issues are still systemic.

Now, this would be the point to talk about why or how Arsenal are bad and need to improve, but I’m going to weasel out. Analysing video and tactical things isn’t really a strength of mine.

For me, Arsenal need to be thinking long-term. It’s not a terrible first season from Emery, but it feels as though next season there should be a much more consistent style of play, with players brought in for that style of play, ideally leading to better underlying numbers.

The Squad

Growing up, it always felt like Arsenal were known for having a young squad, making it feel odd to talk about how Arsenal now have quite an aging squad.

New arrivals Lucas Torreira and Matteo Guendouzi help the problem and have both had good debut seasons, but it is worrying to see just how many players are deemed beyond their peak years on the above graph by @Worville.

It feels as though Arsenal should be looking for younger players who will help freshen their squad, but they still need players who can come in and make an immediate impact.

Before getting into potential targets, however, it’s worth looking over the current squad, highlighting a few talking points.


Arsenal’s defence is probably where they have the most peak age players, with Hector Bellerin, Shkodran Mustafi, Sead Kolasinac and Bernd Leno in goal. Despite this, it’s also the biggest area of concern for Arsenal.

From the above names, Bellerin and Leno are the only two you’d say Arsenal should 100% keep. Leno has had a good season and Bellerin fits the role of a modern full-back well, hitting a good stride of form before he was injured.

Mustafi and Kolasinac are two more interesting names. Mustafi is often criticized for making individual errors or having lapses in concentration, while there’s question marks over how good of a fit Kolasinac is and how well-suited he is to a back four.

I find it hard to assess centre-backs, but I do think there’s a good case to be made for selling Mustafi and re-investing the funds into another young to peak age centre-back. It doesn’t feel like he’s had the effect people hoped for and it doesn’t feel like he’s going to kick on and become the centre-back Arsenal fans were hopeful he would be.

Arsenal would then have two experienced options in Sokratis and Laurent Koscielny, two young to peak age options in a new signing and Rob Holding, while also having Kostantinos Mavropanos and Nacho Monreal (who I think they should keep hold of) available should they play three at the back.

However, there is also an argument to sell Laurent Koscielny, who’s now 33-years-old and has injury troubles. Doing so would likely mean that a second centre-back would also have to be signed.

Kolasinac is more interesting. In Europe’s top five leagues he had the 7th highest xA p90 of all left-backs, as well as the 6th most deep completed passes p90. However, @SaturdayOnCouch found Arsenal’s left side poor when it comes to stopping the opposition gaining yardage, making it seem as though they could definitely improve upon Kolasinac.


Moving into midfield, Arsenal seem pretty settled. Torreira and Guendouzi were two good additions last summer and should continue to grow, while Granit Xhaka is still in his peak years and is generally the best choice at Arsenal for progressing the ball, as he had the 9th most progressive passes p90 in the Premier League.

However, I wouldn’t say Xhaka is indispensable and should the right offer come in, it’s definitely something Arsenal should consider. He has flaws in his game, particularly with his lack of mobility and lapses defensively, but Arsenal do tend to look worse when it comes to progressing the ball when he doesn’t play.

The promise of Torreira and Guendouzi gives Arsenal some wiggle room here. They could sell Xhaka and re-invest in a cheaper, more experienced player (you think of the effect that someone like Joao Moutinho has had at Wolves) who can give them a short-term improvement without hindering the progress of the younger options.

I think this summer it would be best to keep Xhaka, unless someone makes an offer they can’t refuse.

Arsenal also have Mohamed Elneny, who hasn’t featured much in the Premier League this season. Elneny tends to be pretty efficient in possession, but not particularly progressive. With them having better options it feels time for them to move Elneny on and use the funds elsewhere.

Mesut Ozil

Ozil has dominated the headlines for Arsenal this season, frequently finding himself left out the starting eleven or squad, even when there doesn’t seem to be any injury problem. There were stories that it was due to his lack of effort in the defensive phase of play, which is a frequent criticism of Ozil’s game, but that doesn’t feel like a legitimate reason to leave him out.

My argument here is based on the fact that Arsenal were worse defensively than in recent seasons, in which Ozil has been a key player. Leaving Ozil out to benefit Arsenal without the ball would feel justified if Arsenal transformed into a defensively solid team without him, but instead they remained a defensively poor team, now also lacking in creativity.

If Arsenal remain with the 3 at the back formation, there’s definitely space for Ozil. Arsenal will still have three centre-backs and two midfielders, meaning there’s plenty of bodies behind Ozil, while when Arsenal press Ozil can line-up centrally and the ball-side striker can press if the ball is played wide.

With that being said, Ozil had the least productive season (looking at xG + xA) of his Arsenal career last season and he’s now 30-years-old. Not to mention, it feels like for him to truly thrive the system has to be built around him, allowing him to drift into pockets of space, pick up the ball and look to play people in.

If that isn’t going to happen under Emery, it feels like it would be best to move him on, given the substantial wages of his renewed contract. There doesn’t seem to be much logic in paying him so much, if you’re not going to get the best out of him.

Arsenal and Emery need to make the choice of whether or not they’re going to give Ozil the tools to succeed and make him the key figure of the attack, or sell him and reinvest the wages elsewhere.

Alex Iwobi and Henrikh Mkhitaryan

I find Alex Iwobi an interesting player. He often pops up as having impressive numbers, but when watching him it always feels like he’s capable of doing more.

Last season, for players 24-years-old and younger, Iwobi had the 5th highest xA p90 in Europe’s top five leagues. Only Joshua Kimmich, Raheem Sterling, Jadon Sancho and Julian Brandt put up better numbers than him. He also had the 3rd highest passes into the penalty box p90.

I wrote a piece at the start of the season about Iwobi that I didn’t publish, but it was about how 2018/19 felt like it could be a big season for him. It felt like it was his chance to really stake his claim for a starting role and step up from being a good link player to being a key provider in the side.

While many will claim he didn’t step up in this way, he did have the highest xA p90 in the Premier League for Arsenal last season and the 7th highest in the division.

For me, Iwobi’s future depends on what formation Arsenal view as their main one. If they view the 3-4-1-2 shape as their main one, there is an argument to sell Iwobi. Despite the numbers, being the main creator in the #10 role doesn’t feel like it suits Iwobi as much as being wide and linking up with a #10 and an overlapping full-back.

If Arsenal move back to 4-2-3-1, he’s a solid choice on the left, while still having three other attacking players to contribute, should he still not be too consistent or not offer a huge goal threat.

I’m a big fan of Iwobi, so there is some bias here, but I think if I was Arsenal I’d rather keep him. He’s putting up some of the best creative numbers for his age group in Europe and even if he doesn’t kick on further, he’s a useful squad player to have.

Moving on to Mkhitaryan, I think the best option would be to sell. While he’s put up some good numbers for Arsenal, he’s now 30-years-old and it seems a good time to cash in and re-invest.

If Arsenal play 3-4-1-2 it gives them all of Ozil, Mhkitaryan and Iwobi for the #10 role, while in a 4-2-3-1 it feels as though as ‘3’ of Iwobi – Ozil – Mkhitaryan feels too samey. It feels as though they all want to do a similar job and the side lacks any kind of pace or running in behind. Selling Mkhitaryan and replacing him with a quicker, more direct winger feels as though it would be a positive move.


Many people would class Arsenal’s strongest position as their forward line. Pierre-Emerick Aubameyang and Alexandre Lacazette have built a good partnership and are probably the main reason why Emery would consider keeping the 3-4-1-2 shape.

Aubameyang ended the season sharing the golden boot and having the 3rd highest xG p90 in the Premier League and a big part is down to his strong partnership with Lacazette – who puts up good numbers, but they don’t jump off the page as much as Aubameyang’s.

The only real question is whether or not one should be sold to raise funds elsewhere, and again, it all depends on what formation Arsenal view as their main one.

If Arsenal view the 3-4-1-2 as their main choice, it makes sense to keep the pair of them and improve the rest of the team to truly reap the benefits on the pair. If Arsenal want to play 4-2-3-1 they could sell one of them rather than have them play a small role from the bench or be pushed out wide.

The question then becomes which one would you keep if that were the case. It’s a tough question, as Aubameyang has been the more productive, but Lacazette is at a better age. I’d say it’s just a matter of preference and to assess which one is better at leading the line on their own.


Arsenal are in a funny position. I remember thinking at the end of last season that it’s odd because it feels as though there’s an argument to be made for selling any of their players. No one felt truly indispensable. I’d argue the younger additions of Torreira and Guendouzi should now be kept at all costs, while the likes of Bellerin and Leno should only be sold if it’s a crazy offer. Outside of those four players though, it does feel like you could make an argument for selling just about anyone.

Going into next season, taking into account the fact that Aaron Ramsey is on his way out and Danny Welbeck looks as though he could be too, this feels like Arsenal’s current squad situation. (I forgot about Reiss Nelson coming back from his loan at Hoffenheim)

Of course some players can play more positions than listed and I wouldn’t put any weight into the ordering of names, but it feels like this is what they’re working with.

My brief run down of what they could do depending on the formation would be:


  • Sell squad players like Chambers and Elneny to raise some funds
  • Consider selling Koscielny and signing a younger centre-back
  • Sell Mustafi and Kolasinac to raise funds for improvements to these positions
  • Sell Mkhitaryan and buy a winger or two who offer a different threat
  • Look for a versatile forward player to fill the Welbeck hole
  • Only sell Xhaka or one of the forwards if an unbelievable offer comes in


  • Sell squad players like Chambers and Elneny to raise some funds
  • Sell Mustafi and Kolasinac to raise funds for improvements to these positions
  • Consider selling Koscielny and signing another centre-back
  • Possibly sign another peak age centre-back
  • Only sell Xhaka if an unbelievable offer comes in, reinvest in an experienced midfielder and possibly another younger option
  • Sell Mhkitaryan to raise funds
  • Sign a rotation striker to fill the Welbeck hole

It’s mostly the same, the only real difference is if Arsenal commit to 3-4-1-2 they don’t need to worry about wingers, but may benefit from another centre-back.

From this point on I’ll be using TransferMarkt values as a rough guide on player prices. Using these prices Chambers and Elneny could fetch around £20m, add in a £75m budget (I just read that would be the budget if they qualify for the Champions League) and you’ve got a healthy ~£100m without selling any regular players.

So, moving on, who could Arsenal be looking at?

Potential Additions

An Improvement on Kolasinac

Again, I’m not great at judging defenders, so it’s hard for me to say if a player will offer significant improvement in defence for Arsenal. Given the main concern with Kolasinac is defensively, this isn’t exactly ideal.

Three names that seem to stand out as representing good options are Alejando Grimaldo from Benfica, Angelino from PSV and Alex Telles from Porto.

@MoeSquare mentioned Angelino in their review and he seems like he could be the best option for the Gunners. Looking across all competitions last season, with players who have a TransferMarkt value of €5m or more (to weed out the lower and smaller league players), Angelino had the 7th highest xA p90, with only 0.01 less than Kolasinac.

He also managed the 2nd most passes into the penalty area p90, behind only Benjamin Mendy, and the most deep completed crosses p90. He does fall short on deep completed passes, but there’s a chance this is down to team style and at Arsenal there may be more opportunity for deep completed passes.

His defensive duel numbers tend to be around the same as Kolasinac, but what’s encouraging is that PSV had the best defence in the Eredivisie by some distance last season, in which Angelino played every available minute.

He only joined PSV last summer and still has four years remaining on his deal, so is unlikely to come cheap, but if they could sign him for under £20-25m it definitely seems like a deal worth taking – particularly if they can raise £10-20m selling Kolasinac.

Grimaldo doesn’t have as impressive numbers, falling behind Kolasinac when it comes to xA and deep completed passes and crosses, but could still be an option worth looking in to. He has two years left on his deal, is still only 23-years-old and, like Angelino, played every league minute of 2018/19.

His defensive activity and success rate is around the same as both Kolasinac and Angelino, although Benfica’s defensive numbers aren’t quite as impressive as PSV’s.

Moving onto Telles, I initially assumed he’d be either out of Arsenal’s price range or that bigger clubs would come in for him. After a quick search the main clubs mentioned as looking at Telles are Atletico Madrid, Chelsea and Leicester, which makes it seem as though Arsenal could be a realistic option.

Telles has had two hugely productive seasons with Porto. He had the 4th highest xA p90 of the left-backs in the sample in 2018/19 and the highest in 2017/18. He’s in his peak years at 26-years-old and has two years left on his deal.

He does take set-pieces and I’m not sure how big of a proportion of his xA that makes up, particularly as he doesn’t jump off the page when looking at deep completions.

Again, nothing about his defensive numbers is vastly different to that of Kolasinac’s, but, like Angelino, he comes from a side with impressive xG numbers. Porto had the best xG in the Primeira Liga last season, with even better numbers than PSV had in the Eredivisie, and Telles played 95% of the available minutes.

Telles feels like the best immediate option, but would likely also be the most expensive. Angelino is my personal favourite option for the mixture of his numbers, his age and potential cost, while Grimaldo doesn’t stand out as much as the other two, but is definitely an option worth looking into more.

An Improvement on Mustafi

As I’ve mentioned, I find it really hard to find centre-backs using data. A lot of what makes a good centre-back, things like how they cover and defend space aren’t really quantifiable with event data.

One name that may be worth looking into, however, is Jonathan Tah from Bayer Leverkusen. Despite being just 23-years-old, Tah already has around 10,000 minutes played in the Bundesliga, while Leverkusen conceded the 4th least xG in the Bundesliga last season. Tah also played the 7th most passes into the final third p90 for centre-backs under 27-years-old with a TransferMarkt value between 5m and 120m (which I know makes it sound stupid and weirdly specific).

While I don’t put much weight into the reliability of these stories, there is an article stating that Juventus and Atletico Madrid want Tah and he could be open to leaving for around €30m . If Arsenal were to sell Mustafi for around £20m, it definitely seems worth putting the extra £10m into someone like Tah.

A name that Arsenal have been linked with a lot is Samuel Umtiti from Barcelona, who is apparently deemed surplus to requirements there. I’m a big fan of Umtiti’s and at ~£30m he seems like he’d be a great buy, particularly given he’s still only 25-years-old. However, there are stories out there that he has troubles with his knees and it seems a risk to sign him if that is the case. Of course, this isn’t my area at all, but I would be hesitant to spend big on a player if his knee problems are as bad as reported.

Elsewhere, Filip Benkovic has some great passing and aerial duel numbers on loan at Celtic from Leicester, but it seems like Leicester would demand a big fee if they were to sell him, while he does’t really seem like an immediate upgrade on what they’ve got.

A name I mentioned last season that Arsenal could look at is Aissa Mandi from Betis. He has the 2nd best success rate for passes into the final third of centre-backs in the sample and the best long pass accuracy, while still playing the 15th most p90. However, Betis didn’t have great defensive numbers last season.

There’s also RB Leipzig pair Dayot Upemecano and Ibrahima Konate, who both are still incredibly young. Their individual numbers don’t stand out as much, but Leipzig had the 2nd best xG conceded in the Bundesliga last season.

Arsenal have also been linked with 18-year-old William Saliba who recently broke into the St. Etienne team, but I haven’t seen anything of him so I can’t comment on that.

There’s also Martin Pongracic from RB Salzburg who pops up as looking good with passing numbers and defensive duel success. He’s only 21-years-old and be a good younger option, should Arsenal move both Mustafi and Koscielny on and sign two centre-backs.

Lille captain Adama Soumaoro may also be worth looking into. He only played 50% of available minutes this season, but he comes up good when looking at long pass accuracy and duel success, while Lille conceded the least xG in Ligue 1 in this data. He’s 26-years-old with two years left on his deal so may be available for a fair price.

It’s hard for me to say much about these options without looking at a lot more of them individually, but there’s definitely options out there for Arsenal.

Options for Wingers and Attacking Midfield

‘Premier League Proven’

While it can be fun to make fun of the ‘Premier League Proven’ tag, there’s no denying that signing a player who has already performed in the division can eliminate some risk in the deal. The trade-off is that it seems to rarely offer good value. With that being said, Liverpool have found some good deals in the Premier League in the past few seasons, while Arsenal have been linked with a few Premier League attackers, making it worth looking into some options.

Two names that Arsenal have been linked with in recent weeks are Ryan Fraser and Wilfried Zaha.

With the data I’m using, which must measure things different to how Opta do, Ryan Fraser ended the Premier League season with 13 assists, the joint most in the Premier League and the most total xA in the division – with the 4th highest p90.

He’s just turned 25-years-old, only has a year on his deal and has some strong creativity numbers, which makes it seem as though he could be a good value pick-up.

He played the most crosses in the Premier League last season and takes set-pieces, so I’d be intrigued to see what proportion of his xA came from these two sources, but if he could be signed for under £15m it does seem as though he could be a useful addition to Arsenal’s squad – particularly if they opt for 4-2-3-1.

Comparing Fraser to Zaha is interesting. It feels like if you asked most people who they thought was the better player, the more popular answer would be Zaha – and there’s plenty of reason behind this. Zaha is hugely talented, but until 2017/18 didn’t seem to transfer that talent into strong output numbers.

This season his xG figures have dropped from 0.32 p90 to 0.18 p90, while his xA p90 has remained around the same, going from 0.2 p90 to 0.18 p90. On the ball and in 1v1 situations you’d favour Zaha to Fraser, with Zaha attempting the 2nd most dribbles p90 in the Premier League.

Given this was Fraser’s first year with the kind of numbers he’s putting up, there’s a chance it falls again next year, like Zaha’s xG numbers from 2017/18 to 2018/19. If they were the same price it’d be easy to lean towards Zaha, it feels as though surrounding him with better players and a more attacking system could see him produce a better output. The problem is that the reported fee for Zaha is much larger than Fraser with one year left on his deal, not to mention Fraser is almost two years younger.

It feels like there’s better value out there than what Zaha would cost. Zaha feels stuck in a spot where it seems he has enough quality to move to a bigger club, but the deal doesn’t represent good enough value for a big club to go through with.

These are the main two options that Arsenal have been linked with, but are there more players out there in the Premier League the Gunners could be looking at?

One option could be Diogo Jota from Wolves. Only 22-years-old the Portuguese forward enjoyed a good first season in the Premier League, albeit after a slow start. Looking at those 24-years-old and younger, Jota attempted the 4th most dribbles p90, had the 8th highest xG p90 (4th outside top six) and an xA of 0.16 p90, which is respectable but nothing outstanding.

Jota could also function in both the 4-2-3-1 system and the 3-4-1-2 system, having played centrally and from the left.

Like Zaha, my issue isn’t with the player. I think Jota would make a strong addition to Arsenal, but there’s no reason for Wolves to sell, meaning they could demand almost anything they wanted. They’re fine financially and they’ve just qualified for Europe, making it feel as though it’d be a tough task to get them to sell one of their most promising players.

Elsewhere, there’s Ryan Sessegnon who put up some respectable numbers playing for a poor Fulham side – while also going into the last year of his deal. His xG of 0.15 p90 and xA of 0.18 p90 aren’t too far away from Zaha’s numbers this season, yet he’s eight years younger and was playing for a relegated side.

Not to mention, Sessegnon could also be an option for left wing-back should Arsenal play with three at the back, while also being an option for left-wing if they don’t. While it seems Arsenal need someone who is more likely to make an immediate impact and improvement, if he’s available at a cut price due to the relegation and contract situation, it could be worth taking the risk.

Immediate Impact

While I’ve mentioned how Arsenal should be looking at younger options, if they are to move on the likes of Mesut Ozil and/or Henrikh Mkhitaryan, they’d most likely to be looking at bringing in players who can hit the ground running.

A player that has been linked with Arsenal this summer is Hakim Ziyech. The Ajax midfielder has had a great season and is reportedly available at a reasonable price.

Looking at midfielders across all competitions (with a TransferMarkt value between 5m and 120m) Ziyech shines in almost every passing category. He has the 4th highest xA p90 (0.41), the highest passes into the penalty area p90 (7.95), the 4th highest through balls p90 (4.11) and the 5th highest deep completed passes p90 (4.21).

A criticism of Ziyech’s game has been his poor shot selection, and it’s true he does like a long shot, but he still managed an xG of 0.33 p90 across all competitions in 2018/19. For comparison, across all competitions, Alexandre Lacazette had an xG of 0.41 p90, so he’s also getting into good goal scoring positions, rather than his only source of shots being long shots.

He also attempted the most dribbles p90 in the Eredivisie this season, making it seem as though he excels at just about everything you want your attacking midfielder to do.

He’s spent much of this season on the right and, while he’s not the direct winger threat I mentioned earlier, he’d offer a big upgrade on Mkhitaryan and is reportedly available at a reasonable price. If Emery does move Ozil on, he could also be a good choice as his replacement in the #10 role in the 3-4-1-2 formation.

If the £25m release clause that’s being reported is true for Ziyech, I think he should be a top priority for Arsenal this summer. He’s still 26-years-old and would be able to make an immediate impact, while offering good value.

Another option that could interest Arsenal is 27-year-old Pablo Sarabia from Sevilla. Like Ziyech, there’s some articles floating around that suggest he has an incredibly low release clause – €18m is the figure I’ve seen. Not to mention, he’s also entering the last year of his contract, so even if this figure isn’t true, there’s definitely a chance of a good deal.

Sarabia has played both on the right and centrally during his time with Sevilla and has produced respectable numbers throughout. This season has been his most productive, however, with 0.39 xG p90 and 0.31 xA p90 in all competitions.

His other numbers don’t jump off the page as much as Ziyech’s do, but they do tend to be above average. It’s also worth pointing out that he does take set-pieces, which may skew his xA, but he does seem to have plenty to offer other than set-pieces.

The below graph shows the NPxG + xA p90 and passes into the box p90 of wingers across all competitions with a TransferMarkt value between 5m and 120m. You can see Ziyech in an Eredivisie triangle at the top and Pablo Sarabia isn’t too far away.

At 27-years-old Sarabia wouldn’t be much of a long-term option, but if he is available for around £15m it could be a great deal. He seems as though he could play on the right and form a good partnership with Bellerin, as Sarabia cuts inside and looks to pick up the ball in the half-space area.

He’s comfortable with both feet, he can score, create and if Arsenal do alternate between a back three and four, he can fit into both systems.

While there’s a lot more names out there, these two feel like a good trade-off in value for money, quality and age and should both be being looked into by a number of clubs.

Younger Options

PSV Pair

After already suggesting one PSV player in Angelino, why not make it three with Hirving Lozano and Steven Bergwijn?

I should clarify, I’m not suggesting they sign both, but they should both be on their radar. They’ve both had great seasons and are both still young.

Starting with Lozano, I was initially going to leave him out, assuming Arsenal would be unlikely to pay his fee and bigger clubs would be in for him. There is a story that he’s going to Napoli for €40m, but I haven’t seen any official confirmation of that, so I’m not sure if it’s true or not. For that kind of fee, Arsenal definitely seem like they could be in for him.

Using the same sample as for Ziyech and Sarabia, but this time ruling out players older than 24-years-old, Lozano had the 8th highest NPxG + xA p90. He ran slightly hot on his xG, scoring 21 from 16.47, but heavily underperformed his xA, assisting 6 from 12.65 xA.

His other numbers, like dribbles and touches in the box, are also impressive, although don’t rank quite as highly as his output.

Bergwijn’s numbers aren’t quite as big as Lozano’s, contributing 0.50 NPxG + xA p90, compared to Lozano’s 0.66 p90, but are still hugely impressive, especially given he’s still only 21-years-old.

They both put up similar numbers for dribbles and touches in the box, while Bergwijn has the better deep completed passes, but Lozano the better deep completed crosses.

Either of these would be a good signing for Arsenal, with both of them offering them something they don’t currently possess. If they are available for under ~£35m, it definitely seems as though Arsenal should be interested.

Ligue 1 Pair

The obvious Ligue 1 name to mention is Nicolas Pepe. Pepe has had a great season for Lille and looks set to make a big move this summer. Across all competitions this season, he’s scored 24 goals and assisted 12, while still having strong underlying numbers.

After selling players like Alexis Sanchez, Theo Walcott, Alex Oxlade-Chamberlain and Serge Gnabry over recent seasons, Arsenal are now without these kinds of players, making Pepe seem like almost the perfect option for Arsenal’s right hand side. He had a strong return, with a NPxG + xA of 0.53 p90, was above average for both dribbles and touches in the box and is still only 23-years-old.

The catch is the price. The figure that I’ve seen when having a quick search is ~£70m and it seems hugely unlikely that Arsenal are going to go and spend that on one player. If Arsenal could negotiate down to sub £50m it seems like it would be worth it, but it’s hard to see Arsenal spending more than £50m on him – particularly if someone like Lozano is available for around £35m.

The other name from Ligue 1 is Marcus Thuram, who is another player that was mentioned by @MoeSquare, who said he could fill the role left by Welbeck.

Playing for bottom placed Guingamp, it’s not surprising that Thuram’s numbers aren’t as big as the likes of Pepe, Berwijn or Lozano, but they’re still strong in the circumstances.

Across all competitions, playing both on the left and centrally, Thuram managed a NPxG + xA p90 of 0.38 p90. Comparing him with Pepe, he made less dribbles per game (~9 vs ~7), but did have marginally more touches inside the box per game (3.81 vs 3.08), though this is likely due to playing up-front.

Going back to Zaha, Thuram had a similar NP xG + xA p90, but is still only 21-years-old and was playing for a side that finished bottom. What makes the deal even better is, not only are Guingamp likely to sell due to the relegation, but Thuram is also entering his final year of contract. Making it seem feasible that Arsenal could get him for ~£10m. At this price he’d be a great rotation option for them, while still having a lot of room to grow and improve.

I think Thuram is a player Arsenal should definitely look to be signing this summer, in an effort to bolster their attack.

Other Options

Looking in some smaller leagues, two names that could be worth looking into are Dani Olmo from Dinamo Zagreb and Leandro Trossard from KRC Genk.

Starting with Trossard, Arsenal have actually been linked with him this season and it’s not hard to see why. His NPxG + xA of 0.66 p90 was the 7th highest of wingers under the age of 24-years-old in the sample, with 0.41 of this coming from his NPxG p90.

From the few clips I’ve seen of him, he seems to operate from the left but likes to cut inside and look to receive the ball in more central areas. He doesn’t seem too different to how Iwobi plays from the left, except for the fact that he carries a lot more of a goal threat.

Outside of xG, their (Trossard and Iwobi’s) numbers across all competitions are fairly similar. They both attempt ~6 dribbles p90, with Trossard having a better success rate, Trossard manages slightly more touches in the box (4.39 vs 3.9), their xA is ~0.25 p90, only 0.01 separates their number of passes into the box p90 and they both have ~3 deep completed passes p90.

It’s worth pointing out that Trossard is doing this in a weaker league and for a team that won the title, but it’s still impressive stuff from him. A few articles say that he’s open to leaving Genk and the fee would be around £20m, which seems a fair price considering his age and output. I’m not sure if he seems like the answer for Arsenal’s attacking midfield problems, but he does seem like he’d be a useful addition.

21-year-old Dani Olmo has been at Zagreb for the past five seasons, but is apparently open to a move away. He’s spent more time centrally this season, but is capable of playing on either wing too.

His numbers aren’t as impressive as Trossard’s and he has only played 47% of league minutes this season, but he is a talented player. Across all competitions this season he has an NPxG of 0.24 p90 and an xA of 0.18 p90.

I mentioned him in an older piece I did, where I included the below video. While his numbers, particularly for a dominant side, aren’t crazy, they are good and he looks like a player who has a lot of talent.

Dinamo’s record sale was Marko Pjaca to Juventus for ~£20m, while in recent seasons the likes of Filip Benkovic and Marko Rog have let for ~£13m and ~£12m respectively. If Olmo cost ~£15m it does seem like it’d be a good deal, he has a lot of minutes under his belt, can play anywhere across the front and is still only 21-years-old.


Arsenal are in an interesting position. They still have the chance to get back into the Champions League, but a lot of work feels like it needs to be done – both on the pitch and in the window – before we can say they’re back on the right track.

I was also going to look at options for midfield, should they receive an offer for Xhaka, or even as an alternative for what they’re missing now Aaron Ramsey has left the club, but I thought I’ve already rambled on more than enough.

I really hate making decisive statements and much prefer just throwing out a load of names, partially because listing who you’d sell and buy quickly starts feeling unrealistic and like you’re playing Football Manager. Then, like on Football Manager, I very quickly go from thinking ‘those three players would be good signings’ to ‘why not just overhaul the entire side?’. However, to try and summarize the above incoherent mess into something, I’d break it down into whether or not they keep Ozil and what formation, then this would be my thought process.

If They Keep Ozil and play 4-2-3-1


  • Calum Chambers – ~£10m
  • Mohamed Elneny – ~£10m
  • Shkodran Mustafi – ~£20m
  • Sead Kolasinac – ~£15m
  • Laurent Koscielny – ~£5m
  • Henrikh Mkhitaryan – ~£15m

That raises ~£75m, plus an alleged budget of £75m (should they make the Champions League, which should also be given it’s own breakdown, but again, I’ve rambled enough), giving ~£150m. Plus the wages freed up by Ramsey and Welbeck leaving.


  • Angelino – ~£20m
  • Starting centre-back (Tah/Konate) – ~£35m
  • Younger centre-back (Pongracic/Saliba) – ~£20m
  • Hakim Ziyech – ~£25m or Pablo Sarabia – ~£15m
  • Marcus Thuram – ~£10m
  • Sign other attacker (Olmo/Trossard) – ~£25m

Obviously, while I’m using TransferMarkt as a guide, I’m mostly pulling these fees out of thin air, so don’t much weight into it. However, this kind of window does seam feasible. This would give them a squad situation like so:

Again, different players can play in different positions, but this would be the general gist.

The area of concern here would be the back-up ‘keeper (which applies to every section from here on) and the centre-midfield, which looks light after losing Ramsey and Elneny and not signing anyone new.

There are some interesting names out there, ex-Arsenal player Ismael Bennacer seems to have had a solid season for Empoli, Huddersfield’s Philip Billing had decent numbers, is only 22-years-old and is going into his last year of his contract, so could be worth looking into. They’ve also been linked with Christoper Nkunku for a while.

With Torreira and Guendouzi, they are also in a position to just sign a stop-gap option as cover for the time being.

Overall, I quite like the look of this. The left-side of defence is strengthened, they add immediate improvement in attack, while adding some younger options and keeping hold of Ozil, Aubameyang and Lacazette.

If They Keep Ozil and Play 3-4-1-2


  • Calum Chambers – ~£10m
  • Mohamed Elneny – ~£10m
  • Shkodran Mustafi – ~£20m
  • Sead Kolasinac – ~£15m
  • Laurent Koscielny – ~£5m
  • Henrikh Mkhitaryan – ~£15m


  • Angelino – ~£20m
  • Two starting centre-bacs (Tah and Konate) – ~£35m each
  • Younger centre-back (Pongracic/Saliba) – ~£20m
  • Marcus Thuram – ~£10m

This would give a squad situation like this:

This wouldn’t see any improvement for them in attack, with them just swapping Welbeck and 21-year-old Thuram, but is doubling down on the defensive improvements.

This isn’t a bad thing, defence is where they need the most strengthening and this means they can try and get the best out of Ozil, Aubameyang and Lacazette.

If They Sell Ozil and Play 4-2-3-1 v1


  • Calum Chambers – ~£10m
  • Mohamed Elneny – ~£10m
  • Shkodran Mustafi – ~£20m
  • Sead Kolasinac – ~£15m
  • Laurent Koscielny – ~£5m
  • Henrikh Mkhitaryan – ~£15m
  • Mesut Ozil – ~£30m

The Ozil money would raise the budget to ~£180m, while also freeing up some of the wage bill.


  • Angelino – ~£20m
  • Starting centre-back (Tah/Konate) – ~£35m
  • Younger centre-back (Pongracic/Saliba) – ~£20m
  • Hakim Ziyech – ~£25m
  • Pablo Sarabia – ~£15m
  • Starting winger – (Hirving Lozano/Steven Bergwijn) – ~£35m
  • Sign other attacker (Olmo/Trossard) – ~£25m

To give the screenshot like the others:

I’m a huge Ozil fan, but what I like about this scenario is that it feels like there’s much more variety in how Arsenal can line-up and play. Ziyech, Sarabia and Olmo can all line-up centrally or out wide, while Aubameyang and (Lozano / Bergwijn) offer a different threat out wide than the likes of Iwobi, Ziyech or Sarabia, meaning Arsenal can use different combinations depending on the opponent.

One of Torreira or Xhaka could even play as a #6 in front of the back four, then you can have two attacking #8’s. Ziyech, Sarabia, Olmo or Guendouzi could all play as #8’s, while still leaving players like (Lozano / Bergwijn), Iwobi, (Olmo / Ziyech / Sarabia depending on who plays as #8) and again Aubameyang for the wide positions.

I think this would only really be employed in home games when Arsenal expect the opposition to sit back, but it could be a really fluid system.

For instance, a starting line-up like this, where Sarabia and Ziyech could be switched.

(I picked Tah over Konate for this instance just because I was sick of seeing the brackets and slashes)

Then the likes of (Lozano / Bergwijn) (I’d say Bergwijn seems like the more realistic option), Iwobi and Olmo would all be available off the bench should something need to be changed.

If They Sell Ozil and Play 4-2-3-1 v2


  • Calum Chambers – ~£10m
  • Mohamed Elneny – ~£10m
  • Shkodran Mustafi – ~£20m
  • Sead Kolasinac – ~£15m
  • Laurent Koscielny – ~£5m
  • Henrikh Mkhitaryan – ~£15m
  • Mesut Ozil – ~£30m


  • Angelino – ~£20m
  • Starting centre-back (Tah/Konate) – ~£35m
  • Younger centre-back (Pongracic/Saliba) – ~£20m
  • Hakim Ziyech – ~£25m
  • Marcus Thuram – ~£10m
  • Nicolas Pepe – ~£70m

This feels more like the quality over quantity option, going out and getting Pepe rather than signing a couple of players from leagues outside the top five (not to say Lozano/Bergwijn wouldn’t also be good signings).

This would give something like the following:

This definitely looks light in midfield, with Xhaka and Ziyech not having immediate understudies, but does contain a lot of quality. You could argue Aubameyang could be played out wide, with Iwobi being viewed as the back-up #10, while Aubameyang and Pepe on either flank gives Arsenal a huge goal scoring threat from out wide.

If They Sell Ozil and Play 3-4-1-2


  • Calum Chambers – ~£10m
  • Mohamed Elneny – ~£10m
  • Shkodran Mustafi – ~£20m
  • Sead Kolasinac – ~£15m
  • Laurent Koscielny – ~£5m
  • Henrikh Mkhitaryan – ~£15m
  • Mesut Ozil – ~£30m


  • Angelino – ~£20m
  • Two starting centre-backs (Tah/Konate) – ~£35m each
  • Younger centre-back (Pongracic/Saliba) – ~£20m
  • Hakim Ziyech – ~£25m
  • Pablo Sarabia – ~£15m
  • Marcus Thuram – ~£10m

Like if they kept Ozil, Arsenal wouldn’t have the attacking additions here, but would double down in defence. I’d say, with the extra centre-back, Arsenal also have more license to be less cautious and an again line up with a #6 and two #8’s, with the midfield trio decided depending on the opposition. This would give the following:

Concluding the Conclusion

Arsenal have a lot of choices moving forward, partially because it doesn’t feel as though they have a strong tie to any particular shape or style. There seems to be a lot of value out there and a big opportunity to taking steps closer to the top four, while also reducing the average age of their squad.

They have some big decisions to make, mostly when it comes to Ozil’s future, but I don’t think there’s a huge difference in quality between any of the above scenarios. It feels more that Arsenal need to pick one and look to get the best out of it.

Leave a Reply

Your email address will not be published. Required fields are marked *