Houssem Aouar, Kai Havertz, and Who Could Fill the Hamšík-Shaped Hole in Chelsea’s Midfield – Improving Chelsea Part Four

Defending the premise of this series at the start of each post does become quite boring and tiresome, but the deterioration of Chelsea each subsequent week continues to force my hand. That’s how it feels at least. So, if this is your first post of the series and I come across as a mad, rambling, Sarri-apologist, I direct you towards Part Three and Part One of this series where I set out the premise and motivations of this project pretty clearly.

This post was quite the undertaking, so if you do enjoy it I would massively appreciate it if you could share this with anyone else you think might enjoy it. This is my twitter if you want to follow me for similar content.

What’s on the menu for today?

As we established in the previous part of this series, the attributes that Marek Hamšík brought (in spades) to Napoli’s midfield are entirely absent from Chelsea’s current trio of choice. After focusing on those at Chelsea who are yet to be given their opportunity, and potential signings who fulfilled some of our requirements but ultimately weren’t quite the right fit, we can finally have a thorough look at the heavy hitters who could step into Chelsea’s midfield and provide the attributes that are so sorely missed.

How are we going to do that?

The numbers can give us a decent indication of what each player offers in some areas, such as shot volume and quality, defensive output, and ball retention, but they (the stats I have at least) aren’t perfect for evaluating other skills that we are looking for. I used a combination of pass maps and video to evaluate how much each player progressed the ball, and how each player progressed the ball. Video obviously provides vital information about the types of possessions/phases-of-play the player thrives in, how quickly they make decisions, and the techniques they employ to act on those decisions. Whilst the overriding focus of the player clips I compiled was to show how each player progresses the ball, I also make note of other important attributes that are only apparent through video.

To lift the unbearable suspense (because you definitely didn’t skip the last post for its lack of juicy names), these are my favourite picks to fill the Marek Hamšík-shaped hole in our midfield:

(In no particular order)

Julian Draxler

The German midfielder has been fairly widely-known since his breakthrough in 2012 as a teenager at Schalke, but his solid performances at PSG seem to have gone unnoticed in the shadows of his superstar teammates. This season he has played a significant role for Tuchel, at the heart of his ever-morphing midfield. Although it would appear that nominally he hasn’t played all that much as a central midfielder this season, Tuchel will alternate PSG’s shape several times throughout some matches, so the positions on the starting teamsheet are not all that significant.

In a recent match against Rennes, Tuchel employed a 442 in the first half, switched to a 343 at half-time, and then dropped Di Maria into an “8” position to make more of a 3142 formation once they took the lead in the second half. This is clearly illustrated by my blurry screenshots with squiggly lines drawn on them. The point is, in Tuchel’s positionally fluid system this season, Draxler has been picking up the ball in the positions that we are looking for and has been very productive.

Also, PSG’s TV camera angle is amazing for looking at the team-shape. Would appreciate more of this from other football clubs please.

What is impressive about Draxler is that he advances the ball and contributes to shots really well through both his dribbling and his passing, rather than relying heavily on one skill or the other. That makes it much more difficult for the opposition to nullify his influence, as they can’t just cover his passing options, nor can they only aggressively pressure his first touch. They have to find a balance between the two and hope for the best.

Crucial to the effectiveness of Draxler’s passing is how quickly he makes decisions after receiving the ball, a consequence of scanning well, and frequently. He doesn’t dwell on the ball to assess each option, wait too long, and just resort to playing sideways. If the pass is on, Draxler plays it. This is even more important playing for PSG, with how lightning-quick some of his teammates are. If the timing or weight of his pass is even slightly off, the chance is missed.

What really makes Draxler a nightmare for defenders is, when there are no available teammates ahead of him, the German won’t let you off the hook every time with a square pass, he has the confidence to try to beat you, and is successful in doing so a lot more often than he isn’t. He is very unpredictable with how two-footed he is, the speed at which he makes decisions, and how quick his feet are in tight areas. Whilst he can’t just push the ball five meters in front of himself and chase it like a lot of high-volume dribblers can, he has enough acceleration to get in front of defenders once he’s wrongfooted them, or played a one-two around them.

I have no idea how available he would be, or how much he is enjoying his time at PSG, but I would definitely explore Draxler as an option. The quality of PSG relative to their league competition may worry some, but Draxler’s performances at his previous clubs, and in the Champions League with PSG would quell any concerns I might have. He’s been impressive in a variety of playing styles, in both France and Germany, demonstrated by his contribution of at least 0.18 NPxG90 and 0.15 xA90 in every season since 2014/15. I would not be very worried at all about the transferability of Draxler’s skills to the Premier League and Chelsea.

Piotr Zieliński

Signing Hamšík’s literal understudy for the past three seasons does feel like a cheat of an answer to this question… ehh, who cares!

Zieliński was impressive at Napoli in the 16/17 and 17/18 seasons whenever he got on the pitch, but that was the problem, he could never displace Hamšík, Jorginho or Allan. I mean, to have a player as good as Zieliński just not getting minutes, the manager would have to be stricter about rotating his players than Chelsea are under Maurizio Sar… oh, that’s right. Luckily for us, Zieliński has played the second-most minutes in the league of any Napoli player this season, only behind Koulibaly, so we’ve got the all clear to see how he’s been doing.

The Polish midfielder makes the list because this season he has been really good at pretty much everything we’ve asked of our midfielders. That said, I don’t know if he’s exceptional at any one part of the game in the same way that some of the other players on this list are. There are still elements to Zieliński’s game that I do really like though, and I think they would translate well elsewhere.

One of the areas Zieliński does standout in statistically is pure shot-volume, taking ~1 extra shot per 90 minutes over most of the midfielders on this list, but what is notable is how he makes the best of those shooting opportunities. His timing in approaching the ball is impeccable, always arriving at the ball such that he can strike it cleanly and accurately, helped by his ability to connect with the ball well on either foot. This adds value to these chances that might otherwise be spurned by other midfielders, the majority of whom have worse timing and technique than Zieliński. I also like that this doesn’t seem to be overly system dependent, Zieliński pops up on the edge of/inside the box at the right moment in longer possessions as well as transitions.

The Napoli midfielder is a really good ball-progressor, specifically with his ability to carry the ball. He can cover a lot of ground very quickly and uses his pace and quick change of direction effectively, both on and off the ball, to create separation from defenders. As you can see from several of these clips, he even manages to do this from a standing start in non-transition possessions, which is pretty rare.

Although he does have room to improve his progressive and creative passing, specifically in his combination play around the box, Zielinski still has an impressive range and variety of passes in his locker. This, much like his shooting, is also helped by his two-footedness.

There may have been some positional worries with Zieliński, as his creative output (0.16 xA90 for the season) has mostly come from the left of a midfield four (0.29 xA90 from left-midfield and just 0.09 xA90 from central-midfield). But, having now watched a fair amount of Napoli, I have learned that the two central midfielders in the four play as “6s” whilst the wide midfielders really don’t end up in positions that dissimilar to the “8s” at Chelsea under Sarri. If anything, his NPxG90 rate of ~0.20 (which has been consistent regardless of position) is even more impressive given he has spent so many minutes deeper on the pitch. In his previous two seasons at Napoli, he was consistently productive as an “8” in a 433.

The fact that Liverpool (the team who haven’t made a single bad signing since Christian Benteke for £42M in 2015) are reportedly interested in Zieliński probably means absolutely nothing, but I will take it as a sign of encouragement. It also suggests that he ranks really well in some incredibly advanced metric that none of us mere humans could even comprehend, conjured up by the boffins behind player recruitment up in Merseyside.

I really like Zieliński, but he probably wouldn’t be my first choice on this list, not that it would likely matter much anyway given Chelsea’s supposed agreement with Napoli not to sign any more of their players after the dealings over the past summer.

Franck Kessié

Franck Kessié has been one of the best ball-progressing midfielders in Italy since his time at Atalanta, but this is the first season where he has shown enough shot-contribution to squeeze onto this list. To be perfectly honest, Kessié’s 0.3 NPxG+xA90 this season, up from ~0.2 in the previous two seasons, still might not be up to scratch for what we need in this position, but he is such a good player that I had to include him. Under a different manager a double-pivot of Kessié and Kanté would be wonderful, and having a midfielder who advances the ball as much as he does is so valuable because of the burden it lifts off of those around him.

One part of his game that really aids his ability to pass and carry the ball forward, is Kessié’s resistance to pressure. He is the first midfielder that we have covered that I would consider particularly press-resistant. It isn’t a crucial attribute for the type of player we are looking for, but it certainly seems that it is becoming increasingly useful each season, as teams who press out-of-possession are proliferating throughout Europe.

Complemented by his ability to resist pressure and carry the ball is Kessié’s range of passing, which is considerably better than that of many midfielders who progress the ball through carries and dribbles as much as Kessié does. Under Gattuso, AC Milan set up such that the wide-forwards (usually Çalhanoğlu and Suso) often play very narrow and deep in possession, not leaving many players tasked with stretching the opposition in behind. Kessié’s ability to find what might be the sole forward, or even fullback, making that darting run is all the more impressive.

My suggestion of a Kessié – Kanté double-pivot mostly came out of my thinking that Kessié would provide almost all of the ball-progression, with Kanté covering his defensive shortcomings. Whilst I stand by that, Kessié does such great work in the box, that you would be tempted to play him further forward as well. I really enjoy his timing arriving into spaces, his awareness of what is around him, and the disguise he puts on the passes he plays and touches he makes.

I still think Kessié would probably be a less-suited solution to the extremely specialized midfield-role we are looking for, but I also think he is one of the more talented, and tactically flexible midfielders on this list and would thrive under most other managers.

Kai Havertz

There really aren’t many players who make an impact in a top-5 league at 17 years of age, but when they do, they tend to go on to be pretty successful. The list of attackers obviously includes the likes of Wayne Rooney, Kylian Mbappe, Raúl, Michael Owen and maybe a handful more. Kai Havertz contributed 0.2 NPxG90 and 0.2 xA90 (amongst positive output in other areas) over the course of ~1400 minutes in the Bundesliga at 17 years old. He will probably remain in the shadows of other teenage superstars because he doesn’t quite get the goals and assists that create the headlines, not least due to the role he plays, but he gets a hell of a lot closer to that list of 17-year-old superstars than most do.

He also increases the number of reported Liverpool targets on this list to two, which is always nice.

We are quite lucky that Kai Havertz made the box-shot and shot-assist cutoff I set because he is an example of an excellent creative and progressive passer who doesn’t excel in any of the passing stats on the midfielder radar. This is expected given the stats I have access to, included on the player radar, tell you such a limited amount of information about the passing ability of a player. Hence the necessity for me to watch so much video of each player, to try to determine their progressive abilities. But, if you’ve made it this far, you must not mind my eyes in place of those stats, so on we go! 

In longer possessions, Kai Havertz does two things exceptionally well, which allow him to be such a creative and progressive player in the final third. Firstly, he always arrives into the FB-CB vertical corridor at the right moment, rather than waiting in that space and allowing himself to be picked up by a marker. Secondly, his first-touch is perfect, just about every single time. He always receives on the correct foot to present himself with the most options, and takes the ball precisely in the direction of his next action, whether it be a pass, shot, or an opportunity to face up a defender. Havertz also has a fantastic weight of pass, unpredictability in taking on defenders, and a wonderful strike of the ball, but his ability to take each “step” leading up to those actions perfectly, as frequently as he does, is what sets him apart from his peers.

Havertz’s physicality was apparent in his ability to accelerate away from and/or hold of his opposition in some of those clips, but in transition, it really comes into its own. There just aren’t many central midfielders at 1.88 meters tall, as quick as he is, with such good balance and coordination. Combine that athleticism, with his technique, with his knack to make the correct decision on Every. Single. Offensive Transition… and you’re left with a deadly counter-attacking weapon to complement the intricate player he is in longer possessions. His height also creates mismatches aerially from set-pieces and in open play.

I understand that Bayer Leverkusen create a significant number of shots from counter-attacks and that some people may be sceptical about Havertz’s ability to perform against teams who spend more time defending in a low-block. But, Havertz is so exceptional technically and physically, complemented by the best decision-making you will ever see from a 19-year-old, that I would be very confident in Havertz’s skills transferring to the style of play of a more protagonistic team. Whichever team Havertz leaves Leverkusen for will have a midfielder that they won’t need to replace for the next decade, at least.

Houssem Aouar

The makeup of this current Olympique Lyon squad is very reminiscent of the Monaco team of the 16/17 season, in that it is brimming with young talent that looks set to be poached by some of the biggest clubs in Europe. Ferland Mendy, Tanguy Ndombele, Nabil Fekir, (even Memphis Depay, Moussa Dembele and Bertrand Traore to a certain extent) will soon be ready to move on to bigger and better things if they aren’t already. Any clubs that had a close look at one of those players last summer will now likely be kicking themselves for not acting on their interest, as Lyon’s impressive performances against Hoffenheim and Man City in the Champions League group-stage seem to have inflated the individual players’ reputations, and potential transfer fees. Houssem Aouar is the youngest of the current crop, but is clearly one of the best prepared to take on the challenge of playing for one of Europe’s elite.

A graduate of one of the most prolific academies in Europe – that has produced the likes of Karim Benzema, Alexandre Lacazette, Corentin Tolisso, Samuel Umtiti, Anthony Martial and Nabil Fekir – Houssem Aouar has witnessed first-hand the clear pathway to the first-team and the pastures beyond should you take your opportunity. If I had more time I would go back to compile the quotes, but on a recent On The Continent podcast, Andy Brassel spoke of the academy/first-team dynamic at OL. He described how, as much as their footballing-ability ultimately determines every young player’s future at Lyon, the academy engenders an air of confidence, even arrogance at times, in its graduates and having that type of personality is crucial in their progress towards the first-team. Once they make it to the senior group and are surrounded by fellow academy graduates, they really “run the team” and actually hold the authority over their counterparts signed from elsewhere, regardless of how great the discrepancy in age and experience may be.

Regardless of whether he matches up to those stereotypes off the pitch, it is glaringly clear that Aouar carries a certain personality onto it.

Technically, Aouar is quite unique compared to any of the other midfielders we’ve covered so far in his ability to dominate one-on-one situations, with the presence that I’ve been describing. The way dribbles are recorded statistically doesn’t entirely capture this, as Aouar’s 1.8 Dribbles90 is excellent but not streets ahead of what Havertz and Draxler manage. What I am describing that isn’t appreciated through my very binary dribble stats is his ability to beat his man so comfortably from both a standing start and mid-sprint in transition. There is a fluidity and a grace about Aouar’s motion, as he keeps the ball the same distance from his foot at all times, ready to push it in any direction as he reacts to the opposing defender’s movements.

Aouar’s confidence also manifests itself in his willingness to take the ball under pressure, and his ability to still find a dangerous pass or shot from even the tightest of areas. He keeps the ball too close to himself, gets too much of his body between the ball and his marker, and makes decisions too quickly for the opposition to dispossess him with the same ease they would with most players.

Aouar does rely on transitions to progress the ball, but it should be noted that he creates plenty of those transitions himself through his defensive work, and he makes use of those transitions very effectively. Aouar runs more quickly with the ball than most of his opposition can without it, and he picks the right pass more often than not. Essentially, he’s just very difficult to contain once he gets going.

For all of his positive dribbling and passing, in transition and in longer possessions, Aouar also contributes plenty of shot quality and volume indicated by his 0.37 NPxG+xA90.

There aren’t many players out there that demonstrate the qualities necessary to improve the midfield of Champions League contender before their twenty-first birthday, but I’m confident Aouar does. Along with Kai Havertz, the Lyon no. 8 is the best midfielder (in this style) in his age group (+/- 1 year). The opportunity to sign either of them is the opportunity to acquire a midfielder with a decade of world-class performances ahead of them.

Kerem Demirbay

Finally, all the hard work feels as if it’s paid off as we find a midfielder who is neither on the tip of every Chelsea fan’s tongue nor in the title of every “Best FM Wonderkids” article. Having only made two appearances for the German national team, and just five appearances in the Champions League, anyone who isn’t an avid Bundesliga fan probably hasn’t seen Kerem Demirbay play. Which is exciting, you’re about to have a new favourite player!

Demirbay’s slow rise to prominence hasn’t been helped by an extensive injury history, that would certainly be a concern to any potential buyers. Recently, German midfielder has suffered a myriad of soft-tissue injuries, the most significant of which seem to be ankle-ligament related, which have seen him miss 20 matches over the past two seasons. It is a shame because when he does get on the pitch, he’s exceptional.

The phrase “a wand of a left-foot” seems to get bandied about more every season, describing players who don’t actually have anything that special. Kerem Demirbay does have “a wand of a left-foot”. Even if Demirbay had significant tactical or physical shortcomings it just would not matter because wherever he is on the pitch, wherever the opposition players are, and wherever you (his teammates) are, he will find you with his left-foot. It doesn’t even seem to matter if it’s at the end of a counter-attack or part of a twenty-pass move. All of these clips are from open-play, but I should also mention that he has a wicked set-piece delivery, contributing a massive 0.18 xA90 from set-pieces to complement his 0.21 xA90 from open-play.

Even after I built him up so much, I bet you didn’t see that coming.

Demirbay’s passing ability is impressive enough on it’s own to warrant a move to a bigger club, but the German midfielder also contributes 0.21 NPxG, 2.5 dribbles and 3.3 defensive contributions per 90 minutes. His poor ball-retention and diminutive build can cause problems at times, but I still think Demirbay is a massively net-positive asset whenever he plays. His ability to wriggle away from defenders before finding a shot on goal, or one of his signature dinked passes is particularly impressive.

The more you see of Demirbay, the more difficult it is to understand how it took him so long to make his impact in the Bundesliga, and how nobody has made a more concerted effort to take him off of Hoffenheim’s hands. Demirbay would be an excellent option for any team that couldn’t afford/attract Aouar or Havertz, and has young midfielders waiting in the wings to make a significant impact in two to three years (no idea who I’m talking about there…). By then, Demirbay’s powers will be on the verge of waning, but hopefully not soon enough to stop you recouping most of your investment in him. He might be the best peak-age player on this list and is probably the least expensive one as well.

That’s all the video I have time to put together, but there were two more midfielders who did make the cut. I imagine everyone reading this has seen Nabil Fekir play before but, if not, a quick YouTube search will return dozens of videos. Fabián Ruiz also made the list, but I am pretty sure I want to write something on the transition from his role at Betis to his current role for Ancelotti’s Napoli, at some point, so stay tuned for that. This has been quite different from my previous posts, so I hope you still enjoyed it, and if you didn’t I always appreciate constructive criticism. Thanks for reading :).


4 responses to “Houssem Aouar, Kai Havertz, and Who Could Fill the Hamšík-Shaped Hole in Chelsea’s Midfield – Improving Chelsea Part Four”

  1. Hey really love the work you have done for these articles and it was a pleasure reading them. Please keep them coming. Want to point out one thing though, just a line or two more than what you have written on the deficiencies of each player would be perfection.


    1. Thank you! Absolutely, I actually had similar thoughts myself. I will definitely keep this in mind for next time. Really appreciate the feedback.


  2. Brilliant article! I really appreciate the hard-work you put into this. I wish you became more popular!


    1. Thank you, appreciate the kind words!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: