Bruno Fernandes. Signed from Sporting Lisbon by Ole Gunnar Solskjaer, he was seen as a crown jewel in a very transitional Manchester United side. Four years later, he’s captain of the side and a mainstay in Erik ten Hag’s team. However, whilst he was once heralded as the team’s best player, some fans consider him to be a liability in ten Hag’s systems. On top of that, at 29 years of age, Bruno doesn’t have too many more years at the peak of his powers. Now is a good time for United to start looking at new, younger players to take over once Bruno’s time comes. The question is, who is United’s Next Fernandes?
In order to do this, we are going to use two methods to make a shortlist of potential targets. The first method will involve looking at Bruno’s strengths and weaknesses and finding players that match those using Z scores. The second method involves using a machine learning technique called Principal Component Analysis to find statistically similar players to Bruno.
Once that shortlist is complete, we’ll take a closer look at the potential options and come up with a definitive suggestion for who United should sign as an understudy – and eventual replacement – for Bruno Fernandes. So, let’s jump in and take a look at Bruno himself.
Bruno: High Risk, High Reward
Bruno Fernandes is a very stylised midfielder, excelling in creative output and quality. Looking at his percentile rankings this season against other midfielders in Europe’s top five leagues, we can see where Bruno has been elite this season. He ranks in the 98th percentile – that is, the top 2% of players – for key passes, open play shot creating actions, passes into the penalty area and expected goals assisted. He also ranks in the 96th percentile for progressive passes. What we can see here is that Fernandes is always willing and able to find his teammates in good shooting positions, sending them killer passes into the penalty area.
This high-octane creative style does have its drawbacks, of course. His overall pass completion and his specific range pass completion numbers rank him no higher than in the 38th percentile. This makes him somewhat of a liability when trying to control the tempo of matches. United fans have often seen this lead to opposition sides creating chances and goals against the team in the late stages of the game, when the team should have slowed the pace of the game down and kept a hold of the ball.
Bruno’s lack of control can also be seen in his shooting numbers. He ranks in the 91st percentile for shots taken per 90, but in the 9th percentile for non-penalty goals minus expected goals. Bruno takes a lot of shots but isn’t very good at finishing. This leads to a high number of wasted chances. In situations where Bruno tends to shoot, other, more possession-conservative players would find a pass or recycle.
That’s not to say that Bruno’s energy levels are a complete detriment to United’s ability to keep the ball. He’s decent in the air, ranking in the 77th percentile for aerial duel success, and ranks about average for interceptions, fouls committed and blocks. This makes him a capable, albeit not amazing, presser, something that ten Hag wants in his midfield.
Now that we know what Bruno excels and lacks in, we can start looking at players that fit the bill for Manchester United.
Method I: Z Scores
So, what will we want to find in a Bruno replacement? Optimally, we would want a player that shares the same strengths as Bruno. High creative output and an ability to get the ball into dangerous positions are things that his understudy will need.
However, we also want a player that won’t share the same weaknesses as Bruno. Whilst ten Hag has stated his desire to make United the best in class in transition, what United really need is a player that can dictate the tempo of the game and keep things calm and ticking over when they are protecting a lead or holding onto a draw. This means that our Bruno replacement will need to rank decently for pass completion – especially at short range. Someone who doesn’t shoot as much – or as erratically – would also be a plus.
Taking those parameters, I set about finding a shortlist of players that could be a suitable replacement for Fernandes. As we are looking for a player with longevity, I restricted my search to players aged 23 or under. On top of that, I also only looked at players in Europe’s top five leagues. This means that all the players I found have a sense of credibility to their numbers.
To make this shortlist, I went into R and loaded up player data from this season. I then converted those numbers into Z scores. A Z score is a measure of value-location, telling me how a player has performed in comparison to their peers. Unlike percentile rankings, that just tell me a player’s position however, a Z score tells me exactly how much better or worse they are compared to the average player. A player ranked average would have a Z score of 0. I filtered for players who had a Z score of more than 0.5 for key passes, open play shot creating actions, passes into the penalty area, expected goals assisted and progressive passes. This will give me players who rank at least decently in Bruno’s best metrics. I then filtered for players who also had a Z score of above 0 for short pass completion and take on success rate. This would give me players who are also adequate in ball retention.
Four players remained after my filtering process. Leverkusen’s Florian Wirtz, Liverpool’s Harvey Elliot, Jude Bellingham, and ex-United midfielder Angel Gomes. So, for now, we’ll add them to our replacement shortlist. We only found four players using Z scores, but using Principal Component Analysis, we should be able to find a lot more.
Method II: Principal Component Analysis
Principal Component Analysis is a tool often used to find clusters of data observations with similar traits. It uses correlation and matrices to group those observations together. That way, we can easily see which points are like each other. Principal Component Analysis, or PCA, is also good for dimension reduction. It merges groups of variables into PCs (principal components) that have similar effects on an observation’s placement within clusters.
In this context, we can use PCA to take more than one hundred metrics on player style and turn them into fewer than ten new metrics that can tell us exactly what a player plays like. Then, using a mathematical technique called Euclidean distancing and some indexing, we can create a number that accurately tells us just how similar a player is from another. That is our Similarity Score.
So, we can take the numbers of all attacking midfielders in Europe’s top five leagues and do some Principal Component Analysis to see which players play most like Fernandes this season. Note that this score only considers style-based metrics, like passes per 100 touches and not performance-based metrics like pass completion. Having said that though, players that play like each other tend to be good at similar things.
We are now able to create a table of 12 players with the greatest similarity score when comparing them to Bruno Fernandes. Given Fernandes’ unique role in United’s system, the highest similarity score for an under 23 player is only 68 out of 100. But generally, any player with a score of 50 or higher would be able to somewhat adapt to the compared player’s position in their team.
Again, we see Angel Gomes and Jude Bellingham, but we also have the likes of Alex Baena from Villarreal, Chelsea’s Enzo Fernandes, and PSG’s Xavi Simons, who is currently on loan at RB Leipzig. These are all players who have similar in-game tendencies to Bruno and could stylistically replicate him at United.
After adding those players to the list, we now have a shortlist of fourteen players to go through. The players in blue have come from our PCA Similarity Scores and the yellow players have come through our Z score filtering method. Gomes and Bellingham showed up in both methods, so appear in green. But before we go in depth looking at those players, we have some cuts to make.
Trimming the Fat (Fees)
In an ideal world where rivalries didn’t exist and Manchester United were still the most attractive team in the world to play for, all fourteen players on our shortlist would be reasonable targets for ten Hag’s team. However, we don’t live in an ideal world and United are likely playing Europa League football next season, so we have to make some cuts.
Dominik Szoboszlai and Harvey Elliot play for Liverpool, which make them immediate no-goes. Chelsea are unlikely to let Enzo Fernandes leave and Pape Matar Sarr has seen an increase in playing time under Ange Postecoglou, which makes him unlikely to make a move. Jude Bellingham has only just moved to Real Madrid, and even if he had moved to Real Madrid five years ago, I doubt he’d want to leave his sunny Madrid for turbulent United any time soon.
This leaves us with a new list of nine players who we can go through now. Looking at their numbers this season and the shape of their performance charts, we can see who might make a capable understudy for Fernandes.
The Candidates
Villarreal’s Alex Baena ticks all our boxes when looking for a player with high attacking output. The Spaniard ranks in the nineth percentile or higher for key passes, open play shot creating actions, passes into the penalty are and expected goals assisted when looking at European attacking midfielders. Unlike Fernandes, Baena is a ball carrier as well as a passer, ranking in the 85th percentile for progressive carries. However, he does share Fernandes’ flaws, ranking in the 93rd percentile for shots per 90 minutes, but in the 3rd percentile for non-penalty goals minus expected goals. Off the ball, Baena doesn’t show much quality either: the only metric in which he ranks above average is fouls committed per 90. For a pressing team like ten Hag’s United, Baena may not be the perfect fit.
Clermont Foot’s Muhammed Cham is similar to Baena in that he excels in creativity. Whilst his rankings are generally lower than the Spaniard’s, Cham’s possession numbers are much more promising. A 72nd percentile ranking for short pass completion is nice to see when added to his creative numbers. He can be relied upon to make those simple passes as well as killer balls. However, his defensive numbers are worrying, without a single ranking above the 20th percentile, Cham would need a lot of coaching to fit into United’s out of possession system – although the same could be said for United’s current forwards.
Enzo Millot’s player pizza looks a lot more complete than the previous players we’ve seen so far. Creatively, Millot ranks highly in all areas, including carries and passes into the penalty area, shot creating actions and expected assists. A high ranking for xG also suggests that his movement has been good for Stuttgart this season. In possession, Millot is very comfortable making short passes as well as progressing the ball forwards. His take-on success, whilst below average, isn’t a massive cause for concern either, making him a somewhat capable holder of possession. Enzo’s defensive numbers are nothing special, ranking in below-average positions for most metrics. His low aerial duel win rate may also be an issue, unless United deploy a taller midfielder like Scott McTominay to compensate. Unlike the previous candidates and Bruno Fernandes himself, Millot is pretty sensible with his shots. Taking an average amount of shots for his position but finishing chances at a better rate, United are likely to find use in his attacking play and at 21 years of age, he has a lot of room for growth.
As an ex-United player, Angel Gomes’ inclusion on this list is very interesting. His player pizza is also very interesting in its difference to all the other players we’ve looked at. Whilst Gomes’ creative numbers are very good, his possession numbers are his selling point. His 92nd percentile ranking for total pass completion is comprised of pass completion rankings above 85 for all lengths. A take-on success ranking of 80 sprinkled into the pot and we’ve got the taste of a press-resistant and comfortable midfielder, who can control the tempo of the game with an extensive passing range. He can bring the ball forward and into dangerous positions through his passing and carrying. Adding a player with that skillset and playing them alongside someone like Kobbie Mainoo would give United’s midfield an element of calm that they’ve lacked over the years. However, one problem that Gomes wouldn’t be able to fix is a lack of physicality. He ranks low for defensive numbers and is quite short. Off the ball, he is likely to be bypassed, increasing the burden on the rest of the midfield. Having said that, Gomes seems like he would be an undeniable asset to United’s midfield arsenal.
Athletic Bilbao’s Oihan Sancet falls into the same category as Alex Baena and Muhammed Cham: high volume creators. Sancet’s quirk is that he is also a very good ball carrier, ranking in the 94th percentile for progressive carries. He’s also very good in the air, which may be useful in set pieces – something that United haven’t impressed in this season. Sancet’s defensive numbers aren’t great, but at this point, it’s clear that the sort of player we’re looking for to replace Bruno Fernandes isn’t going to have the pressing intensity and off-the-ball awareness of the Portuguese. Apart from his poor finishing ability, Sancet appears to be a good option to step in for Bruno.
Ivan Ilic is probably the best defensive player on the list. Above average rankings for aerial duels and interceptions are signs of a player that can be coached to press effectively. His finishing is good, ranking in the 80th percentile for xG overperformance. High rankings for key passes and expected goals assisted, and good rankings for take-on-success and short pass completion make Ilic a valuable asset in multiple phases of the game. Whilst he was on the books at Manchester City for three years, he never made a senior appearance for the club, so a move to Old Trafford likely wouldn’t ruffle too many feathers.
According to transfermarkt.com, Lorient’s Julien Ponceau is worth 3 million euros. For a player with rankings above 40 for all key attacking midfielder related metrics, that is a bargain price. However, some context is needed. Lorient aren’t a strong Ligue 1 team. They are currently in a relegation battle and have the league’s worst defensive record. Whilst Ponceau ranks in the 63rd percentile for interceptions per 90 and in the 44th percentile for blocks, these are to be expected for someone on a team that is constantly playing on the back foot. Having said that, when they do press opposition back lines, they press very well, with Ponceau often benefitting from this. Lorient have also created the fewest shot creating actions and have one of the lowest xG totals in the league. That being said, Ponceau has generated 15% of the team’s SCA and contributed to about 20% of the team’s xG. An integral part of a weak team, it would be interesting to see how Ponceau adapts to playing with better players at a better club.
Xavi Simons is a very good player, who would command a very large fee of around 70 million euros. With a 98th percentile ranking for progressive carries and a 71st percentile ranking for progressive passes, he provides a slightly different creative threat to Bruno Fernandes. However, diversity is something that the very one-dimensional United need right now. On top of play style, Simons is elite in creative output, even surpassing Bruno in key metrics like penalty area touches and carries into the penalty area. His finishing is much better as well, with his 90th percentile ranking for xG overperformance placing him 81 percentiles higher than Fernandes, despite only ranking 2 percentiles higher for non-penalty xG. Xavi Simons just takes his chances much better. His take-on success is great, making him quite press-resistant when carrying from deeper positions, but his 55th percentile ranking for short pass completion isn’t particularly astonishing. Since Simons’ playstyle is suited for carrying, he may struggle against low blocks in the Premier League – something he rarely has to face whilst at RB Leipzig. Whether he can adapt and still maintain his high level of output is a risk that United would have to take if they considered signing the Dutchman.
Florian Wirtz is the final player on this shortlist – and I’ve saved the best until last. The golden pizza slices indicate that Wirtz ranks top for that metric. That is, he is the leading attacking midfielder in Europe for open play shot creating actions, passes into the penalty area and progressive passes. He also ranks in the 99th percentile – top 1% – for penalty area touches, and rankings in the 90th percentile or higher for metrics like progressive passes, key passes and expected goals assisted. In other words, Wirtz is a creative machine. His pass completion is at least average for all lengths, and he succeeds in his take-ons more than the average attacking midfielder. His ballstriking is just below par, but his 92nd percentile ranking for non-penalty xG suggests that he is adept at getting into good goalscoring positions often. Sure, he isn’t the best defensively, with only his blocks reaching the average European standard, but seeing as Bayer Leverkusen are still unbeaten in the Bundesliga this season and have conceded the lowest number of expected goals this season, this defensive passivity can be excused. Wirtz would be a fantastic addition to any side in Europe this summer – United included.
The Final Selection – Who should United sign?
Now that we’ve analysed all of United’s potential options for a successor to Bruno Fernandes, I can finally present my three top choices. I have selected one player for each price range, so that the club – if their scouts have found this – can adjust their options for the budget they have in the summer.
For a very high budget – something quite unlikely given financial fair play rules and United’s inability to cash in on players – my choice has to be Florian Wirtz. Whilst Xavi Simons has a strong level of creative output, Wirtz’s ability to progress the ball through both passes and carries makes him much more reliable against different types of defences. For 100 million euros, United would get a player that has dominated the Bundesliga, and at such a young age, still has the potential to grow to greater heights. There would undoubtedly be competition from the likes of Bayern Munich and other Premier League clubs that would drive the price beyond 100 million, but if the Glazers suddenly decided to push all of their funds into the transfer market to help United compete on the pitch, Florian Wirtz should be one of their targets.
With a more realistic budget that doesn’t allow for such massive transfer fees, I think that a 20-million-euro investment on the return of Angel Gomes to Old Trafford would be my preference. Gomes gives you want Bruno Fernandes gives you – a lot of progressive passes that lead to dangerous chances – but he also provides a calming presence in possession, something that Bruno struggles to do. Gomes may not want to rejoin United given the nature of his exit and in that case, Enzo Millot would be my backup target, but if he was up for a redemption story, I think that it has every chance of going more like Jonny Evans rather than Paul Pogba.
Finally, for a very low budget option, Julien Ponceau is definitely a must-buy. I’d argue that even if Angel Gomes or Florian Wirtz were purchased, or if the club had a large budget, Ponceau would be a smart singing. A very low risk move given the 3 million euro price tag, even if the Frenchman failed to make the jump from Lorient to United, there would be heaps of resale value. What United fans would hope for, however, is that this could be one of the best bargain signings of the Premier League era.
So after a long scouting process, several thousand numbers and a lot of pizza, we can finally say that United have the ability to replace Bruno Fernandes with an up and coming star. However, whether we do see a succession plan for Bruno or not is a different question.
Why None of This Really Matters
Over the past few seasons, Bruno Fernandes has rarely missed a game for United. In sickness and in health, through good form and through bad form, United’s captain has pretty much always been the first name on the team sheet. Whilst the saying “availability is the best ability” does hold true, Erik ten Hag’s inability to rotate Bruno out of the team could be detrimental in finding a player to take over once Fernandes moves beyond his peak.
A player like Florian Wirtz wouldn’t leave his starting role at Bayer Leverkusen to sit on the bench for Manchester United behind Bruno. Angel Gomes left United for this exact reason – he needs to play games. This would likely affect Ponceau to a lesser extent, but game time aids in player development and ten Hag is notorious for not giving much time to players that aren’t mainstays in the first team. Amad Diallo is a great example of this, and Isak Hansen-Aaroen, a player who could have been seen as Bruno’s successor, has just left in search of real playing opportunities. Erik ten Hag would cite injuries as a reason behind his lack of rotation, but the decline of Donny van de Beek would serve as a counterexample to this.
Replacing Bruno Fernandes theoretically, using data, is something that can be done quite easily – we’ve just done it in a few steps. Replacing United’s ever-present, vocal talisman and captain on the field in reality is definitely a much bigger task.
