With the transfer of Leroy Sané to Bayern Munich now confirmed, Barcelona seem to have missed out on yet another young player who could have greatly improved the team. When comparing the German’s stats to Barça’s left winger this season, Antoine Griezmann, it’s clear that Sané would have been a brilliant signing.
On Thursday, Bundesliga champions Bayern Munich made official the signing of Manchester City’s Leroy Sané for 49 million euros plus 11 million in variables. It is true that the 24-year-old German left winger had missed most of the season in England due to an anterior cruciate ligament tear. But, for his price and Barcelona’s problems out wide, the Catalans should have set their sights on Sané.
Antoine Griezmann was purchased by Barça last summer, but, with Luis Suárez and Lionel Messi often having a reserved spot in the starting line-up, the former Atlético de Madrid man has been forced to act as a left winger on multiple occasions. Nevertheless, that’s not Griezmann’s best position, but it is Sané’s.
While this article will be comparing Sané to Griezmann, this is not an attempt to ridicule the Frenchman. It has been clear to all Barça fans that Griezmann has been well out of his preferred position and role this campaign. However, he has played most of his minutes on the left of the front three, which was the preferred role for Leroy Sané at Manchester City.
Thus, comparing the numbers of the two players allows one to see the different strengths that make Sané more suited to that role. This doesn’t mean the Griezmann should be out of the team. If anything, Leroy’s signing would have allowed Antoine to move to a better position, while providing more balance in the side.
The heatmap of Leroy Sané in the 2018/19 Premier League season (left) vs Antoine Griezmann in the 2019/20 La Liga season (right). Both players operating primarily down the left | Heatmap by SofaScore
Due to Leroy Sané’s knee injury this past August, he has only played twelve minutes of Premier League football this season. Because of this, his stats from the 2018/19 Premier League campaign will be used, and compared to Griezmann’s from this season. So, what do the numbers say?
Goal scoring and finishing
The finishing aspect of the game is the one where most would expect Griezmann to outperform Sané. The Frenchman has a far greater reputation as a striker and goal scorer due to his time playing up top for Atlético de Madrid. But, the stats show that Sané has just as much, if not more goal scoring ability from the wing.
First things first, the charts in this article for Sané show his stats in comparison to all outfield players who played at least half the available minutes in that Premier League season. While Griezmann’s show his stats in comparison to all La Liga outfielders who have played at least half their team’s minutes so far this season. By splitting up their two respective leagues, this takes away much of the differences in stats produced by different styles of play between the two countries.
The charts also utilize the fifth percentile for each stat as the minimum, and the ninety-fifth percentile as the maximum. This is done to remove outliers that would skew the data. Think about how comparing some of Lionel Messi’s stats to other players, even if they are very good, will always make them look poor in comparison.
Stats are correct as of July 4, 2020 | Stats from Football Reference
At a glance, it’s clear that both players perform very well in this part of the game relative to their leagues. Going further into the numbers, though, can reveal some interesting facts.
In terms of getting into better scoring positions, Griezmann’s stats are the slightly better of the two. He beats Sané, barely, in non-penalty expected goals per 90 minutes. This means that from the shots taken by the two players, Antoine is expected to score slightly more goals. Griezmann also has a higher non-penalty expected goals per shot, meaning that the average shot he takes is a better chance, and is more likely to go in.
Unfortunately for the 29-year-old, he has not done as much with these chances as Sané did in 2018/19. Despite being expected to score less, Sané’s actual goals per 90 minutes and goals per shot are higher. In fact, Sané’s 0.48 non-penalty goals per 90 were the eighth best in the Premier League that season. Griezmann, despite having more history as a striker, is currently seventeenth in La Liga.
Sané also has the advantage in terms of his shots on target. With 1.16 per 90, he has a significantly higher volume of shooting, but his rate of getting them on target is still just 0.2% below Griezmann’s.
With these stats in mind, it’s clear that Sané takes better advantage of his shots, even if Griezmann’s tend to come from better positions. This is a crucial trait to have at a side like Barça, where many teams will try to defend with a low block and everybody behind the ball. Leroy Sané clearly has the ability to make something out of nothing in these situations, which would have provided Barcelona with a much-needed attacking spark alongside Lionel Messi.
Chance creation has been one of the most criticized aspects of Grizmann’s performance this season. He often seems too pragmatic with the ball, and compared to the likes of Messi and Suárez he very seldom makes that final pass or creative play. While the Frenchman’s stats aren’t bad at all, Leroy Sané’s completely blow him away.
A clean sweep for Sané, as he betters Griezmann in every stat, and by a comfortable margin. It’s clear to see that the German is the more adept of the two at providing for teammates.
For expected assists per 90 minutes, Sané’s tally is more than double that of Griezmann. It was also good enough to be the third highest in the Premier League that season – Griezmann is currently forty-third in La Liga.
In terms of key passes – passes leading directly to a shot –, passes into the penalty area, and through balls, Sané’s numbers are at least 50% better than Griezmann’s for each stat. This trend continues for their shot and goal-creating actions, which are the two offensive actions – pass, dribble, etc. – leading to a shot or goal. Sané’s 0.77 goal-creating actions per 90 minutes were third in the league, while Griezmann is currently thirty-third.
Ultimately, these numbers reflect what has been clear for all Barcelona supporters to see: Griezmann lacks the dynamism and creative ability to play out wide. A player with the profile of Sané can be far more involved in the attack, and uses his skill set as a natural winger to open the game up for his teammates.
It’s surprising that Griezmann’s dribbling does not get talked about more often. Maybe it’s because he doesn’t even attempt it enough for people to notice it as a weakness, or maybe it has just been ignored. Either way, the Frenchman’s stats for carrying the ball and beating defenders have been atrocious this season, and they reflect another key characteristic that a Barça winger has to have.
Right off the bat, Sané completes an astonishing five times more dribbles per 90 minutes than Griezmann. For more perspective, that 2.9 dribbles per 90 was the fourth highest in the Premier League that season. Meanwhile Griezmann’s 0.57 dribbles have him one-hundred fourteenth in La Liga. One-hundred fourteenth!
Even more impressive is the fact that Sané’s dribble success rate is significantly higher, so he dribbles with greater volume and efficiency. This shows a real lack of ability to take defenders on for Griezmann, and an abundance of that ability for Sané.
Also important is the progressive distance of the two players’ carries, or the total distance they dribble the ball towards the opposition goal. Lionel Messi currently leads La Liga in this stat, but with him aging, it’s vital for Barça to get a pair of fresh legs in the attack who can help share the burden of advancing the ball. Sané, who progresses the ball 2.86 times further per 90 minutes than Griezmann, could have been this player.
Griezmann does draw 0.04 more fouls per 90 minutes, and gets dispossessed far less. Nonethless, that isn’t much of a positive when you factor in how little he does on the ball compared to Sané. The value gained by beating a defender in the attacking third can be far greater than the value lost by getting dispossessed. Ultimately, risks have to be taken to create goals, results, and even entertainment.
These dribbling metrics point out another vital flaw that prevents Griezmann from fitting in as a winger. As previously mentioned, Barça forwards need to be able to produce something against a deep-lying defense. But when Antoine plays out wide at Barcelona, he lacks the ability on the ball to break down a defence, an ability which Leroy Sané definitely has.
While it is most of the time the least important aspect of the game for a Barça forward, and the least entertaining, defending is still a job that must be done. Especially with Messi conserving his energy more and more for the attack, he needs players around him who put in a lot of effort. Defensive effort is the one area in which nobody can fault Griezmann this season, but Sané can do this job as well.
Note: For these charts, the defensive stats of the two players are shown only in comparison to the other players in their league with forward listed as their primary position by Football Reference
While Griezmann’s ball-winning stats – tackles and interceptions – stand out more in relation to the league, Sané’s aren’t too far behind. In fact, the German had the highest combined tackles and interceptions per 90 minutes out of Manchester City’s primary forwards that season: Sergio Agüero, Gabriel Jesus, Raheem Sterling and Leroy Sané himself. So while there were other Premier League forwards winning the ball more than Sané, he has definitely shown that he can handle being the most defensively active player in a front three.
Additionally, the area of defending in which Leroy Sané seems to excel the most is the most important area for Barcelona forwards: pressing. A successful pressure is given when the player’s team wins possession of the ball within five seconds of that player applying pressure on the ball carrier. Antoine Griezmann is currently sixth out of La Liga forwards for successful pressures per 90 minutes, and first for Barcelona forwards. Sané was eighth in the league and first for Manchester City.
Both Sané and Griezmann also press with very high efficiency. Sané’s success rate of 33.50% was the best in the whole Premier League for forwards, and a whole 4.2% higher than second best Sadio Mané, from Liverpool. Showing more similarity between their defensive output, Griezmann is currently second out of La Liga forwards for this metric.
While Sané has some very good defensive numbers, it’s undeniable that Griezmann appears the stronger player here. Especially when looking at the Frenchman’s numbers relative to other La Liga forwards, his defensive output is truly elite. Rather, these numbers serve to show that Leroy Sané is comfortable and willing to do defensive work. He’s a great presser, and took on a lot of defensive responsibility in Manchester City’s attack, so these numbers are more to refute the claim that Leroy Sané wouldn’t fit Barcelona due to his work rate.
With all of this in mind, it’s clear for anyone to see that Leroy Sané ticked all the boxes required to be a great Barcelona winger. Even without looking at any statistics, it’s obvious that he would have added width, pace, and far more dynamism in attack than is currently supplied on the left by Griezmann. The numbers just back this up, and make it even more frustrating as to how the Barça board didn’t make more of a push for the German.
The main benefits of signing Sané would have clearly been a vast improvement in terms of creative play and dribbling. On top of this, he still gives his side brilliant finishing and goal scoring output, as well as the pressing intensity and defensive responsibility that Griezmann has. The transfer of Leroy Sané to Barcelona seems like it should have been a no-brainer, but that isn’t really how the board operate.
Especially when you take into account that Sané is still only twenty-four years old – five years younger than Griezmann – and ended up costing Bayern a bargain price of just around €50 million, it seems this was just too smart of a deal for Bartomeu and co. to make. Instead, that money is being invested into aging players who don’t fill any necessary roles. Compared to a club like Bayern who have made their transfers with supreme intelligence and negotiating to replace their older generation, Barcelona are miles behind.
Using Machine Learning to predict Barcelona’s 2021/22 league season
Johan Cruyff famously said, “I’ve never seen a bag of money score a goal.” It is indisputable that money is not the only deciding factor behind a football club’s success. Time and time again, we have seen that football is chaotic and absurd. Anything can happen.
That being said, it is evident that statistically, the clubs with more money are usually the more successful ones. While one cannot predict the unpredictability that is engraved within football, an outcome that is statistically probable can be predicted. This article will explore the relationship between the points a club accumulates in a league season and the total value of the club’s squad. Then, Python will be used to build a linear regression model in order to predict the number of points Barcelona will obtain in the 21/22 season based on the club’s hypothetical squad.
Let us begin with visualizing the relationship between the total league points obtained by a club and the club’s total squad value. Since the amount of money spent on transfers has increased tremendously over the course of the last decade, in this article, only seasons 2014/15 – 2019/20 will be considered. Upon plotting the total points against the total squad value, it is evident that there is a positive correlation between the two. Granted, the correlation is not extremely strong, but it exists nonetheless.
A line of regression has been generated. The line’s gradient is approximately 0.05138986, and its y-intercept is approximately 44.5470726. There is only one dependant variable in question, club value/total value of squad, and so our line of regression can be modelled by the equation:
P = (0.05138986 )V + 44.5470726 where P is the points and V is the total squad value.
club_values = np.array(df['Values']).reshape(89,1) club_points = np.array(df['Point']).reshape(89,1) from sklearn.linear_model import LinearRegression linear_model = LinearRegression() linear_model.fit(club_values,club_points)
In order to predict the points accumulated based on Barcelona’s squad value next season, the squad value won’t directly be inputted into our equation above. The coefficient and the y-intercept have been rounded up after a certain number of decimal places, and so, to preserve the precision of the prediction, Python’s LinearRegression predict method will be utilized.
Now comes another challenging part of coming up with Barcelona’s hypothetical squad next season. Based on reports, it can be assumed that Samuel Umtiti, Miralem Pjanic, Martin Braithwaite and Phillipe Coutinho will be some of the notable players to leave. To compensate for the outgoing players, it can be assumed that Barcelona will sign Eric Garcia, David Alaba, Georginio Wijnaldum and Memphis Depay. Here is a table that contains all the players in Barcelona’s hypothetical 21/22 squad and their corresponding market values:
|PLAYER NAME||MARKET VALUE (MILLION POUNDS)|
|MARC ANDRE TER STEGEN||79.34|
|FRENKIE DE JONG||84.63|
|KONRAD DE LA FUENTE||2.64|
The club’s total value comes up to be £786.60 million. It is important to note that a player’s value is volatile. Transfermarkt updates player values every few months, and so it is extremely probable that this squad’s market value will be drastically different next season.
It is also important to understand that Transfermarkt’s market values are not official and so they are not always an accurate representation of a player’s true market value. Moreover, a player’s value is not always an accurate representation of their quality. Age is a key factor in determining market value, and so while a young player might have a high market value indicating that they are a bright prospect for the future, they might not necessarily be an excellent player at the time when the value is assigned.
Now that all of this has been established, the only thing left is making the prediction.
The machine learning model predicts that Barcelona will obtain 85 points in the 21/22 La Liga season. In the 19/20 season, Real Madrid won the league with 87 points, and in the 18/19 season, Barcelona won the league with 87 points as well. The difference between the points accumulated by the last two winning La Liga squads and the predicted points that Barcelona’s 21/22 squad will obtain is a mere 2 points. This shows that it is possible for Barcelona to win the league next season.
Since the correlation between the two variables, as observed from the data from the last 6 seasons, is not even close to being extremely strong, Barcelona’s 21/22 league points tally may be drastically different from what was predicted by the machine learning model. However, at the end of the day, the model can only work with the data it is exposed to.
Erling Haaland can change the course of Barcelona’s season. It would take forever to make predictions based on Barcelona’s hypothetical squads that include all of the club’s potential transfer targets, however, due to Haaland’s influence, an exception can be made for him. Laporta’s interest in signing Haaland has been widely reported.
However, he will cost a fortune this summer, and with Barcelona’s debt situation, it is unlikely that the Catalan club will pursue him in the upcoming transfer window. That being said like mentioned previously in the article, football is chaotic, and anything can happen.
Assuming that Barcelona does end up signing Haaland in the summer, Antoine Griezmann will most definitely have to leave (due to wages and the money that can be generated from his sale). Haaland’s current value is £99 million, and so Barcelona’s total squad value for next season, after being corrected for Griezmann’s exit and Haaland’s arrival is £831.60 million.
Haaland’s addition can put 21/22 Barcelona on level with the previous two Spanish champions in terms of total points won. The difference in points between a squad containing Haaland and one not containing him is only two. However, an argument about whether signing Haaland is worth it, is not in the scope of this article.
“Transition seasons do not exist at Barcelona.”Joan Laporta | Presidential campaign
A club of Barcelona’s stature will always compete for titles. The machine learning model shows that Barcelona is in contention for the 2021/22 La Liga title. Ronald Koeman has done a splendid job at the Blaugrana club this season, and there is no logical reason for him to not do an even better job next season after getting fresh players in the summer.
While football at its core is unpredictable, statistically, it would not hurt to trust Barcelona to win the 2021/22 La Liga title.