Dusan Vlahovic — Analysing a Potential £70,000,000 Striker
A deep dive into the numbers behind this transfer window’s hottest name.
All stats used in this article were collected before Fiorentina’s 0–4 loss to Torino on January 10th, during which Vlahovic, obviously, did not score.
Dusan Vlahovic’s name has been in the social media transfer rumour mill this winter. Earlier in the window, he’s been touted as a potential replacement for Romelu Lukaku at Chelsea, should the Blues decide to part way after his less-than-well-received interview (calmer heads prevailed), and now he’s being linked heavily to Arsenal.
21 year old Vlahovic has done good numbers in Serie A in the last two years, scoring 15 league goals + 6 penalties last season, and is up to 11 league goals + 5 penalties so far for the distinctly midtable Viola. That has naturally lead to interest in him, which a started as early as this summer, with some asking whether he’s ready for a step up to a bigger club.
The best scouting piece written about Vlahovic is an analysis by Phil Costa for Scouted Football. You should go and read the whole thing, but Costa’s conclusion is that he is promising, but questions remain and it is unclear whether he’ll be able to easily transition into a style of play required in a top side. Costa concluded his article by saying,
“Of course, whether his future lies at any other club, be that Liverpool or Leicester, depends on what happens next season. Both Vlahović’s ability to replicate his goalscoring exploits, and Fiorentina’s ability to demonstrate just how well rounded he is, will prove vital in determining at what level his future lies.”
In this article, I’ll attempt to look at how he has fared in terms of scoring this season, how his all around game has progressed and whether he’s a safe bet or still a gamble for a top side. While I will mostly focus on stats, I will attempt to partially validate with the all-important eye-test.
Goalscoring — A statistical look
Of course the thing that jumps out at your first about Vlahovic is his goalscoring exploits. Here, to remove at least some of the mystique from the numbers, it is again important to note that he did not score 21 goals last season and 16 this season so far. He scored 15 goals and 6 penalties last season and 11 goals and 5 penalties this season. The ability to convert a penalty is important, of course, but one you can find in other players in your team, including a defensive midfielder who doesn’t score much otherwise and a sweet-shooting right back.
But what about non-penalty goals? This summer, as part of my semi-serious attempt to explore whether any of the Chelsea strikers on the team at the time are capable of scoring 20 non-penalty league goals (sad lol) in a season, I’ve looked at all the players who have scored 20 or more non-penalty league goals in the last 4 seasons. In that article, I’ve highlighted the three statistical components that go into the total number of goals a player scores in a season — total number of shots (shots_total), naturally excluding penalties, the averaged xG value of each shot (npxg_per_shot) and by how much the player overperformed his expected goals on a per-shot basis (npxg_outperform_per_shot). I have then compared these statistical components to these generated by Chelsea players. I am going to do the same analysis for Vlahovic.
The 20 non-penalty league goal mark is significant, because if Vlahovic continues at his current pace, playing precisely the same amount of minute per match he’s currently playing, he will join this fairly exclusive club, just barely. Here is a full list of the players in it, alongside their statistical metrics. I’ve added Vlahovic’s numbers from last season, as well as his projected numbers from this season to this table.
As you can see, continuing at the current pace, Vlahovic is projecting to finish the season with 21 non-penalty league goals, though it is worth noting that he will have played the most minutes to do so than any other player on this list.
But how does his goal scoring, when broken into its three statistical components compare to the rest of the players on this list? To examine it, I’ve first normalised these to a standard normal distribution. This means that for each statistical component, 0 means the average, 1 is one standard deviation above average, etc. I’ve then created a score metric for ranking goalscorers by averaging¹ the three normalised components. This part is a little technical, so the TL;DR is that this is simply a way to more easily compare results to one another without changing the ordering (the best player in a certain statistic will still be best in the normalised statistic, and the worst one will still be worst). The results are displayed in the plot below.
Each white dot represents one of the other 20 goal scorers from the set above. Red dots are Vlahovic’s stats from last year, and blue dots are his stats from this year.
Last year’s Vlahovic has the lowest score out of any player being compared to, which makes sense since he is the only one who hasn’t reached the 20 goal plateau. The cause was both his low number of shots compared to others, and his NPxG overperformance, which was also very low compared to others. However, his NPxG per shot was actually very slightly above average compared to the typical 20+ goal scorer.
This year, the story is different. Vlahovic has upped his total number of shots per game, though it remains below average. His NPxG overperformance actually shot up, but concerningly, his NPxG per shot went down significantly. In other words, he shoots slightly more, takes significantly harder shots, but finishes these shots better than expected.
This is a concerning trend, in my opinion. While it is expected that the quality of the shots taken decreases as you take more shots (if they were easy shots, you would have already been taking him), the drop in NPxG from 0.16 to 0.12 per shot is quite large and cannot be adequately explained by either him shooting more, or Franck Ribery’s departure, especially as Fiorentina as a whole are doing better in attack and generating roughly the same NPxG per shot as a team as last year. The improvement in his finishing, as evidenced by a rise in NPxG overperformance compensates for this, but NPxG overperformance is the component that has most season-to-season variability and least likely to endure, so this is not a good trade off.
It is hard to say how his stats project going forward. He has not shown the statistical profile of an elite goalscorer yet. In fact, his statistical profile resembles those of one season wonders such as Krzysztof Piatek. Some of the responsibility for his low shots and poor shot selection lies with the fact that he plays on an average Fiorentina team. A better team with better providers would be able to get him better quality shots. On the other hand, on a top team he would likely not be able to take 25% of his team’s total shots like he is doing right now for Fiorentina. And his finishing currently is at a level that is probably unsustainable. He has a total NPxG right now is 7.3 and he has scored 11 goals. In percentage terms, that’s an NPxG overperformance of 51%, a level that only two players above have reached in the last 4 years — Lionel Messi in 2018 and Ciro Immobile in 2017. In fact, the chart below displays NPxG overperformance of every player on this list in terms of percentage.
One thing that is striking is that while world class strikers may occasionally have seasons where they greatly outperform their NPxG, they don’t need to do so. Robert Lewandowski, in fact, has been in the top half of all NPxG overperformances just once (2020, where he overperformed by 33%). On the other hand, nearly every single one hit wonder on this list fits the pattern of greatly outperforming their NPxG.
All Around Play
The other concern Costa expressed in his article from this summer is around Vlahovic’s ability to mould into the more all-around player that a forward at a top team has to be. To look at his progression in various areas, I’ve looked at a his percentile ranking relative to other forwards in a variety of statistical measures.
Dusan Vlahovic statistical profile, last season vs this season
While Vlahovic has improved some aspects of his all around game, notably his passing and ability to get touches in the box, other aspects of his game, such as pressing, winning headers and dribbling, have declined. Some of that could be due to changes in the team around here and his role, but what’s particularly notable is how non-elite he is in all these aspects. In his article from this summer, Costa noted that ,“we haven’t seen anything yet to suggest he’s capable of being the complete package for the very best teams”. So far, half a year into his second full season at Fiorentina we are still not seeing that.
The Dreaded Eyeball Test
Statistical analysis is fine and informative, but you still have to have a look at the player to see if its missing anything. Now, I will admit that I have not made a habit of watching every game of a mid-table Serie A team, so I did the next best thing. I went and I watched every goal he has scored for Fiorentina since the start of last year, courtesy of this compilation and took some notes.
First the good things.
- He has a cannon of a left foot.
- He has good movement. There are some goals in this clip where he made smart runs into space that have paid off, either though a non-obvious pass or by being in position to to capitalise on a deflection or a defensive miscue. His far post run against Spieza (2:23 in the video) and Genoa (3:00) are an example.
And some not so good things
- He is ridiculously one footed. I think I counted 2 right-footed goals and that was him tapping into an empty net. There are several examples of him having to take an extra touch to get the ball onto his left foot before shooting (Consenza, 4:19 and Spezia, 5:20).
- He is fairly lumbering, both in terms of his straightaway pace, which is average, but not terrible, and especially in his dribbling ability which is very awkward, perhaps explaining why he’s in the bottom half of all strikers in terms of dribbles completed.
- Nearly every goal is him getting on the end of a pass and blasting the ball past the keeper. With the exception of 2 or 3 goals, and of course his long range golazos, there were very few examples of him creating space for himself to get a shot off with the ball at his feet.
- Which brings me to my final point, the standard of defending and goalkeeping in Italy continues to leave a lot to be desired². A huge amount of these goals had completely disorganised defenses, Vlahovic being allowed to run or simply stand in the box completely unmarked, comical goalkeeping errors or all three. A good example is the goal against Torino at 4:35. He is allowed to get a free header off with enough distance between him and surrounding defenders to make the strongest social distancing advocates pleased, and still it probably should have been saved had the keeper not just sort of comically flopped at it.
Conclusion
So is Dusan Vlahovic the next Ibrahimovic or the next Andrea Belotti or Krzysztof Piatek, the last few young Serie A strikers to cause a sensation only to fall back down to earth and never reach the same level again? It was not clear last summer, and it is still unclear right now. His high goal scoring this year comes mainly off unsustainably hot shooting, while the rest of his skillset remains unrefined. He will likely provide decent or even good goal scoring with the right service, because he can absolutely blast the life out of a ball, but any team paying what Fiorentina will likely demand for him will be gambling on him becoming an elite complete forward, and that seems like an awfully big gamble right now.
[1] I’m actually calculating the score by summing the three normalised components and dividing by √3. This is done for variance vs standard deviation reasons that are beyond the scope of this article, but the important thing is that it only makes the results easier to interpret by spreading them out, it does not change the ordering of where each player lands compared to each other. As you can see in this plot, the calculated score correlates well with the number of goals scored, with the exception in the bottom right being Luis Muriel’s absolutely ridiculous last season, where he scored only 20 goals, but did so in only 1436 minutes!
[2] Last 3 years, Serie A and Bundesliga have been trading places for the title of the highest scoring league amongst the Top 5 European leagues. This year, Premier League has caught up a little, but still remains behind these two
