Analysis iD

Using Data to Find Players Who Could Be Great in Another Position

Can we find the next Alphonso Davies?

INTRO

I’ve always been interested in players who could potentially excel in another position. Player recruitment is all about find good players for good value. The opportunity here, is to find players who are undervalued in their current position but could be of considerable value in another.

Anyone who follows me on Twitter will likely be aware that I (far too often) bang on about how Willian should play as a number 8. Physically he’s superb, fast, strong, great stamina, he’s very tidy technically, he can pass, he can dribble… and the area of his game he gets most criticism for – his end product – is less crucial if he plays as a central midfielder, as opposed to as a wide forward. I think 3000 words on WHY WILLIAN SHOULD PLAY AS THE 3RD CM IN A 4-3-3 may be too niche a blog, even for me.

The idea of getting more from a player than he is currently giving is intriguing, though.

This concept takes place organically within clubs. Coaches may see a player who is not affecting the game hugely in his current role but may have the skillset to be more effective elsewhere.

From a recruitment point of view, the theory is: could you find a player who is not a “star” in his current role at the moment, and hence his market value is relatively low, who could be bought cheaply and successfully repurposed into a new position? Hopefully, this player could either become an important cog in your team (great) or be flipped for a profit (also pretty great).

For the likes of PSG and Manchester City, this probably doesn’t apply. These clubs have the budget to simply go out and buy the best players on the planet. If you do not have financial firepower of Europe’s super-clubs, there is a need to be a bit more creative in the market.


 

WHERE TO BEGIN?

The winger to full back transition seems to be a good place to start. This is a well-trodden path in football. What is common, in a lot of these cases, is a young midfielder who perhaps doesn’t quite have the quality in the final third to be a winger at the very top level but has great athleticism and decent defensive instincts – think Aaron Wan-Bissaka.

There are numerous other examples of players who have taken to the full back role later in their career.

  • Marcelo – former winger/forward
  • Kieran Gibbs – former left or central midfielder
  • Emmanuel Eboue – former winger
  • Bellerin – former winger
  • Trent Alexander Arnold – former winger or central midfielder
  • Dani Alves – former winger
  • Jordi Alba – former winger
  • Patrice Evra – former winger
  • Lauren – former winger
  • Achraf Hakimi – former winger
  • Alphonso Davies – former winger

Both Bukayo Saka and Ryan Sessgnon, could potentially be placed in this category also. It could be that argued that both could play at a higher level at full back than they would reach as wingers – though both do have very good instincts in the final third.

Alphonso Davies is a particularly interesting case here. He’s become on the of the best left backs on the planet after shifting back from his accustomed wide-left role. He can be quite positionally naïve. It’s not uncommon to find him slightly out of position defensively, but his insane athleticism gets him out of trouble. And to be clear, that is not a criticism, that is a fantastic asset to have. He’s not a hugely natural defender, but why should he be? This is a guy that would have been by far the best offensive weapon on his school team, his local Sunday team, his youth team and likely his early professional teams. Thomas Muller spoke recently about how Davies must continue to work to improve his positioning, whilst also referencing his road runner-like ability to recover.

SIDENOTE: It’s either a great or terrible time to post this blog in the context of a tweet that went semi-viral recently. The tweet, from EiF (https://twitter.com/EiFSoccer/status/1264274946173394944?s=20), referenced Davies and talked about how 1v1 defending is all about athleticism now. This was widely derided on Twitter, and in the main rightly so. The assertion that 1v1 defending is “not inherently a skill” is clearly complete nonsense. However, being a monster athlete DOES make defending easier. This is something that is true in all sports, not just football. Van Dijk is great but wouldn’t be the same player if he was not 6”4, fast, and a powerhouse of an athlete. This doesn’t mean you can stick any old sprinter at full back, but that level of athleticism is undoubtedly a huge asset, when coupled with other skills.

Everything we have seen and heard about Davies so far indicates a player who is insanely fast, powerful, agile, can run all day, has quite a natural game understanding, is good in possession, with good technique and general coordination with the ball… and is a quick and willing learner.

While it’s unlikely we can find the next Alphonso Davies easily, this provides a decent blueprint for the kind of player who could make a similar positional transition.

METHOD

In the search for a player that could be repurposed as a full back, I decided to start with the Bundesliga (2019-2020 season) – not least as I was able to obtain some rare physical data for the league. Useful.

A minimum threshold of 500 mins played was applied to the data set, which enabled some of the outliers, those players that charge around like mad men in their brief cameo appearances, to be filtered out.

[Updated data definitions (28/06/20) below – thank you to the analyst for a top Bundesliga side, who I will assume wants to remain anonymous]

The physical metrics used are as follows:

Sprints p90 – how many times, per 90 minutes, a player makes a run of at least 2 seconds faster than 14.4 km/h and at least 1 second faster than 22.7 km/h (a new sprint cannot start until the players speed has decreased below 14.4 km/h)

Tempo Runs p90 – how many times, per 90 minutes, a player makes a run of at least 2 seconds faster than 14.4 km/h and at least 1 second faster than 18 km/h, but without reaching 22.7km/h for more than 1 second (a new Tempo Run cannot start until the players speed has decreased below 14.4 km/h)

Intensive Runsthe sum of Tempo Runs and Sprints

To simplify, a max-intensity run would be a “sprint” and 70% intensity run would be a “Tempo Run”. Collectively, these numbers should give a good view of a player’s sprinting capacity and their level of stamina.

DATA

The below plots Bundesliga outfield players on their performance in the aforementioned two metrics, Sprints p90 and Tempo Runs p90. The node size is related to how many kilometres they cover per 90 minutes, providing another view of each player’s fitness level. Essentially, the larger the dot, the greater distance that player runs on average per game. Age is also displayed on this chart, the darker red the node is, the younger the player.

This graphic outlines the players with the most impressive athletic profiles in the league with some of the best performing players highlighted by name.

Achraf Hakimi, himself a winger/full back hybrid, profiles well here as a high-volume sprinter. In case you were wondering where Alphonso Davies is – he’s the dot right next to Hakimi, profiling almost identically, in terms of sprints, tempo runs and km covered, to the Moroccan speedster.

An obvious but important consideration when attempting to convert a player to a new position is whether they would actually want to play there. There are rumours that Ainsley Maitland-Niles had a falling out with Mikel Arteta over a lack of enthusiasm for playing full back and looks likely to leave the club this summer, despite (in my opinion) performing extremely well in that role. Professional athletes have egos, and are human beings, if they genuinely do not want to play a specific role, there seems little point persisting with the idea.

It seems unlikely that 31 year old Ivan Perisic would fancy a late change to full back, and given his profile within the game, the club he plays for and his market value and wages, it is probably worth excluding him from further analysis at this point.

Ismail Jakobs is particularly striking here. 2nd best in the league for Sprints per 90, extremely impressive for Temp Runs also and top ten in the league overall for Intensive Runs p90. All while covering 11.2 km per game. That is a monstrous profile.

Felix Klaus is off the charts when it comes to Intensive runs p90. Karim Bellarabi profiles very well and is a player who’s played a number of roles for Leverkusen, including right back on occasion. Benito Raman is a sprinting machine but may be difficult to persuade to move to full back, as he scored a bunch of goals as a forward for Schalke last season. Worth keeping in for now. There are some other interesting names in there, too.

So, we’ve identified the most physically impressive players in the league who don’t currently play as defenders.

If we take these players (those highlighted by name) as a sort of shortlist of options, we can dig deeper into the data and see if we can narrow the net further. We’ll include Hakimi and Davies, as blueprints of the kind of players we are looking for.

That dot right in the bottom left hand corner is not a goalkeeper by the way. It is in fact one-time Liverpool loanee Nuri Sahin, who prefers to do things at his own pace, clearly.

If you want to have a play with the dashboard and see how different teams, players, positions and ages compare, you can do so here:

DIVING DEEPER INTO THE DATA

As mentioned previously, what we are not looking to do is find any sprinter and shove them at full back. There is a need to get view of the other skills of the chosen players to inform how successful a positional change may be.

The following graphic plots each of our shortlisted players, in relation to each other, across 4 different metrics, and split by positional group.

The four metrics and the rationale for each are as follows:

Expected Goals per 90 (xG p90) and Expected Assists per 90 (xA p90)

The modern full back needs to be able to contribute in the final third, so how well they get into scoring or assisting positions is important. It’s worth noting that if we are looking for value in the market, we aren’t necessarily looking for the best performing players in these metrics as these players are likely to be expensive. It’s important to consider also that by moving a player from a midfield or forward role to a full back role, you are likely to see a change in the xG and xA they are producing currently. However, some visible indication that they could potentially contribute in the final third from full back would of course be a positive.

Defensive Pressures per 90

Using this metric will aim to get a view of how active and willing players are to contribute in the defensive third. How often are these players tracking back into the defensive third and pressuring the ball carrier? As with all pressure metrics, this is playing-style dependent. A low score in this metric may be more of an indication of what he is asked to do by his coach, rather than what he is capable of. A high score here would suggest an active and willing defender.

Dribbled past per 90

How often, per 90 minutes, is the given player dribbled past by an opponent? Defensive metrics are hard. I’ve spoken before about volumes of tackles or interceptions not being correlated with defensive ability. A player who is dribbled past at high volume, clearly, is not desirable; a player who is not dribbled past often, may indicate some potential as a 1v1 defender – or at least a player who doesn’t allow opponents to drift past him easily.

How our shortlisted players compare to each other can be seen below:

The two players who stand out from within our shortlist are:

Both are relatively low in xG and xA per 90, with Becker the more impressive creator here with a respectable 0.18 xA p90.

Jakobs is third best in the entire league for defensive pressures (8.4 p90) and the best “forward” in the data set in this regard, illustrating a player who, though positioned further forward, is highly active and willing in the defensive areas of the pitch.

Jakobs (0.62 p90) and Becker (0.73 p90) profile extremely well when it comes to how often they are dribbled past per game, suggesting two players who make it difficult for opponents to go past them. This is a promising asset for a would-be full back.

For reference, Alphonso Davies (0.96) and Hakimi (1.58) are dribbled past more regularly, per 90 minutes, than Jakobs and Becker. There are two contextual factors at play here: Davies & Hakimi play as full backs so may face more situations where an opponent is attempting to dribble past them. Though they both play in ball dominant teams, so potentially face less defensive situations in general than the average full back. Regardless, these are impressive numbers for Jakobs and Becker.

How Jakobs and Becker profile comparative to all Bundesliga players in their position group can be seen below:

Jakobs is particularly impressive in the numbers. The volume of defensive third pressures coupled with his low dribbled past data, and a frankly ridiculous physical profile.

I had no idea until I checked TransferMarkt, that Koln have actually beaten me to it, Jakobs has played 1 top-flight game at left back, completing 90 minutes there at home to Augsburg in November. And from what I can discern, has played a bit there in youth football also. Not a completely novel discovery by me then, but more evidence, I suppose, that he could potentially be suited to that role.


 

VIDEO

The natural second step after a data filter is to look at the tapes.

So, what am I looking for in Jakobs & Becker?

Let’s imagine we are doing this for a club. We’ve sat down with the head of recruitment and coach and highlighted the key attributes we want in a full back to fit our game model and style of play.

Speed – He needs to be quick to help us in defensive/offensive transition

Power – We need a combination of speed and strength to ensure he is robust in duels

Stamina – He’s going to be an important attacking outlet for us so he needs to be able to get up and down the flank all day

Agility– It’s important that he is able to change direction quickly to defend 1v1. He needs good general body coordination in his movements. We don’t want out of control limbs giving away penalties and dangerous freekicks

Technical competence Our full backs are going to see plenty of the ball when we build up so he needs a level of technical competence (first touch, passing, ability to dribble to maintain possession, crossing etc)

Game understanding – we have good coaches so we can work with him, but he needs a level of game understanding (positioning in and out of possession, understanding of space, anticipation, alertness to danger etc)

The above can be broadly shaped into 3 categories: Physical, Technical and Tactical. If I can get a reasonable view of personality/mental characteristics through the video, then I will pass comment on this also.

The data presented up front suggested both Jakobs and Becker are high volume sprinters and extremely fit but couldn’t give us a view of their top speed. On first view, both players have impressive speed and acceleration, a key criterion for the type of full back we are looking for, as demonstrated below:

 

Indeed, Becker has been credited with the 8th fastest sprint in Bundesliga history at 35.43 km/h (https://www.bundesliga.com/en/bundesliga/news/achraf-hakimi-sets-new-speed-record-fastest-player-borussia-dortmund-8808) and Jakobs has clocked 34.5 km/h this season (https://twitter.com/Bundesliga_EN/status/1237030381611704323?s=20)

They are both suitably rapid, then.

The data has given us confidence that both possess high levels of stamina as indicated by their excellent volume of their sprint, tempo runs and distance covered per game numbers. Any supporting information from the video will be called out for each player as appropriate.

 

IN POSSESSION

Dribbling and general technique

Jakobs is mostly secure technically. He has a decent first touch, is able to receive the ball and protect it with his athletic frame and has fairly good coordination with the ball at his feet. One area for development for Jakobs has is that he is heavily one footed. He is generally reluctant to pass with his right foot and his technique can let him down when he does. As a dribbler, he can use body feints to dribble and maintain possession. Though he is able to beat a man in transition with his explosiveness, he is not a player who can regularly beat a defender 1v1 in tight spaces, often preferring to look for a pass backwards or inside.

 

Crossing

Jakobs has a good left foot, albeit with slightly inconsistent technique. Due to inconsistent ball striking, he has the occasional issue in not beating the first man. At times, he lacks a little bit of guile to create space off the dribble to get his cross in, meaning the cross is blocked and a potential opportunity missed. He gets his head up before crossing and is aware of teammates movement; and when he gets it right, his crosses are played in at pace and dangerous.

 

Passing

His passing is mostly safe/conservative, he often has a picture in his head before the ball arrives and looks to move it quickly. He is reasonably adept in tight spaces and able to move the ball effectively in one or two touches. I was impressed with the way he was able to combine with teammates down the left-hand side, finding overlapping/underlapping players and making both of those types of runs himself. He lacks a little bit of quality in playing the “killer pass”, often the execution is ever so slightly off (e.g. just behind a teammate) but encouragingly the idea is often good, demonstrating good game understanding.

 

OUT POSSESSION

Defending – Physical

Another box Jakobs ticks is that he is agile. In the first few clips in the below, he demonstrates his ability to change direction quickly allowing him to stick closely to opposition forwards.

It was highlighted by the data and backed up by the video… Jakobs is a monster athletically. He is a pressing machine. He can sprint 30 yards forwards to press the full back, turn on a six pence and sprint 40 yards back and force a turnover. His pace and athleticism are huge assets defensively. He can stop transitions, he can press opponents and force errors and, much in the same way that Alphonso Davies does, he can make rapid recovery runs that get himself and his team out of trouble. He is obviously hugely quick, but he is powerful as well. He is aggressive, committed, willing, and defends with great intensity – you can see why the data suggested he does not get dribbled past very often. Jakobs also competes well in the air. At 6”0 tall, this is a useful asset for a full back to defend crosses at the back post.

 

Defending – Technical/Game Understanding

Jakobs has the occasional issue in not setting himself when defending 1v1 following a recovery run. By not slowing down and taking an active defensive stance, Jakobs can be off balance and be skipped past by an opposition forward, as evidenced by the first clip in the below. Overall, I was impressed with the young German’s defensive awareness, positioning and anticipation. His previous experience playing as a full back is evident – he has a good nose for danger and is able fill in gaps and take up largely good positions defensively. Where he has a sub-optimal starting position, his speed allows him to recover and make a block/interception.

 

Becker makes it into this analysis having played just over the 500 minutes threshold. The fact that he has not been a regular this season (only 17% of the time has he been in the starting XI for Union Berlin) tells us:

  • His market value would likely be low
  • He may be open to a move
  • There’s a question as to why he hasn’t played more for a team that is not far off the relegation places in the Bundesliga

IN POSSESSION

Dribbling and general technique

Technically, Becker is largely clean, demonstrating a pretty reliable first touch and good coordination with the ball at his feet. Becker is a better dribbler than Jakobs, both in quantity and quality. This is reflected in his numbers also, completing 3x more successful dribbles than Jakobs per 90 this season (1.58 compared to 0.51) and at better success rate (68.2% compared to 47.1%). While not spectacular, Becker is a competent and agile dribbler. He has quick feet and is more of a threat 1v1 than Jakobs. Certainly, he is more willing to use his weaker foot, as demonstrated in the below by his propensity to go on the outside when playing from the left.

 

Crossing

Generally, Becker has pretty good technique on his crosses. He is able to shape them nicely and does, at times, put the ball into the box with good pace. He has a slight issue when running at full speed in that his ball striking can be inconsistent, though this is, of course, a difficult skill. His delivery with his left foot is just ok, but his willingness to go inside or outside when 1v1 is a positive and makes him less predictable to defend. Some of his crosses are evidenced below.

 

Passing

Union Berlin are not a possession dominant side and with his limited minutes, it is difficult to get a substantial view on his passing quality. From the clips shared above, it suggests a player who can do the basics with his passing, but lacks quality/guile with his final pass.

OUT OF POSSESSION

Defending – Physical

Like Jakobs, Becker’s pace is a big asset defensively. His speed can help to force turnovers and allow him to recover into position in defensive transition. Becker does however lack intensity in his defending, at times. He is certainly not as tenacious or willing a defender as Jakobs, picking and choosing his moments to defend aggressively. It is important to remember the data does suggest that, comparative to the rest of the Bundesliga he does produce a very high number of sprints per game, and a good level of Tempo Runs. The caveat here is that he doesn’t start a lot of games so he may be “fresher” for the minutes he is on the pitch. With regard to ground duels, Becker is athletic but not hugely powerful. Generally, he competes fairly well, though he can be outmuscled by more powerful players. The Dutchman is fairly poor in the air, and often looks uninterested in competing, in truth.

 

Defending – Technical and game understanding

The first clip in the below is an example of Becker momentarily switching off and being a little slow to react, which leads to a midfield runner going past him and a dangerous situation for the opposition. At times he shows fairly good awareness of danger, filling gaps and taking up some decent positions to defend. He is not a natural defender and appears is keener to go forward than he is to run back the other way (which is to be expected given his current role). It would depend then, whether this gap could be covered with good coaching and his considerable athleticism, though he would certainly need to find more consistent intensity in his defending.

 

Transitions

One big asset Becker would provide from a full back position is his ability to turn defence into attack following a turnover and his ability to carry the ball over long distances. He is an excellent athlete and would suit a team that target transitions as a means of progressing the ball.

Some examples of Becker’s transitions from defence to attack are evidenced below.

 

CONCLUSION

The question this piece aimed to answer was:

“…could you find a player who is not ‘star’ in his current role at the moment, and hence his market value is relatively low, who could be bought cheaply and successfully repurposed into a new position?”

What we got to was two superb athletes, with slightly different skillsets, both of whom could be interesting full back options. Both players have a market value of less than £2.5m, with the caveat that Transfermarkt values are a guide and it is not uncommon for actual prices to be north of these valuations.

We found a potential left back in Ismail Jakobs who is a phenomenal athlete, with good defensive instincts and intensity, and decent technical ability. And he’s 20 years old.

We found a potential right back in Sheraldo Becker who would be cheaper, is still a great athlete, has slightly more quality as a dribbler – though would need more work on the defensive side of his game.

If I was to compate the two, I believe Jakobs could operate at full back at a higher level than Becker, sooner, and be less of a risk defensively, though Becker would likely be considerably cheaper and still possesses an interesting skillset.

Did we find the next Alphonso Davies? Not quite. I’m not sure Jakobs has the quality with the ball to play for a real top team, even as a full back. But he is still a hugely impressive player. We found someone who could rival Davies physically, if not technically. And that in itself is no mean feat and of considerable value.

Not bad for a first cut.

There are plenty of ways this study could be improved. You could do a more thorough first filter with the data, to look at more in-possession variables. I might next time up the minutes threshold to more than 500 also, to look at players who have more data points (and video available). As always, feedback is most welcome, and it would be great to see others looking at creative ways to find value in the market.

The follow up to this blog will certainly include more creativity in the use of numbers, considering leagues where physical data is less easy to obtain.

As always, thanks for reading and do share your thoughts on twitter here: https://twitter.com/SanchoQuinn/status/1269297826669965313?s=20

 

Sancho Quinn, 6/6/20

Share this article

Share on facebook
Share on twitter
Share on linkedin
Twitter feed
Get in touch