Using data to identify a new central defender for Arsenal
We should start here with an insight into how to use data when looking to recruit players for your club. In no way should data be used as the definitive and only source of information when looking to take new players into your club. Data has to be combined with video and live scouting as well as with background checks on a player before you can have enough information to decide whether you are going to sign that player or not.
Why then is data so important as a first team in the recruitment process? Using and filtering data allows you to narrow and filter your search range to a point that is manageable. The purpose of this piece, for example, is to identify central defenders for Arsenal. The London club has the financial means to recruit from every market in the World but scouting all players in those markets, either via video or live, would be extremely time-consuming. So, instead, we use data to narrow our shortlist and identify players that we feel fit the playing style of the club that we are working for.
Why central defenders? Well, it’s quite simple really. Since Mikel Arteta took charge of Arsenal we have seen the immediate implementation of a more possession-based style of play with the central defenders taking responsibility for playing vertically to progress the ball into the opposition half. At the moment the options available to Arteta are somewhat limited. David Luiz has impressed, to a point, and Sokratis Papastathopoulous and Shkodran Mustafi are serviceable. None of those three feels like a medium to long term option under Arteta though. Rob Holding has shown potential but only in limited periods. We also should acknowledge that the club finalised a deal for the 18-year-old French central defender William Saliba last summer although the Frenchman is spending the 2019/20 season on loan back at St-Etienne. So with Saliba coming in next season and Holding showing promise we can assume that the club will choose to keep one of Luiz, Sokratis or Mustafi but they must then recruit at least one, if not two, central defenders to make a push on the Champions League places next season.
Before we consult the data though we need to understand what Arteta is likely to expect from his central defenders going forward. Firstly we know that Arteta prefers to play with a four-man backline and not back three. So far the Spaniard has shown a preference for playing a double pivot in midfield with the overall structure of the side being 4-2-3-1. The defensive line is relatively high and as such any central defenders that are signed will have to be capable of defending in space and when isolated against attacking players. In possession the central defenders have been specifically told to not release their passes until an opposition player moves to engage the ball.
In order to form our initial shortlist therefore we will be using two sets of data. First, we will consider the defensive line of the game and look for central defenders who perform well in terms of defensive duels. Secondly we will consider how they perform in possession by considering the amount of progressive runs or passes they make per 90 minutes.
In this data analysis, we will initially filter our data to only consider players under the age of 25 who have played a minimum of 1000 first-team minutes this season.
The initial data check
Here we see the initial data check for players plotting their defensive duels per 90 minutes and their % of defensive duels won. Initially when running these checks you are looking for any outliers or high performers who stand out in order to add them to your scouting shortlist. For the purpose of this article, however, I am combining the data with prior knowledge to identify players within the data set that I believe would fit with Arteta’s preferred style of play.
The first name that stands out and performs best in the data is actually an English player. Ben White is currently playing in the English Championship on loan from Brighton and his performances so far this season have led to him being widely regarded as one of the most promising young English players. In terms of playing style we know that Arteta has learnt a lot from his time working with Pep Guardiola and Guardiola, in turn, lists Marcelo Bielsa as one of his primary influences. Ben White is currently learning under Bielsa at Leeds and he is expected to defend when isolated on a regular basis. The next player that I have picked out is Clement Lenglet of Barcelona. The 25-year-old central defender is currently on the books of Barcelona and he plots just below the average line in the bottom right quadrant. Just below Lenglet is something of an outlier as I have highlighted the statistical output of Takehiro Tomiyasu of Bologna. The 21-year-old Japanese international is capable of playing as a right-back or as a central defender. he has been identified as he offers flexibility within the squad. Next is another English player and another who is on the books at Brighton in the 25-year-old Adam Webster. Brighton only signed Webster last season from Bristol City but he has already shown himself more than capable of playing at the top level. The final player that we are identifying from this dataset is the 21-year-old Stade de Reims defender Axel Disasi.
Next, we need to drill down into how they perform in possession.
The two standout players from our initial shortlist then are Clement Lenglet and Takehiro Tomiyasu both of whom grade out well in terms of progressive passes. Tomiyasu is slightly better in terms of progressive runs but this is likely attributed to his time spent at right-back. Adam Webster is next and he is still in the golden top right quadrant with Ben White right on the average line. Axel Disasi plays less progressive passes than any other on our five-man list.
When using data to identify any targets we have to put the data into the context of the team that they are currently playing for. Clement Lenglet, for example, plays for Barcelona and in a possession orientated team that defends high. Axel Disasi, on the other hand, plays for a more pragmatic team that defends in a low to medium block. As such you would expect their statistical outputs to be different. We have already mentioned that Ben White plays in a system similar to the one that Arteta is likely to use and Tomiyasu plays in a system that is less progressive than some of the others. Adam Webster is an interesting player given that he plays for a coach in Graham Potter who is tactically versatile and flexible.
So we have five options all of whom have different skillsets. Next, we use more data to filter them down still further.
First of all, let’s consider Clement Lenglet. Now we have added extra metrics to further delve into his playing profile with aerial duels and passes into the final third giving extra context to his performances. This season Lenglet is averaging 4.08 aerial duels and 7.39 defensive duels per 90 winning 62.50% and 65.52%. Defensively then Lenglet is not a standout but in possession, he does impress. So far he is averaging 8.60 passes to the final third per 90 minutes with an 82.2% success rate. He is also averaging 11.90 progressive passes and 1.40 progressive runs per 90 minutes. In possession then Lenglet certainly meets our profile. Defensively he is less of a sure thing and we have to consider the potential cost of getting him out of Barcelona with Quique Setien having just taken charge and given that the Spaniard also favours a possession-based style of play.
Now let’s turn our attention to Takehiro Tomiyasu. The Bologna defender is in his first year in Serie A having signed from the Belgian side Sint-Truiden. His aerial duels per 90 are slightly higher than we saw from Lenglet with 4.44 per 90 minutes but his success rate aerially is much lower with 48.48%. His defensive duels grade out similarily to Lenglet with 7.60 defensive duels per 90 and a 63.72% success rate. In possession, he plays 9.07 passes into the final third per 90 with a 77.78% success rate. Tomiyasu plays 11.02 progressive passes per 90 minutes which is an impressive number.
The aerial duels are a concern given the nature of the Premier League and the fact that against some teams (Burnley) defenders will be bombarded aerially. Tomiyasu is perhaps best served moving to right-back permanently although the flexibility that a player who can play two positions has can be appealing.
Axel Disasi is the biggest outlier of the group. The 21-year-old is enjoying his breakthrough season with Reims in the French top-flight and scouts are flocking to watch him in action. That does not, of course, mean that he would be a good fit to this Arsenal team under Arteta. Disasi is more dominant aerially than Lenglet and Tomiyasu with 5.70 aerial duels per 90 and a 57.39% success rate. He only makes 4.81 defensive duels per 90 minutes and this is likely a result of the defensive structure used at Reims. His success rate in terms of duels sits at 73.20% which is extremely impressive. In possession, Disasi is less productive than the others on this list. he is averaging 5.31 passes to the final third per 90 with a success rate of 61.68%. He is playing 7.59 progressive passes per 90 as well.
Without more of a sample of his ability in possession, it would be difficult to sanction a move tor Disasi based on data alone.
Adam Webster is the statistical stand out from the five players that we have identified. He is consistently strong in all of the metrics that we have identified and his experience already in the Premier League makes him an interesting prospect. He averages just 4.95 aerial duels per 90 but he has a 66.67% success rate. He is also averaging 5.64 defensive duels with a 67.68% success rate. In possession Webster also impresses with 6.39 passes to the final third per 90 and a 68.75% success rate. He also plays 9.45 progressive passes per 90 minutes.
Ben White is the only player on the list without top-flight experience. Despite that, he is an interesting prospect with a skillset that appears to match Arteta perfectly. He is averaging 5.23 aerial duels per 90 with a success rate of 52.63%. His defensive duels are the highest in our shortlist with 9.02 per 90 minutes and a success rate of 74.05% in this respect he is an absolute output monster. In possession, he is averaging 5.33 passes to the final third per 90 with a 60.66% success rate. He also plays 8.05 progressive passes per 90.
So, out of the five players above which would we put forward for our recruitment team to follow up on which are we picking as the best fits for Arsenal? We could take the easy option and target Clement Lenglet relying on the fact that he is coming from a huge club with a significant reputation. Lenglet is naturally left-footed and would fit nicely as the left-sided central defender but is he the right fit for Arsenal at the moment? First of all his cost would be prohibitive and as mentioned above Barcelona are now under Quique Setien who is likely to change the style of play to one that suits Lenglet perfectly. We are disregarding Tomiyasu at this point given our need for a central defender above all else although he will be kept as an option should we need to add squad depth at a later date. Axel Disasi is too raw at this point but should be monitored when he takes the next step in his career.
From the data above we would likely look to move forward with video and live scouting on two players in Adam Webster and Ben White. One is a fit immediately for Arsenal, Webster, while the other, White, profiles perfectly as a Pep Guardiola defender. That fact alone is attractive when looking to recruit a player for Arteta to build his defensive unit around.
Adam Webster fits the profile for Arteta. Comfortable in possession and capable of bringing the ball out from the back he fits with the need for defenders to be composed in possession and step out towards the opposition to open up the field and attract pressure. He is also quick over short distances and capable of defending back towards his own goal or when isolated in space. Capable of playing in a back four or a back three he offers versatility and the options for mid-game switches of system. Ben White is in contention simply because he is a monster when it comes to defensive duels. We do of course have to acknowledge that White’s data is at Championship level but the fact that he is thriving under Marcelo Bielsa makes him an interesting target. White is stronger defensively than Webster and is a good passer of the ball in possession. he is also younger and should develop more while Webster is about to enter his peak years.
The next step is for me to go away and consult the video on both players to produce initial scouting reports that will give more insight into the data that we have used above. Check back in next week for these scouting reports to be up on the site.