Replacing Andy Robertson: Part 3
The third instalment of a series dedicated to finding a new left-back for Liverpool.
Alright, here we go. We’ve reached judges’ houses. A sample of over 2,500 players has been whittled down to just 12. The guys who are still standing are legitimate candidates as we search for our long-term Andy Robertson successor.
If you missed the first two instalments of this series, I recommend catching up below before you read this edition.
So, here are the final 12:
Facundo Medina, Lewis Hall, Milos Kerkez, Andrea Cambiaso, Miguel Gutiérrez, Diogo Leite, Alejando Balde, Rayan Aït-Nouri, Diego Moreira, Sergi Cardona, Tyrick Mitchell, Leif Davis.
In this post, I’m going to use the data to compare the remaining players across a select few metrics, with a view to determining how each left-back performs for his respective team.
It’s not entirely about who is good and bad. It’s not quite as black and white as that. We want to know how each player offers value on the pitch. Which players are likely to be worth more points? Do any of the options have glaring limitations? Who personifies Liverpool’s playing style?
Think along those lines.
Let’s make a start. First, I’m curious about how often each left-back touches the ball, purely as a means of determining which players are heavily involved in possession, and which players are passengers.
But we’ve got a problem. Obviously Barcelona see far more of the ball than Ipswich Town, for example. So it’s hardly fair comparing the number of touches posted by Balde versus Davis.
To fix that, I’ve created my own metric. It’s basically my own take on usage rate, which is a basketball metric used to determine the percentage of a team's possessions that a player uses.
That might sound complicated, but it’s really not. In simple terms, Liverpool average 680 touches per match in the Premier League, cool. Trent Alexander-Arnold averages 84.5 touches per 90. Thus, as a percentage, that means the Scouse defender uses about 13 per cent of his team’s touches. And that’s quite a lot.
Below, you’ll see how each left-back compares for usage rate.
Nobody quite as involved as Alexander-Arnold on Merseyside — which isn’t a surprise — but it’s worth taking note of Hall. The ex-Chelsea defender is seeing plenty of the ball for Newcastle United. He’s used to being pretty integral to his team.
Balde, by contrast, looks like more of an outlet for Barcelona. Far less involved in possession, which perhaps offers an insight into his playing style while also acting as a nod in the direction of Pedri and Co.
Alright, now let’s take a glance at progression. We know which players are seeing the ball, but who is moving forwards towards the business end of the pitch?
I’ve used the same logic as usage rate for this one. Using progressive passes alone would’ve been unfair. So in addition to that metric, I’ve created my own again to identify how each player performs within the context of his respective team.
Medina ranks top. Hall looks good once again, with Balde perhaps further emphasising his profile as a player who stays high and receives progressive passes rather than playing them himself.
And just as an extra layer of evidence, you’ll find progressive passes combined with progressive carries below. The guys in the bottom right are essentially the passers, and those towards the top left are the carriers.
Thus, the players in between offer a solid blend of both. Shoutouts to Kerkez, Gutiérrez, Cambiaso and Aït-Nouri.
Now for the more important stuff. Well, in my head, at least. How much does each left-back contribute to his team’s ability to score goals? Pretty weird given we’re looking for a defender, but who cares?
We’re signing a starter for Liverpool, so we want to buy threat. It doesn’t matter where you play. Moreover, Robertson has contributed with his fair share of offensive output over the years. You don’t want to lose that.
Below, each player has been ranked according to his non-penalty xG and xG assisted — combined — per 90.
And below that, you’ll see how each player ranks as a percentage of his team’s attacking output.
Some interesting findings here. Gutiérrez doing a lot for Girona. And Notice the top progressive guys from earlier in this piece — such as Hall and Medina — now look pretty ordinary. It’s not easy to move the ball forwards from deeper areas AND deliver in the final third. You tend to do one or the other.
Also, I feel like I expected more from Kerkez. Based on what I’ve seen of him playing for Bournemouth, he feels more dangerous than Mitchell for Crystal Palace, for example, but the numbers kinda conflict with that.
At this point, given what I’ve just addressed, I’m interested to determine who is the best at combining ball progression with attacking output. Sure, you tend to do one or the other, but does anyone do both?
Aït-Nouri looks impressive here. And it’s worth noting that he’s posted these numbers while representing Wolves, who are currently 17th in the Premier League table. Gutiérrez also good, same as Cambiaso, Hall and Moreira.
That’ll do. We know plenty about our candidates now. In fact, I’m comfortable removing a few more names from our shortlist. Say goodbye to Medina, Leite, Mitchell and Balde.
After doing a bit of homework, I found out that Medina is primarily a centre-back. Leite, the same. Mitchell showed up as pretty vanilla across every department, which is unlikely to change given he’s 25. And I probably should’ve removed Balde last week. He’s just not leaving Barcelona, we all know that.
Dim the lights. Eight players remain.
Hall, Kerkez, Cambiaso, Gutiérrez, Aït-Nouri, Moreira, Cardona, Davis.
See you next week. Bring snacks.
I've publicly stated that I would have Alt-Nouri for 3 seasons now. I also like the "played in the PL" factor.
Great stuff! It would be awesome to see 2 players added to the graph, for comparison: Peak Robbo, and Robbo of this season. How would they compare to this pool, using these parameters?