Replacing Andy Robertson: Part 1

The first instalment of a series dedicated to finding a new left-back for Liverpool.

Alright, time for another series.

When I first launched this Substack, I used data to find a potential successor for Jürgen Klopp. And last summer, I searched for a wide player who suited Liverpool, after news emerged suggesting the Reds wanted a new attacker for Arne Slot.

This time around, I’m going to focus on our left-back spot. Those who have followed my content over the course of the current campaign will know my thoughts on Andy Robertson, who has been elite but now seems to be experiencing the autumn of his career. I wrote about that here.

The Scot is 30 years old now, and Kostas Tsimikas is 28. Both of them are largely fine, but this is Liverpool. We aspire for greater than fine. And if we’re hoping to upgrade our starting eleven this summer, the left-back spot is an obvious place to start.

---

It’s not entirely clear what Slot wants from his left-back. You’ve got to be supremely technical to play for him — that’s for sure — but the rest is kinda open for debate.

Maybe the Reds will follow the crowd by fielding a centre-back over there, who knows? But in my opinion, the Dutchman will want a relatively traditional full-back.

This season, we’ve seen Robertson stay behind the ball as part of a safety net, and we’ve also seen him run beyond to form part of Liverpool’s attacking bank of five, with both examples pictured below.

So you want someone who can do both, but let’s not overcomplicate things here. We want a player who is capable of overlapping and underlapping. Someone who offers a degree of threat when in possession. Delivers with quality from the wide areas. The obvious stuff.

Crossing in particular forms quite a central part of Robertson’s arsenal. Slot’s game generates certain opportunities for the captain of Scotland to feed teammates who are lurking around the penalty spot, with two specific avenues used pretty often.

As shown below, there’s the cross from the byline after running beyond, and there’s the more withdrawn half-space cross. They aren’t always successful, but our left-back will always get chances to use them, so we need whoever signs to possess enough quality to punish opponents in those situations.

I’m basically looking for another Robertson, in essence, albeit one who is closer to the age of 25. And if the player in question is capable of offering more solutions for Slot in a tactical sense — such as inverting into the middle like Trent Alexander-Arnold does on the opposite side — great.

---

But here’s the problem. As a general rule of thumb, the further back on the pitch that you go, the harder it becomes to accurately assess players using publicly available numbers.

It’s relatively easy when you’re looking for an attacker, because the guys who do loads tend to shine in standard event data. The more, the better. Goals, assists, shots, chances created, xG, xA, passes into the penalty box, progressive passes, whatever. It’s all there.

But if we take Virgil van Dijk as an example. Yes, he’s the best defender to ever play the sport, but he’s also averaging fewer tackles per 90 than Mohammed Kudus, Raúl Jiménez and Brennan Johnson this season. Wait, aren’t defenders supposed to tackle?

Indeed, Van Dijk averages fewer interceptions per 90 than Lisandro Martínez, Maxence Lacroix and Kenny Tete. But we all know he’s better at defending than all of them.

As you can see below, there’s no correlation whatsoever between doing stuff on the defensive side of the game and being a good defender. Using basic numbers to pick out the best defenders on the continent is like walking across a minefield.

My point here is that when you’re analysing defenders, the data can be noisy and deceptive. So in our search for a new left-back, rather than focusing too much on whether the next Robertson can regain the ball, I’m going to dedicate more attention to his possession game.

I don’t wish to overlook the defensive side, but we’re just making this whole process a bit more bulletproof by prioritising on-ball stuff. Don’t forget, Robertson himself was famously labelled as a terrible defender before his switch to Merseyside in 2017. You can coach that sort of thing.

---

That’s enough for this week. It’s all about setting the scene and understanding the problem at the beginning. Next time, we can start the scouting process from an initial pool of just over 2,500 players.

See you then.