Benchmarking is very important in data analysis. It lets us use a known quantity to compare against.
Specifically, I reckon many Celtic supporters would agree that Mikael Lustig and Kieran Tierney are a recent high-bar for full-back operators.
Neither were perfect, few players are, but not only did they perform at a consistently high level (Lustig is a much-decorated international tournament player and Tierney is a stand out at English Premier League level), but they were an excellent combination.
Lustig is a good benchmark of what a defensive full-back looks like.
Tierney is a good benchmark for what an attacking full-back looks like.
This season we’ve seen the rebirth of Anthony Ralston, the flexibility of Josip Juranovic asked to fill both full-back berths, the continuing development of Greg Taylor, and the signing of Liam Scales from the League of Ireland.
How are they doing versus our benchmarks?
Defending
I use two heavily aggregated measure.
Defensive Action Success Rate (and breathe!) is the percentage of defensive actions that are successful – e.g. tackles and aerial duels won.
Possession Won from Defensive Actions (lie down!) is the percentage of defensive actions that result in Celtic winning / maintaining possession.
Neither includes passing or other attacking actions.
I have also included other recent luminaries – Boli Bolingoli, Diego Laxalt, Jonjoe Kenny, Hatem Elhamed and Jeremie Frimpong.
At the margins we see Frimpong, who was really a winger positioned at full-back. And at the other extreme Lustig, who was arguably a centre back repurposed as a full-back.
In the middle, in green, is the current cadre.
All much of a muchness in all honesty. Scales, being the most physical and athletic, is slightly ahead by these measures.
But none of the current crop are first in both DASR and Possession Won percentage amongst their peers.
Note how Tierney, despite doing so much attacking, would still be up there in defensive terms.
That’s why Lustig and Tierney are a tough benchmark!
Build Up
The full-backs are often the first point of build-up to attacks. Even central ones when inverting under Postecoglou’s system.
Certainly, we can agree that modern Celtic full-backs need to be good at linking the play.
To measure this, I use:
Completed Passes – very simple measure of how many passes were completed and a useful proxy for influence and ability to get on the ball (also style dependent – I’ll come to this)
Secondary Assisting Passes – number of passes per 90m before the pass that creates the chance.
A little more differentiation here. Again, Frimpong is a complete outlier. He had relatively low involvement, but the most Secondary Assisting Passes. (Many Celtic fans will be hating that Kenny would be next!).
The current cadre are also difficult to split. The fact they are all to the right of the Completed Passes axis speaks to team style more than individual contribution. That is, under Ange Postecoglou, the team is more possession/control dominant and therefore, as a team, completes more passes.
Greg Taylor completes the least but has the highest secondary assist rate. His creativity is perhaps undervalued.
All current fullbacks have a slightly higher Secondary Assist rate than Tierney. Again, this may speak to style and their involvement in central areas.
We can also see the “defenders first” Laxalt, Elhamed, in the bottom left of this view.
Ball Progression
Ball progression recognises the ability to move the ball up the park towards the opponent’s goal!
This is measured using:
Pack Passes – the number of line-breaking forward passes per 90m Ball Carries – number of progressive runs + number of pack dribbles (runs that take opponents out the game) per 90m
Note: Lustig and Tierney are excluded as I do not have the Ball Carries metric for them.
Again, Frimpong is in a little island of his own on the top left. He is actually elite level at ball carries across world football. That is what that World Class looks like in graph form. You are welcome! At bit like a Marvel character, he has one unique superpower, and it isn’t passing the ball.
The bunching of the current crop bottom right is again indicative of style as much as ability. Players under Postecoglou are encouraged to pass vertically aggressively and all have higher Pack Passing rates than previous full-backs as a result.
Within that bunching, Juranovic has the better Pack Passing volume, and Ralston is the one who makes most progressive runs. His short dash past the full-back for a cross has become almost a trademark.
We can see how some of the other full-back selections were a poor fit for Celtic’s attacking style.
Creativity
Finally, we look at creativity:
Chances Created – the number of passes leading to a shot per 90m xA – the Expected Goal value of those chance creating passes, again per 90m
Now we start to see some differentiation amongst the current crop. What may astound is that Kenny (oh, look, there he is again!), Ralston, Frimpong and Taylor all have higher xA rates than Tierney although the now Arsenal stalwart has the highest volume of chances created.
A case can be made for Taylor to be the most creative chance maker of full-backs in recent times. Remember back to his Build Up stats as well.
Ralston also scores really well against this benchmark.
Juranovic has not been as creative as some might expect and indeed is below Elhamed and Bolingoli by these measures.
Scales scores more like a centre back than a full back.
Summary
A lot of data to digest there but I would draw attention to the following summarising points:
• Holistically, and explicitly on defensive strengths, Celtic have not replaced Lustig and Tierney with equivalent full backs
• The current crop are performing well by their peers but this is often down to performing to order, i.e. meeting team style requirements
• Taylor is potentially the most creative full-back Celtic have had in recent times but also the defensively weakest of the current crop
• Ralston continues to exceed expectations, especially creatively
• Juranovic shows great promise as a ball progressor but has more to come creatively
• Scales profiles more like a centre back than an attacking full-back
In the short term the current situation whereby Juranovic and Taylor are first-choice is arguably correct, with Ralston still getting plenty of game time.
I suspect the full-back area will be one that will continue to receive focus from Postecoglou over the summer.
Comments: Our rules
We want our comments to be a lively and valuable part of our community - a place where readers can debate and engage with the most important local issues. The ability to comment on our stories is a privilege, not a right, however, and that privilege may be withdrawn if it is abused or misused.
Please report any comments that break our rules.
Read the rules here