In the geeky world of data analysis and football stats gathering, it is not often that I am genuinely taken aback.

But it is becoming more regular. Firstly, there are/were enormous disconnects between ‘what the data says’ and general supporter sentiment over Carl Starfelt and then Reo Hatate. Lately, there have been a quite viscerally negative reaction to the new summer signing Auston Trusty after he’d made a couple of appearances.

Don’t believe me? Check out the direct quotes from Celtic supporters on “X” as I laid them out in this piece on the Huddle Breakdown. By the way, now you are here, please subscribe to the new Huddle Breakdown channel! That ‘further analysis’ teed up at the end is appropriate given he has passed the 900-minute threshold in all competitions - what else are we going to do in the latest in a long line of international breaks?

Defensive actions

The bedrock of my analysis framework for central defenders is to strip out everything but defensive actions. I measure two slightly related and highly aggregated metrics. Defensive action success rate is the percentage of defensive actions that were successful e.g. an aerial duel won, a tackle won, or a successful interception.

Possession won from defensive actions is the percentage of all defensive actions that result in Celtic gaining or maintaining possession. Over time this view tallies well with the eye test.

As the data set grows it gets more cluttered. This time I have stripped out many players and left in Jack Hendry (what ‘bad’ looks like), Christopher Jullien (what ‘good’ looks like) and Dedryck Boyata (what ‘average’ looks like). I have included all the centre-backs from the recent Ange Postecoglou and Brendan Rodgers periods as the current team play a similar high line, high press and possession-based style of football.

The good news is that Celtic’s benchmark for what “good” looks like seems to be rising. Whilst possession won seems to have peaked with early Cameron Carter-Vickers, defensive action success averages are rising.

All three current centre-backs are performing well in this comparison. Liam Scales wins more of his defensive duels, but they do not necessarily result in possession to Celtic, as I have highlighted before. Overall though I’d say he has been an upgrade on Starfelt, whose better second season placed him firmly in Boyata's ‘average’ territory by this view.

Whilst Trusty can improve on his possession retention dates, his defensive action success is as good as we have seen from ten years of Celtic centre-backs. An encouraging start.

Ball progression

In recent Celtic teams, it isn’t all blood and snotters defending. The majority of on-ball actions are in passing and generally progressing the ball. I measure this by illustrating the number of pack passes each player achieved per 90 minutes, and the number of ball carries.

No one will match Kristoffer Ajer’s ball-carrying ability, I suspect, as he is a clear outlier by that measure.

Pack passing volumes reflect the style of play preferred by the manager as much as any individual ability. The Postecoglou and Rodgers era defenders are far more likely to rack up forward line-breaking passes than those prior where there was not such an emphasis on playing out from the back.

However, within those populations, there is still differentiation. Scales in particular has taken on board the requirements of his manager to be a prodigious pack passer by the standards of his peer group. And Trusty is not far behind. Carter-Vickers’ numbers are reducing season on season maybe reflecting the ongoing management of his physical state and therefore him taking more of the easier options as regards passing forwards. Despite his ground speed, we have not yet seen a high volume of ball carries from Trusty.

If we consider the impact of those passes as expressed by the pack passing score:

Again, we see the more recent centre-backs at the top and the pre-Postecoglou ones at the bottom. I suspect Shane Duffy and Josip Simunovic were never going to be ball-playing centre-backs… By this measure, we also see that Scales has been an upgrade on Starfelt. Trusty has slotted in as the second most effective pack passer after Scales.

So far, all good news in terms of alignment with the style of play Rodgers demands as regards his new (and expensive) American recruit.

Peer compare

How do they compare between themselves? Using (newly branded) Hudl StatsBomb data, we can compare Trusty with both Scales and Carter-Vickers. Note that all the data above was for all competitions. Hudl StatsBomb data is focused on SPFL matches only.

Firstly, Trusty versus the player some thought he may replace, Scales.

Both players are 26 years old. According to Hudl StatsBomb, in league play, Scales is more aerially dominant in terms of the volume of duels won, and quality as expressed by the percentage of duels won. They also prefer the Irishman’s carrying the ball out from the back. Also, he affects more clearances than the American.

Trusty scores better overall on Hudl StatsBomb’s proprietary on-ball value model with 0.25 to 0.19. Specifically, he concedes fewer fouls and blocks more shots. Carter-Vickers is a year older than his colleagues.

He is even more aerially dominant as regards outcomes winning 82 per cent of his duels. His volume of aerial battles is like Trusty’s reflecting the extent to which opponents target Celtic’s left side.

Carter-Vickers actions more clearances and, like Scales, is less likely to be dribbled past than Trusty. But again, taking a holistic approach, Hudl StatsBomb rates Trusty’s overall contribution as expressed by OBV as higher with Carter-Vickers scoring 0.2 per 90 minutes.

All in all, similar strengths and weaknesses between Trusty and his two colleagues.

Hudl StatsBomb OBV

We’ll dive into a peer-wide comparison of the main components of on-ball value impacting centre-backs – defensive action OBV and passing OBV.

Data for these metrics is covered back to the 2018-19 season.

 

Focusing on this season’s performances only, both Carter-Vickers and Scales trend towards the middle of the pack by these measures. Whilst Trusty edges towards the virtuous top right-hand corner.

Stephen Welsh does particularly well by these metrics which tend to focus on the xG impact of actions as opposed to the merits of each action in isolation. This is a logical approach accepting that the objective of football is to score goals at one end and prevent them at the other. My issue is the correlation between this view and the eye test, as exemplified by the defensive action focus of the first view.

I would point to Filip Benkovic 2019-20 as further evidence for that disconnect. But these are well-thought-out metrics devised by folk far cleverer than I. So, it all adds to the mix. Back to our subject. This is another view by which Trusty is viewed favourably by the data compared to his peers.

Summary

How to summarise? Context, context, context.

Trusty is not a young player (26) and has just come off a difficult season as a member of the worst defence ever to play in the English Premier League (Sheffield United last season). He has been asked to play on the right side of the centre-back pairing whilst previously being part of a back three and often deployed as a left-sided wing-back. 776 out of 995 minutes has been as a right-sided centre-back at Celtic.

His partners have alternated between Scales and Carter-Vickers across the 12 matches he has participated in. His appearances include Celtic’s three most difficult Champions League ties plus five minutes versus Slovan Bratislava. Seven of his 11 full matches have been away from home and include a League Cup semi-final.

It is therefore slightly remarkable that his numbers so far after a small sample of games are so promising. Already in line with the better centre-backs in terms of defensive actions, he is only behind Scales in his ability to progress the ball.

He is profiling very similarly to what we all agree is an improving Scales and a consistent Carter-Vickers according to Hudl StatsBomb in league play. By their OBV model, he is a standout relative to Celtic centre-backs historically.

This is a highly promising start to his Celtic career.  Will Carter-Vickers continue to need careful time management, or will he eventually supplant the admirable Scales? File under good problems to have.