It seems like Celtic's captain Callum McGregor may be missing until after the next international break due to a calf/Achilles problem. That covers at least two matches, one each in the Scottish Cup and the SPFL. Of course, the injury break may be longer.

Based on the recent selections, the replacement for McGregor would appear to be either Tomoki Iwata or Paulo Bernardo. Young Odin Thiago Holm has rarely featured whilst it would be highly surprising to expect even younger Daniel Kelly to fill that void. It remains unknown how long Reo Hatate will be absent.

Bernardo or Iwata, then? Bear in mind that this is a mixture of StatsBomb data (SPFL only) and my own (all competitions). Iwata has only 549 total minutes, so a health warning is attached to his data as regards the small sample size.

Defending

We’ll start with the central defender framework of defensive action success rate and possessions won from defensive actions. An important consideration here is we remove passing.

Iwata is something of a defensive monster using these metrics. McGregor’s numbers are a little below what Scott Brown used to post. However, with Iwata, that profile is more akin to a central defender. It is slightly surprising Bernardo lags when we consider the volume of defensive activity each is involved in.

Bernardo has been involved in around 50 per cent more defensive activity than McGregor despite playing a more forward role.  And that is a key consideration as well. Bernardo has been used often as the most advanced of the midfield three whilst McGregor and Iwata have generally been the most disciplined position-wise. We can see this when we consider the StatsBomb pressing data.

As with Iwata’s defensive action success, Bernardo is a pressing monster. He rivals Daizen Maeda and Matt O’Riley in this discipline.

The sheer volume of pressing is heightened by his data disproportionately being drawn from Champions League games. Which will serve to dampen down most of his numbers compared to those whose workload has mainly involved combatting SPFL-level opposition. Bernardo also has commensurate regain rates associated with his much higher press and counter-pressing volumes. It is probably his ‘super skill’. Iwata has higher volumes than McGregor as we may expect from both a younger man and one more defensive in playing history.

Overall, you’d want Iwata as a more defensive, traditional ‘six’, whilst Bernardo offers some pressing excellence and sheer defensive endeavour.

Ball progression

In which we compare ball carries and deep progressions from StatsBomb.

Carries are simply controlling the ball with the feet whilst standing or moving, so this is more about control. Deep progressions are instances of moving the ball into the opposition's defensive third, so this is about progressing the team in attack.

This is very much McGregor’s ‘super skill’. His technical control and ability to find and manipulate space as his trademark strengths.

Bernardo’s numbers by this view would be depressed by his more advanced position. However, my feeling is both he and Iwata are less adept at finding space in which to receive the ball. McGregor has 71 on-ball actions per game compared to 57 from Iwata and only 37 from Bernardo.

Creativity

Firstly, let’s look at deeper build-up as evidenced by the volume of pack passes (forward passes that take opponents out of the game) and the expected assist value generated from secondary assisting passes (the pass before the one that creates a chance).

Again, McGregor is difficult to replace given his natural creative tendencies.

Whilst Bernardo does provide a reasonable quality of passes before chances are created, the volume of his pack passing is eye-catchingly low. I’d go so far as to say that his lack of forward passing is on the margin perhaps the lowest of any attacking midfielder I have ever seen. It is a real standout feature of his play and would on its own be enough for me to question a permanent signing.

With Iwata you get a reasonable volume of forward passes with minimal involvement in actual chance creation opportunities. In terms of direct chance creation, I’ll compare the volume of chances created with the quality as expressed by the expected assist value.

Here Bernardo’s numbers are more akin to the attacking role that he plays but bear in mind he is not reaching the volume of McGregor whose role is very much deeper.

Iwata does not provide the creative threat of the other two players and so with him in the side, you are sacrificing some attacking productivity.

Goal threat

Arguably this is the least important aspect of play for the role of replacing McGregor given it is a 'nice to have' rather than core capability of the holding midfielder.

Bernardo does carry a goal threat as was evidenced in his data from Portuguese international age-grade football. Again, his more advanced positioning influences these numbers. McGregor’s goal threat has been steadily declining with age whilst Iwata offers minimal threat in this regard.

On-Ball value

Finally, let’s consider the StatsBomb composite on-ball value metric that considers the impact of all on-the-ball actions for each player.

Oh boy, does StatsBomb hate poor Bernardo? His OBV of -0.22 puts him in the bottom percentile of all players in the league. This is mostly driven by his -0.2 shot OBV. Despite having decent expected goals and volume of shots, their model hates his shot selection and execution. He also has a negative pass OBV reflecting his negative (i.e. backwards) passing.

McGregor’s overall OBV of 0.21 puts him in the 67th percentile of all players in the league and is mainly driven by his ball progression. Iwata comes out at 0 OBV, in the middle of the three, which places him in the 6th percentile in the league. They like his ball carrying less so his defensive actions, surprisingly.

Summary

Replacing McGregor is a thankless task and as a like-for-like fruitless, and his excellent ball progression is not directly replaced by either candidate.

We have a choice between a traditional defensive-minded holding midfielder who will offer little in the attacking third. And a player with excellent pressing abilities and some goal threat but very little in the way of progressing the ball.An important consideration is also what the impact would be on this season's standout performer O’Riley. Selecting Bernardo probably means O’Riley must play a more defensive and disciplined role. Whilst picking Iwata should allow O’Riley to supplement the attack as he has done to great effect.  

Given Celtic’s weaknesses this season are in giving up too many chances compared to their rivals for the title I would be minded going with the more defensively solid Iwata and allowing O’Riley to do his thing further forwards.

Bernardo remains a bench option when chasing a goal and hounding the opposition with pressing excellent late in the game.