Like each year’s turn of the calendar from winter to spring, July as a Celtic supporter brings forth feelings of renewal and a fresh start to a new season.

Talk of young players and new signings helps fill the daily void prior to the start of competitive games beginning and the blank canvas of the new season offers inspiration for optimism.

As TCW's own Tony Haggerty wrote last week, Mikey Johnston is a player for whom pre-season optimism has become an annual unrequited routine since he made his Premiership debut in May 2017.

As Tony laid out, this season’s renewed optimism includes the arrival of Harry Kewell and today we will examine what areas of Johnston’s game may be most in need of improvement. 

The sample size for Johnston is unfortunately not large, as his introduction to the team came at a relatively young age and injuries have interrupted his development on multiple occasions. His first season of material minutes happened to coincide with the launch of StatsBomb data, so the visuals here include his data for league games since the beginning of the 2018-19 season. 

Celtic Way:

This first radar shows attacking-related metrics over the 22.1 90-minute equivalents Johnston has played in the league over the period. As one may expect playing at Celtic, his output was reasonably high on most of the metrics, though it came with a high volume of possession lost. 

To analyze his performance so far, an important question is what benchmark to select for comparison purposes. This introduces a related issue, which is system versus player attribution relative to performance output. The 2021-22 vintage of Celtic generated materially higher attacking output under Ange Postecoglou compared to other recent sides. To compare Johnston’s output over the period to Jota or Liel Abada last season would introduce this systemic variable.

To try to neutralize that complexity, I decided upon comparing Johnston with James Forrest’s 2018-19 league season, in which Forrest produced one of his best seasons. Here is the same attacking radar:

Celtic Way:

This radar offers hope as to the potential Johnston offers, as his overall attacking profile is comparable to this aspirational benchmark. His relative issues with ball security are once again evident.

Next, we take a look at defending and pressing:

Celtic Way:

Here we see that Johnston is more active defensively overall and also generally more efficient. However, it once again comes with a ‘cost’: a relatively high rate of fouling. 

The following OBV-related radar pulls these two radars together:

Celtic Way:

This radar ‘drops the hammer’ a bit on Johnston, as we see the relative impact of his high rates of fouls, turnovers and dispossessions. To produce near-zero OBV in passing and defending as a Celtic winger playing in the league is simply not good enough, which is a conclusion many supporters likely already have made. 

But what is the case for optimism and is it a reasonable one? If we are trying to ascertain a common thread across Johnston’s performance profile, it would be that his decision-making is resulting in his considerable physical gifts being neutralized.

For example, here is his shot map over the period:

Celtic Way:

This offers one window into the decision-making issue, as an xG per shot of 0.091 is relatively low for a Celtic winger.

The map shows the volume of shots taken from relatively poor areas, including angled from long distances. Each one of those decisions results in a potentially higher quality chance being taken by a team-mate. 

His statistical profile also offers some clues in this regard as well. An open-play xG assisted per key pass from open play calculation offers an average quality of chances a player ‘creates’ for team-mates. Johnston’s comes in at 0.107 compared to Forrest at 0.12, about 11 per cent lower. 

Celtic Way:

Taken in total, the relatively low quality of average chance Johnston has taken and created, along with his significant rates of fouling, turnovers and dispossessions suggests that decision-making should be his main focus. 

Fortunately, cognitive-related development curves can be varied. On average, the human brain continues developing through age 25. Some footballers mature later when it comes to decision-making too. This could be the case with Johnston.

As Tony's article suggested, he is likely to get the coaching he needs to help. Should Johnston achieve the necessary improvements, his profile suggests his upside could bring performance levels comparable to Jota.

After nearly a half-decade of unfulfilled promise and frustration, Johnston likely has it within his own control to make the ultimate good decision: unleashing his considerable talent.