Celtic must address pressing problem - James Dailey

Why include the 2019-20 iteration of Celtic? That team was the product of a coherent transfer window, including significant investment that resulted in several first-team starting upgrades, and the team’s domestic performance levels were high up until the point which the league was ended due to the pandemic.

First, let’s take a look at broad performance metrics from Saturday and compare them to last season’s averages:

Given the disappointing league campaign last season, Celtic underperforming across the various xG metrics is instructive. In particular, conceding 1.28 in xG warrants further examination. As this is broken down further, we can see that both open play and indirect free kick were poor, while corner was at least in a similar range. The average xG per shot conceded was high.

Possession was high, however, Celtic had difficulty in converting it into quality chances. The xG per shot was poor, as 7 of 19 shots taken were of poor quality. Just 2 of the 7 were on target, with both having post-shot xG of just 0.01, which suggests they had around a 1% chance of resulting in a goal.

Finishing was an issue as well. For example, Celtic’s highest quality shot by James Forrest in the 93rd minute had xG of 0.41 but post-shot xG of 0.31, which suggests Gordon’s save attempt was far more likely than not to be successful.

READ MORE: Leggy Celtic's physical conditioning is another concern needing immediate action - Alison McConnell

The passes per defensive action (PPDA) and aggression metrics suggest that Celtic’s pressing intensity was marginally higher than last season’s average, and the directness metric reflects how much more deliberate the team was in possession and build-up.

Next, we look at defending metrics and how they compare to last season and also 2019-2020 levels:

With the broader metrics indicating high pressing intensity, what jumped out to me with the defending metrics was how ineffective that pressing was. This was evident in the defensive distance metric, which measured how close to Celtic’s own goal line the average defensive action took place - nearly 10 metres closer than the average of last season.

Also, pressure and counterpressure regains were significantly below the averages of both seasons. I believe this was at the root of Celtic’s issues in the game, as successful regains often lead to the highest quality chances as defences tend to be more vulnerable in transition. For example, Starfelt’s lost one-v-one occurred in transition and was key to Hearts’ first goal.

I think we can think of this dynamic using a risk/reward framework. High pressing intensity creates defensive vulnerability if unsuccessful, and Celtic’s was not very successful on Saturday. This resulted in Hearts being able to generate well above average chance creation from open play, which was reflected in the broad metric radar above.

Now let’s take a look at attacking metrics:

Here again, we can see the ineffectiveness of the Celtic press reflected in the high press shots metric, as well as average shot distance and clear shots, in these radars. Along with the directness metric, pace towards goal also indicated how deliberate Celtic’s possession was. From a style perspective, the lower box cross percentage value showed that crosses represented a lower proportion of the passes into the penalty area.

Lastly, Celtic’s decline in volume and effectiveness in the dribbles metrics are notable given Abada’s selection and the fact his performance record in Israel was poor for a winger/wide forward. On Saturday he was successful in just 3 of 7, which compared to Forrest successful in 5 of 6.

Many will understandably point to Celtic’s continued issues with defending set-pieces, which was reflected above in the broad metric radar. However, I am concerned with the degree of ineffectiveness in Celtic’s press. The issue calls into question team selection, as Ryan Christie has been Celtic’s best player in pressing metrics in the past few seasons.

This is likely to be a vital aspect of Celtic’s play given Postecoglou’s style and must improve in the coming weeks. Decision-making in the final third and finishing has certainly been an issue, but I believe the relative quality and productivity of Celtic’s press is a far bigger issue if not addressed.