Just over one year ago, this column examined the concept of performance volatility and the potential implications of the ‘Ange-ball’ playing style through that lens.
The subsequent period offered a practical realisation of these issues, which surprised me. Domestically, Postecoglou's style has manifested in squashing opposition chance creation to a severe and consistent degree.
At the same time, many of the issues I laid out in that August 2021 column have been equally as severe and consistent versus higher-quality European opponents.
Let's begin with the good news.
This first graphic shows the rolling 11-game average non-penalty xG difference and goal difference for the three domestic league seasons prior to Ange Postecoglou’s arrival at Celtic:
It shows the degree to which the 2020-2021 season was a drop off in both underlying net chance creation, but also goal difference.
In contrast, here is the Postecoglou era to date:
The absolute level of non-penalty xG difference has been very high and trended towards modest improvement. Goal difference has been impacted by the huge 9-0 result versus Dundee United and, while unlikely to persist, is also trending positively.
If we breakdown the non-penalty xG difference into its two components, we see some additional context:
The non-penalty xG and non-penalty xG conceded are shown using a shorter six-game rolling average in this graphic.
I selected a shorter periodicity to try to gauge volatility more and a couple of issues stand out.
Chance creation has been relatively high and consistent but opposition chance creation has been extremely low. In fact, I am not sure it can get much lower.
For example, over the 44 league games under Postecoglou to date, opponents have averaged just 0.32 xG from open play on just 4.68 shots from open play. To put it bluntly, domestic opposition has been wholly unable to cope with 'Ange-ball'.
Many of the risks seemingly inherent in the style of play have remained under wraps versus league opponents. Unfortunately, the polar opposite has transpired against higher-quality European opponents.
READ MORE: Why Hoops are perfectly placed to disrupt 'dangerous' Shakhtar Donetsk in Warsaw
What is meant by ‘higher-quality’ opponents? I have defined that as those teams with rankings by objective models such as various ELO models and 538’s Soccer Power Index which are relatively close to or better than Celtic’s over Postecoglou's tenure.
This results in the following opponents: AZ Alkmaar, Real Betis, Bayer Leverkusen and Real Madrid.
Opponents such as Bodo/Glimt and FC Midtjylland fall a level or two below that group. Shakhtar Donetsk fall in a sort of ‘in-between’ level between those and AZ Alkmaar, while RB Leipzig are among the higher tier.
Those four opponents offer a sample of seven games in the Postecoglou era. Here were the non-penalty xG conceded in those games listed in chronological order: 1.04, 2.45, 2.37, 3.14, 2.24, 2.76, 2.51.
The first of the games was the 2-0 victory at home versus AZ last August and was the only match in which Celtic did a reasonable job limiting opposition chance creation. The seven-game average works out to about 2.36.
To place that level into some context, last season’s Champions League group stage had only two teams concede more non-penalty xG, with Besiktas at 2.68 and Club Brugge at 2.44. Unfortunately for those sides, goals conceded were over three per game.
Yes, there were some positives to take away from the performance versus Real Madrid last week - particularly in the first 50 minutes - but ultimately Celtic conceded 2.51 in non-penalty xG (and, obviously, the three actual goals).
At Champions League level, conceding that volume and quality of chances can result in scorelines like those suffered versus Paris Saint-Germain and Barcelona under Brendan Rodgers.
Celtic have done a good job in creating chances against this higher echelon of opponents, with a seven-game average of 1.67 in non-penalty xG.
However, given the relative deficit in chance creation, some combination of superior finishing and/or goalkeeper shot-stopping is likely needed in order to extract results with consistency at the level.
So far during his tenure Joe Hart has been a competent but unspectacular shot-stopper and finishing is infamously volatile in small sample sizes.
Shakhtar may be the ‘swing’ opponent this season, as their level of quality is likely below Celtic and even arguably AZ.
More controlled defensive performances closer to those enjoyed versus Ferencvaros last season would offer some hope that the risks seemingly inherent to 'Ange-ball' may be at least somewhat manageable at this higher level.
Barring that, a consolation is that there is little evidence of the domestic landscape shifting as long as Postecoglou's style remains at Celtic.
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