Celtic must embrace volatility and risk a more purified version of Angeball

Both clubs have experienced significant changes over the period, so we will be slicing and dicing things in various ways in order to attempt to gain insight.

First, here are two custom radars I maintain and run for the full season to date:

A quick reminder: OBV is designed to measure the value of playing actions and is distinct from xG models. For example, shot OBV will be reported as positive even over stretches in which teams significantly outperform their xG, which may suggest that performance level is more likely to mean revert.

From an xG perspective, Celtic have outperformed across the various ways of slicing and dicing. However, Rangers have had a slight advantage in overall OBV difference, with them benefiting from relatively poor keeper play from opponents.

But looking at things over the full season so far leaves out two material developments, which leave the robustness of comparative analysis less than ideal. Celtic’s squad composition changed materially at the end of the summer transfer window, and Rangers had a new management team arrive at the end of November. The following are the same radars but comparing the two periods for each team.

There are obvious issues with this comparison; the samples do not align relative to number of games, opponents faced, etc. However, the comparison offers some value contextually, as Rangers’ recent stretch of wins has been accompanied by excellent performances by the keeper. Another standout over the seven games for Rangers was the relatively low amount of passing OBV conceded, which drove this inquiry:

Those radars compare each team to itself pre and post changes, using defensive and pressing-related metrics.

Similarities with themselves and each other are evident across periods. For example, both teams have ratcheted down pressing actions but ramped up counter-pressing at the same time. One metric which jumped out was the defensive distance for Rangers, which suggests they may have advanced their lines on average.

Metrics in the prior radar comparing chance creation and shots also reflected this, as Rangers’ shots from the high press have jumped considerably, which also correlates with them creating more clear shots (when a player shoot one-v-one with the opposing keeper).

READ MORE: Will Celtic's new signings banish the ghost of Ryan Christie?

Otherwise, attacking metrics are relatively similar. Rangers have also become far stingier relative to conceding shots from counter-attacks since the managerial change, averaging just 0.14 per game.

For Celtic, the shifts have been more nuanced, but then that was arguably to be expected given the shift has been in personnel rather than in manager.

Referring back to my Jenga analogy, the January transfer window has already brought with it three ‘better-shaped’ pieces for the Ange Postecoglou tower.

For example, go back to the defensive and pressing radar and compare the relative pressure and counter-pressure regains for Celtic versus Rangers. Celtic have been less efficient while deploying Postecoglou’s pressing system, as the frequency of regains versus pressing actions has been at a lower rate.

That probably is not all that surprising given that the two primary pressing midfielders have been David Turnbull and  Tom Rogic over most of the period. Even Jota and Liel Abada have been functional, but not high-end, in that phase of play with Kyogo having been left on a sort of pressing ‘island'.

The introduction of Daizen Maeda and Reo Hatate, in particular, may increase those efficiency levels for Celtic. That is to say it is fair to expect the Hoops to regain possession more frequently relative to pressing actions.

Teams tend to be more vulnerable to conceding higher quality chances in periods of transition, and particularly in their own half, so this uptick could also further improve Celtic’s quality of chance creation.

Both teams have been moving in directions where they are likely to get more out of their playing squads, with Celtic having materially improved depth relative to the fit of players with Postecoglou’s preferred playing style.

Conversely, Rangers’ new management team appear to have improved the alignment of player selection with pressing-related tactics, in particular.

So how is all this likely to play out in the coming months? It appears to be two very evenly matched teams that are both gearing up for a title race which may go down to the wire.

As covered in prior columns, Celtic’s risk profile is likely higher due to relative squad depth, potentially persistent injury issues and the inherent volatility of a more purified version of ‘Angeball.’

However, being the club six points behind at this stage, embracing risks with commensurate potential upside may be just what is required.