It’s a bit scary, isn’t it?

Expectations rise after a performance like that against RB Leipzig, a comprehensive dismantling of a tier-one club in Europe’s premier club competition, don’t they? The seeds of a tactical plan glimpsed on the hybrid playing fields of obscure American universities wrought large under the Celtic Park disco lights.

Can the Celtic of Greg Taylor, Liam Scales, Daizen Maeda and Yang Hyun-jun be capable of such feats regularly? That now becomes the question.  Time will be the provider of answers and in the meantime, we should enjoy but pause to reflect on where such a result fits into the recent context of Celtic’s European history.

My detailed database started in the 2014-15 season, which I accept is a little arbitrary as a starting point. I think we can all agree the Martin O’Neill era were ‘different days’ in terms of the relative wages and therefore player quality within reach of Celtic.  And practically, as Celtic drew back from Rangers’ Employee Benefit Trust-induced arms race in the Gordon Strachan and Neil Lennon eras, we simply don’t have the data!

So, arbitrary or not, that’s our starting point.

Top five

In the last 10 seasons, Celtic have had 37 matches against teams from the ‘top five’ leagues. During recent history, those leagues have mainly been:

  • England
  • France
  • Germany
  • Italy
  • Spain

16 of those ties have been in the Europa League and 21 in the Champions League. Celtic’s overall record is (if there are young children watching, you may want to avert their eyes):

Those six wins have been home and away against Lazio (2019), home to RB Leipzig in 2018 and 2024, home to Stade Rennes (2019) and home to Real Betis (2021). Slim pickings. Given that RB Leipzig was second in the Bundesliga at Tuesday’s match, arguably that was the most impressive win.

Notable draws have been achieved home and away to Pep Guardiola’s Manchester City in 2016, home to Internazionale in 2015, home to Athletico Madrid in 2023 and the away draw in Bergamo this season. Brendan Rodgers has been at the helm for both wins versus RB Leipzig and four of the five notable draws. Or, of Celtic’s best 11 results, he’s been the gaffer for six of them.

You know where we are going next! Yes, performances rather than outcomes.

Performances

Celtic generated 2.25 xG according to my model versus RB Leipzig – and I need to use this for comparative reasons as I do not have e.g. the StatsBomb data or Opta data going back that far. As a rule, my model is usually similar in differential to the StatsBomb model.

This is only the sixth time in 37 ties Celtic managed to generate more than two xG. To give an indication of how brutally difficult this level of opponent is, in amongst those six ties, Celtic have lost 0-4, 1-5 and 1-7. Blimey. Indeed, only the 3-2 home win against Real Betis Balompie, where both sides were under strength, resulted in a win apart from Tuesday. If we look at xG difference, the story becomes remarkable. In 37 ties, Celtic have ‘won’ the xG difference battle on 11 occasions. Wins from that situation? One - against RB Leipzig on Tuesday.

We talk about the danger zone in xG difference on a single-game basis being between +1xG and -1xG. In between those values, there is sufficient room for the vagaries of football variance (luck) to have a greater influence than chance creation quality.

If you want to reduce the impact of such variance significantly, you want to get to above 1.5 xG difference. Celtic’s highest positive xG difference against a top-tier club was 1.45 away to Internazionale in 2015 – they lost 0-1. Their difference of 1.07 against RB Leipzig was the next highest and represents the only time Celtic has triumphed over a tier-one club when winning the xG battle.

The other key performance indicator is packing data. That is the measure of players taken out of the game from passes, dribbles, turnovers, and recoveries. The sharp end of the packing data set is measuring how many defenders you take out of the game. Once you have the opposition defence wrong side of the ball, you are causing mayhem.

On average in these tier-one ties, Celtic have 26 defenders packed during a match and pack 15 opponents. Against RB Leipzig, Celtic packed a record (amongst this sample) 27 Saxon defenders whilst only allowing 13 of their defenders to be put ‘wrong side’ of the ball. Both are records in this data set.

Celtic achieved an element of game control against RB Leipzig. Rodgers likes to defend with the ball, handy against more talented sides. Celtic’s 612 completed passes are second only to 734 when losing 0-2 at home to Valencia in 2019. It was only the fourth time they had achieved 60 per cent possession in these matches.

24 touches in the opposition box equalled the tally at home to Stade Rennes in 2019 as the highest. 17 total shots equalled the total achieved at home to Bayer Leverkusen in 2021 when goalkeeper Lukas Hradecky made six saves in a remarkable display. A match where Celtic generated three ‘Big Chances’ to match the tally the other night.

Summary

Celtic have put together two back-to-back performances of note against high-level opponents at the Champions League level. They managed to achieve four points as a result.

The degree to which variance can impact such matches remains high given the quality of the opposition and the low likelihood of Celtic generating the xG differentials needed to mitigate the variance. Therefore, a consistent run of positive outcomes against this level of opponent remains a low probability overall. However, the beauty of the new Champions League format is that Celtic does not now face another tier-one club until the final fixture in Birmingham versus Aston Villa.

If they can maintain the performance levels seen on Tuesday, positive outcomes against Club Brugge, Dinamo GNK and Young Boys are possible. Our expectations have been raised, and Rodgers and the team deserve that extra pressure.