What You Could Have Won

Some Celtic numbers to alleviate the ever-lengthening nights as we wait for international football to be over and the real stuff to start again!

I look at the creative aspect of the game.

For all the categories analysed, I have taken the cut off at Roberts who has played the equivalent of 4.19 90-minute matches. So, I have excluded the data for Ajer, Dembele, Miller, Benyu, Sviatchenko, Edouard, Boyata and Kouassi.

The data is for ALL competitions. Celtic have played 18 matches so far.

Passing

Pass Completion %

The more creative players are at the lower end of the scale – they take more risks. At the top end of the scale we should expect to see the centre backs and defensive midfielders. And McGregor.

Pack Passing

I’ve introduced the packing concept a few times now, but if you are still unsure, please read Concerning “Packing”. We are getting into the territory now of players who unlock the defence by taking out opposition players with their forward passing. The higher the Pack Pass rate, the more opponents taken out per 90m. The higher the Pass Impect score, the more defenders and midfielders taken out.

Forward players are unlikely to score high as they are ahead of their opponents and few passes will take out many opposition players. Players who play conservative passes will also score low. The high pack passers are the defenders and midfielders who try the higher risk passes through the lines of the opposition.

Chances Created

This represents the sum of all passes leading to shots at goal plus assists. Who is providing the scoring opportunities?

Despite concerns over his form, Armstrong remains the player with the highest chance creation per 90m. Note that Hayes is second on 2.91. Either he is providing poor quality chances, or has been very unlucky his service has not been converted by the shooter. Tierney’s attacking role is highlighted as he provides over 2 chances per 90m

 

Assists

Finally, the rubber hits the road! Who is setting up the goals?

The assist league table:

4 assists in Champions League is impressive from Griffiths. Increasingly on set piece duties, he nevertheless is a consistent creative force.

Note that despite leading the way in chances created, Armstrong and Hayes have only 1 assist each. On such statistics can perceptions of form, err, form. Had the receivers finished some of those chances, our perceptions of their respective form may be different.

Roberts on 2 already. Just sayin’.

Of course, regular readers will know that this is interesting but misleading, and we should consider the assists achieved per 90m to remove the impact of time on the pitch being a variant.

As we have seen with Armstrong and Hayes, you don’t always get what you deserve. Horse to water and all that. And we have a metric for that! Expected Assists (xA) tells us about the speedboat you could have won had the striker not orbited the moon with his attempt at goal.

Lastly, a comparison between actual assists per 90m and xA per 90m. The ones with high positive differences can perhaps expect a regression to the mean. The ones with high negative differences can hope for a change in fortune.