Packing a Pass
Measuring effectiveness of Ante Palaversa and Leighton Clarkson
Aberdeen will face off against Ross County this weekend, and one of the main topics will be the central midfield area. Last week, we saw an incisive attacking performance with Leighton Clarkson at the heart, pulling the strings.
Two weeks prior the Dons had a lot of the ball against St. Johnstone, completing more passes than they did against Motherwell, but they failed to create any clear-cut chances and managed only two shots on target.
What was the cause of this?
Partly, the Dons performed much better on set-pieces against Motherwell, scoring twice from corners. Additionally, the team looked much more dangerous with the ball, passes seemed to break the lines as opposed to circulating in front of them.
I wanted to take a look at a different metric to try and compare the performances of Palaversa and Clarkson, principal playmakers against St. Johnstone and Motherwell respectively, and see if this could offer us any additional insight but first let’s look at what we can ascertain via FotMob.
Freely Available Data
As you will all be aware, I reference a lot of the publicly available data we can access for free. Sure there are flaws but compared to what we used to get, it’s remarkable to be able to access so much information in real-time in games.
If we compare the two performances, there’s not a lot to choose between the two in regards to passing stats:
Clarkson shades it with pass accuracy and certainly wins on accurate long balls. Palaversa has a higher expected assists which I think would have been generated from corners, so maybe Clarkson is higher in open play? Passes into the final third are neck and neck.
But we all watched the game, and Clarkson seemed to make things happen a bit more than Palaversa did vs. Saints.
Introducing Packing
I’m indebted to Alan Morrison for introducing me to this concept and helping me understand it in his recent podcast episode.
Packing is a metric developed by the German company Impect, an organisation set up by two former players - Stefan Reinartz and Jens Hegeler. It measures the amount of opponents taken out of the game by a pass or a dribble, giving value to those passes that progress the ball. See the graphic below.
The pass from player 1 to player 2 is not a pack pass as it does not bypass any opponents. The pass from player 1 to player 3 is a pack pass as it bypasses two opposing players.
This little video summarises the idea pretty well too.
I’d also recommend this great article from Leicester analyst Dominic Wells which explains the concept in greater detail.
I used a positional modifier to measure. This is shown in both the video and the article and gives a greater value to bypassing a player in the defensive or midfield line. So passing the ball past a forward is worth 1, a midfielder is 2 and a defender is 3. See an example from the weekend to illustrate.
Notes on the Data
As this was my first time and I wanted to do a comparison, I only had time for 45 minutes. So it's just the first half of both games. I also only have these two sets of data as a reference point so there’s not a whole lot of context that can be applied.
Nevertheless, I think it’s interesting to look at and it’s something I know I’ll be keeping an eye on moving forward, even just to mention in an article.
The major conclusion that can be taken from this data is that Clarkson’s passing was more focused on playing forward with over half his passing attempts resulting in a positive pack pass. Palaversa, also scores high on his packing numbers but clearly there are more of his passes circulating the ball as opposed to penetrating. I think this supports what the eyes saw during the game.
It would be dangerous to make sweeping conclusions from this small review. We could attribute it to Clarkson being a much better passer with greater vision. We could give the credit to the different personnel that played in this game versus St. Johnstone. Or, it could all be down to Motherwell’s setup compared to Saints.
And this is just one game. Palaversa may score similar numbers to Clarkson in other matches and vice versa.
Palaversa Pack Passes
D = Defensive half, A = Attacking half
Palaversa’s top target was MacKenzie (5) followed by Keskinen (3). A lot of those passes were wide passes to the left.
Clarkson Passes
Clarkson’s number one target was Gueye (6). In theory, he was playing centrally but he did drift to the inside left channel.
It’s notable that both players focused a lot of their passes to the left side. It does feel that most of Aberdeen’s build-up play, at least in the first half of matches, is concentrated there.
I included the half of the pass and the reception. Clarkson made a lot more passes that crossed from one half to the other. Does this tell us anything? I’m not totally sure. Maybe that he was more direct with his passing than Palaversa?
What’s Missing?
A fundamental component that’s not covered is the player receiving the ball. I believe Impect categorise this in their software. For example, if Clarkson played a long ball to Gueye and he got the first touch on it but then lost possession due to being tightly marked and/or the pass being hit too hard etc. then Clarkson would not get full credit for that. In my study, this was not considered.
Additionally, I did not include dribbling or ball-carrying. Clarkson had a few good runs with the ball forward vs. Motherwell. Finally, there’s no note on distance. This might be good to see with some passing maps.
Ross County
On to more ‘pressing’ matters. It will be interesting to see how County approach this game out of possession. Here’s an excerpt from the game recap back in September.
The Staggies could opt for a similar approach. The difference this time is that Dorrington and Knoesters are both better on the ball than Rubezic and Molloy. Not just in terms of passing but in ball carrying too. We could also see Clarkson drop into the back three again like he did against Motherwell, and the two centre-backs go wider. In that memorable match, Nilsen did that as the game wore on.
What County did well was deny space once the ball progressed into their half. If the Dons can pull them apart and similarly break the lines to Saturday, it would be very encouraging for next season. I say next season, as this is likely to be the last game for a while (maybe Tannadice) where the Dons would expect to have most of the ball and be tasked with breaking down a deep defence.
Enjoy the game. I’ll be keeping an eye on Clarkson and Palaversa (assuming it’s not a Shinnie/Polvara midfield). My post-game thoughts will be on the P&J/Evening Express website on Tuesday evening.
COYR