One of the aspects of conducting analysis publicly is the accountability which accompanies doing so. For example, a relatively flippant Twitter thread I conducted at the time of Liel Abada’s signing ended up with egg on my face, as it expressed my opinion that his profile should not have made it through a robust player recruitment screening process.
With a few exceptions, I think the profiling exercise (which I developed after the Abada clunker) conducted has done a pretty decent job, while using very flawed Wyscout data, at analyzing player signings. Two of Celtic’s mainstays in midfield this season were subjects back in September, with record signing Arne Engels and Luke McCowan both put through the exercise.
We now have a decent sample size of performance data at Celtic with which to gauge how their profiles, and my analysis, have held up. To summarize those analyses, I was concerned about the lack of evidence for Engels relative to be a good strategic fit for Celtic’s style of play and his likely role within it (i.e. attacking central midfielder), while optimistic on McCowan’s profile if deployed as a valuable squad member, while remaining skeptical at a European level.
The first quantitative exercise I conducted for this column was to revisit the degree to which team and associated competitive dynamics impact player performance levels. This remains a fundamental issue within football analytics, in my opinion - and one that I believe is central to how someone like Tony Bloom thrives by exploiting.
For example, across more than 130 statistical metrics I reviewed for this exercise, Engels’ and McCowan’s data displayed a simple correlation of over 98%.
Keep reading with a 7-day free trial
Subscribe to The Huddle Breakdown to keep reading this post and get 7 days of free access to the full post archives.