Dummy events

With no football to entertain us, football analytics nerds got busier than ever in this quarantine period: Metrica Sports released high quality full match tracking data, David Sumpter and his Friends of Tracking have been releasing incredible work every week, and Seattle Sounders organized an online conference with some of the biggest names in the analytics world (and also the COVID “death charts guy”).

But even with increasing availability of tracking data, the most valuable work in football analytics is still — and will continue in the foreseeable future —being done with events.

Therefore, one of the most important outputs of tracking data analysis is to improve the definition and richness of events.

Indeed, codifying and recording events is hard work, and sometime not even that objective.

Take this example: Alaba’s pre-assist for Bayern Munich’s first goal against Eintracht Frankfurt. This was the event:

Seems readable: Alaba saw Muller’s movement and made a pass to the back of the defender.

However, that is not what happened.

You can see that the pass was made to Perisic, who threw off the defender with a dummy.

Analyzing this event with tracking data allows us to get the full picture. First, considering location and velocity of each player, there was no space behind the defender:

That is until the ball gets near Perisic, who is marked by the defender and lets the ball go to Muller faking a reception:

So, how would you codify this play? Let us know in the poll.