An advanced defensive data personal library

A mini library of interesting advanced research around defensive football data

Last week I sent out a defensive stats personal library. I got a few suggestions for further pieces to add to it, but generally found myself wanting to separate them out into a separate space, for a couple of reasons.

The first is that all the work in this post was written or presented in 2020 or later, and a big motivating factor for the first 'library' post was bringing some old(er) work back into the light. The second is that there's also a fairly clear difference in the type of work and type of presentation here, which is why I've titled this 'advanced defensive data'. I want people to do good work regardless, but in my mind these two lists are like baking and barbecuing. Technically both cooking, but very different.

Like with the other library post, this isn't a definitive and entirely-complete list, more a travel-ready pack of ideas.

'The right place at the right time: Advanced off-ball metrics for exploiting an opponent's spatial weaknesses in soccer', Sergio Llana, Pau Madrero, Javier Fernández (2020)

The paper has more than just defending, but the defensive element is similar conceptually to PATCH, on the other list. Players have zones of responsibility assigned using tracking data and stuff that goes behind them in that zone is a red mark against their name.

'Measuring and modelling defensive efficiency with only event data', Abhishek Amol Mishra and Soumyajit Bose (2022)

A bunch of interesting ideas and a reference to similar work in this Opta Pro Forum presentation. The approach is basically to take the sequence of events that has led up to a defensive action to predict the quality of chance it may have led to.

'Player Chemistry: Striving for a perfectly balance soccer team', Lotte Bransen and Jan Van Haaren (2020)

Looks at both sides of the possession coin but as applied to defending this paper uses two concepts that I imagine people might find intriguing. One is matching defenders up against a direct opponent and attributing a better/worse on-ball performance than usual to that defender. The second is joint defensive impact; if defending is less of an individualised endeavour than attacking, could looking at pairs or units of players be a useful avenue of research?

'Pressing Times: Can data tell us when & how to navigate out of a counter press?', Gerald Lim, Ashley See, and Zhi Yuan Chua (2022)

Although framed as looking at how to get away from counterpresses, this Opta Pro Forum presentation is interesting in that it looks at different types of counterpress using tracking data.

Please get in touch with any pieces that may make good additions. You can do so on Twitter (@get_goalside), Mastodon (, or email (

Get Goalside is just about the only place you can get this kind of work. If you like it, you can become an official supporter for just £2 a month.