Limit of Voronoi diagrams in soccer analysis, 1, essence

My claim in this article is that "do not rely on Voronoi diagrams unlimitedly in soccer."

I am concerned about criticism to players by using Voronoi diagrams under blind reliance.

As a fact, when 22 players do not move, a Voronoi diagram indicates regions that the players can cover.

On the other hand, when some players move fast, a Voronoi diagram is far from regions that the players can cover.

I think, Voronoi diagrams are not compatible with soccer.

This is because the essence of soccer is in the movement of players.



The reason why Voronoi diagrams are used in soccer analysis is because "likely regions are obtained only from players' positions."



In this article, I roughly explain what the Voronoi diagram is.

In the next article, I will show difficulties in applying Voronoi diagrams when players move, with simple numerical calculation.



The contents of this article are as follows.

• Why are Voronoi diagrams used?
• Questionable scenes.
• Scenes where Voronoi diagrams can be applied, during a game.
• Scenes where Voronoi diagrams can be applied, in our lives.

Analysis of open tracking data, 03, data of DEBS 2013

In this article, introduced is open tracking data of DEBS 2013.

Mutschler et al.(2013) describes this data set precisely.

However, this data set cannot be downloaded from this original site.

One can download this data set from another site described in German. (Fraunhofer Institute for Integrated Circuits)

HEAD and TAIL of the data set.

Analysis of open tracking data, 02, several plottings

On the paper and data, this page。

"Soccer video and player position dataset": S. A. Pettersen, D. Johansen, H. Johansen, V. Berg-Johansen, V. R. Gaddam, A. Mortensen, R. Langseth, C. Griwodz, H. K. Stensland, and P. Halvorsen, in Proceedings of the International Conference on Multimedia Systems (MMSys), Singapore, March 2014, pp. 18-23 

Analysis of open tracking data, 01, Data of Petterson et al. (2014)

How to make the following figure.

Environment, Google Colaboratory (in the following, `colab')

Language, python 3

Tracking data of a player.

This is the reference paper.

"Soccer video and player position dataset": S. A. Pettersen, D. Johansen, H. Johansen, V. Berg-Johansen, V. R. Gaddam, A. Mortensen, R. Langseth, C. Griwodz, H. K. Stensland, and P. Halvorsen, in Proceedings of the International Conference on Multimedia Systems (MMSys), Singapore, March 2014, pp. 18-23 

The data may only be used for non-commercial research purposes.

What is "Activity time spent" ?

In this article, introduced is Activity time spent shown in 'Tracking Statistics' PDF file on 2018 FIFA WORLD CUP RUSSIA.

This PDF file and other data are open for public consumption.

You can find these data in 'Match Facts' page of each match. (e.g., 'Match Facts' page of Farnce vs. Argentina, fifa.com)


Players and a ball always move in a match.

Activity time spent indicates time percentages of players' velocity zones.