Creating and sharing knowledge for telecommunications

uPATO applies social network analysis to team sports


on 06-05-2018

... Developed by a research team from IT in Covilhã, the uPATO - Ultimate Performance Analysis Tool is a software tool for social network analysis applied to team sports. There are other software tools available but normally they are not free, or they do not have all the metrics need. For example, the metrics that used the geodesic distance sometimes are not available in some tools.

To do a social network analysis it is necessary a representation of the network (i.e. a graph or a digraph). Thus, a first module was developed to allow codifying the network, which in the case of team sports is given by the sequences of interactions between teammates. This module, for example, allows the user to codify a sequence of passes during a match and to generate an adjacency matrix, which represents the weighted digraph.

A second module of the tool involved the programming of all metrics available for network analysis applied to team sports based on adjacency matrix. The metrics were codified for all network possibilities, i.e. for graphs and digraphs, weighted or not. This means that uPATO tool can be applied for social network analysis in team sports, but also for any network that can be represented by an adjacency matrix (e.g., a computer network or a telecommunication network, etc.). Thus, the metrics are divided into three types: individual metrics, sub-group metrics, and team metrics. These make possible to compute different network measures, providing scores to classify the collective and individual performance over the network.

It also does the graphical representation of the graph or digraph (weighted or not) allowing an easier understanding of which nodes in the network (i.e. the players in case of team sports) are more important.

In the future, the uPATO tool will have an additional module for geolocation data processing. Currently, some teams use GPS devices to have the position of the players during the match (e.g., FieldWiz and TraXports formats will be supported). Thus, uPATO will have a set of metrics based on geolocation data of players. This new functionality extends the uPATO capacities for team sports analysis but also for other activities where GPS data is available. However, it does not consider yet the possibility of the ball with a GPS device.

On the photo: Layout showing the weighted diagraph representation of passes between teammates during a football match