What is it about?
We started from the database of results of official matches in tables tennis Italian champtionships and constructed the graph (or network) in which players are the nodes of the graph and edges between two nodes connect players who played against. The edge might be directed and weighted according to the meaning: it can be the number of won matches, the difference of won sets, and so on and the edges can be directed or not. In this paper we showed how various metrics computed on the graph nodes (such as centrality, number of edges, etc) are correlated to the players' performances (measured by the number of points, position in the ranking at the end of the season, and so on). We also tried more complex challenges such as prediction of the players' position in the ranking for the next year and prediction of emerging talents.
Featured Image
Why is it important?
The outcomes of this paper are important because they show that it is possible to apply with success the techniques of data analysis on the networks structure to discover with a certain, useful precision, some important measures of the players performance.
Perspectives
My personal perspective on this line of research is that this work can be further extended to prediction of other players' performances, such as the outcome of matches against players never encountered and also in other sports.
Rosa Meo
Universita degli Studi di Torino
Read the Original
This page is a summary of: The role of the network of matches on predicting success in table tennis, Journal of Sports Sciences, June 2018, Taylor & Francis,
DOI: 10.1080/02640414.2018.1482813.
You can read the full text:
Contributors
The following have contributed to this page
