The basis of the technology is hierarchies and such information fits well with competitive sports, especially team based ones (rank distribution is a common way of measuring sports team performance). This is because performances are measurable and the performance of a team can improve or decline over time. It provides ‘big data’ for analysis. The researchers were interested to understand whether the formation of hierarchical complexity is determined by the particularities of each area (such as whether sports are different to the economy) or whether there are generic mechanisms of stratification that are common to different systems (such as a set of universal rules).
Using this data gathering approach Mexican researchers have taken the performance of national teams in sports like tennis, chess, golf, poker and football to test out a new computer model. They have used the data to explore universal features in the creation of hierarchies. An area they have looked at included how similar sports are to the stratified structure found with the distribution of wealth (billionaires down to those in absolute poverty).
With this in mind Dr. José Morales, who works at the National Autonomous University of Mexico, has found the model can predict changes in rank occupancy over the course of a contender’s lifetime. This is in relation to any of the sports or activities. The research is said to illuminate forecasting how stratification occurs in competitive activities like sports.
Here the researchers set out to predict statistical regularities that can explain how competition shapes the hierarchies of players and teams. Here the researchers focused on how performance rankings of players and teams across sports and for specific matches evolved over time. They call this evolution “rank diversity”, and they took this concept from linguistics (where linguists research how vocabulary changes over time).
The primary outcome from the sport-centric research was that ranking hierarchies appear to be driven by similar underlying mechanisms to rank formation. This appeared regardless of the characteristics of the players. In other words, the a given performance rank, when compared over a period of time, gives the same functional form as in languages. With linguistics competition is determined by the use or disuse of grammatical structures; with sport, competition is structured over a series of rules.
The research has been published in the journal EPJ Data Science. The research paper is titled “Generic temporal features of performance rankings in sports and games.”
