Prediktivna analiza i modeliranje teniskih mečeva

master's thesis

• Title (english): The Predictive Analysis and Modeling Tennis Matches
• Author: Ana Šarčević
• Mentor: Mihaela Vranić (mentor)
• Committee member: Mihaela Vranić (predsjednik povjerenstva)
• Committee member: Boris Vrdoljak (član povjerenstva)
• Committee member: Damir Pintar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Nakon kratkog uvoda o pravilima u tenisu prikazana su i opisana dva hijerarhijska Markovljeva modela za predikciju ishoda teniskih mečeva. Prvi omogućuje predviđanje ishoda meča prije nego je sam meč započeo dok je drugim modelom dodatno omogućeno praćenje meča uživo. Pozornost je u radu posvećena i računanju ulaznih parametara modela, te je prikazan način kombiniranja statistika svakog meča ovisno o danu odigravanja povijesnih mečeva, o odstupanju od prosječne igračeve statistike te o broju odigranih poena u meču. Model je prilagođen da prati tijek meča te ažurira vjerojatnosti osvajanja vlastitog servisa nakon svakog odigranom poena u meču, uzimajući određenom težinom statistiku trenutnog meča kao i statistiku povijesnih mečeva, a prikazan je i način određivanja optimalnog težinskog parametra kojim bi se to i ostvarilo. Za kraj su implementirane različite strategije klađenja u svrhu ocjenjivanja kvalitete izvedenog modela.

Abstract (english)

After a brief introduction to tennis rules, two hierarchical Markov models for predicting the outcome of tennis matches are presented and described. The first one is used for prediction of the outcome of a tennis match before the match has started, while the other model also alows live match tracking. Attention is also paid to the calculation of the input parameters of the model, as well as the method of combining the statistics of each match depending on the day historical matches were played, the deviation from the avarage player statistics and the number of points played in the match. The model is adjusted to track the progress of the match and updates the likelihood of winning its own servise after each point in the match, taking with specific account the current match statistics as well as historical match statistics. Also a method of determining optimal weight parameters is presented. In the end, different betting strategies were implemented for the purpose of evaluating the quality of the derived model.

• Keywords: Markovljevi lanci; sportsko klađenje; tenis
• Keywords (english): Markov chain; sport betting; tennis
• Language: croatian
• URN:NBN: urn:nbn:hr:168:973497
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-14 01:07:00