|
|
Predicting the outcomes of tennis matches. How important is the factor of different surfaces?
Sklenička, Jan ; Kurka, Josef (advisor) ; Palanský, Miroslav (referee)
This thesis focuses on predicting the outcomes of ATP tennis matches with the aim of investigating the effect of different tennis surfaces. We define nine surface variables based on two different clustering methods, and incorporate these variables into models that were estimated using logistic regression. Such approach allows us to robustly observe how the effect of specializing on a spe- cific surface translates into the match winning probability. The accuracy of the models ranges from 63% to 65%, with the models incorporating the effect of surface specialization displaying a superior predictive accuracy. Based on several evaluation metrics, the results confirm that surface specialization is a crucial variable when predicting tennis matches. We are also able to identify that the most accurate metric to measure surface specialization is the difference in normalized winning rate. Moreover, the analysis reveals that the specializa- tion on grass courts is the most important compared to clay or hard courts. Lastly, we formalize betting strategies based on the predicted probabilities from the models, and we are able to achieve positive out-of-sample return on invest- ment (ROI). Accounting for the effect of surface specialization increases ROI on the non-naive betting strategies. JEL Classification...
|
|
|
Comparison of Models for Probabilities in Football Betting
Kožnar, František ; Večeř, Jan (advisor) ; Hlávka, Zdeněk (referee)
The aim of the thesis is to compare different statistical models for football betting odds and determine the best performing once based on the historical performance of sport teams. There are at least two possible approaches for computing the odds, namely Poisson regression and methods based on statistical machine learning. The idea is that the historical performance of teams is a good predictor of the future performance. Thus we can take the past performances, say all matches in the full season of the Bundesliga (306 matches), and use these data for predicting the odds for the following season. The resulting odds should be compared with the actual results using the scoring rules, which will identify the best performing model. 1
|
|
|
Comparison of Models for Probabilities in Football Betting
Kožnar, František ; Večeř, Jan (advisor) ; Hlávka, Zdeněk (referee)
The aim of the thesis is to compare different statistical models for football betting odds and determine the best performing once based on the historical performance of sport teams. There are at least three possible approaches for computing the odds, namely logistic regression, Poisson regression and methods based on statistical machine learning. The idea is that the historical performance of teams is a good predictor of the future performance. Thus we can take the past performances, say all matches in the full season of the English Premier League (380 matches), and use these data for predicting the odds for the following season. The resulting odds should be compared with the actual results using the scoring rules, which will identify the best performing model.
|
|
|
Prediction of Betting Odds Based on Logistic Regression
JANDA, Pavel
Betting odds in various areas of human life is becoming very popular. This Bachelor's Thesis deals with the prediction of betting odds based on logistic regression. This work describes general information about betting odds, the history of betting odds both here in the Czech Republic and abroad, the forms of betting odds, who is a bookmaker and how betting odds are created. There are 3 sets of data. Specifically, it is the advantage of the home ground, the pre-match table and previous matches. Data is processed by lo-gistic regression to predict betting odds. The three above conditions are included in the Statistica program, which calculates the probability of each variant based on logistic regression. The results from the Statistica program are then compared to the actually listed betting odds.
|
guest ::
