CMStatistics 2016: Start Registration
View Submission - CMStatistics
B1116
Title: A new Markovian model for tennis matches Authors:  Marco Ferrante - Universita di Padova (Italy) [presenting]
Giovanni Fonseca - Universita di Udine (Italy)
Abstract: A generalisation of previously considered Markovian models for Tennis is presented that overcome the assumption that the points played are iid. Indeed, we postulate that in any game there are two different situations: the first 6 points and the, possible, additional points after the first deuce, with different winning probabilities. We are able to compute the winning probabilities and the expected number of points played to complete a game and a set in this more general setting and we test our results considering 57 matches between Novak Djokovic, Roger Federer and Rafael Nadal.