Authors: Juan Manuel Mart\u00edn Gonz\u00e1lez, Yves de Sa\u00e1 Guerra and Juan Manuel Garc\u00eda Manso. IX World Congress of Performance Analysis of Sport. 1, Worcester, West Midlands (UK). Ed. ISPAS, 2012. ISBN 978-1-1361-6737-9<\/p>\n
<\/p>\n
Abstract<\/strong> Authors: Juan Manuel Mart\u00edn Gonz\u00e1lez, Yves de Sa\u00e1 Guerra and Juan Manuel Garc\u00eda Manso. IX World Congress of Performance Analysis of Sport. 1, Worcester, West Midlands (UK). Ed. ISPAS, 2012….<\/p>\n","protected":false},"author":27,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[9],"tags":[],"yoast_head":"\n
\nThe score of a basketball game is a direct reflection of the dynamic and non-linear interactions of the teams and its components (Kubatko et al. 2007; George et al. 2009; Yarrow et al. 2009; Ziv et al. 2010). However, the evolution of the score in time seems to have certain patterns or properties that confer identifiable characteristics of each league. We tried to identify them in order to know in detail the internal logic of competition. We studied a total of 5 seasons (1230 games per season, with a total of 6150 games) of the NBA regular season. In every game we analyzed the game transcription published by the NBA in which are described in detail, all incidents that occur play by play (NBA). In basketball, the time between points is a random process, which a part follows an exponential distribution defined by \u03bb. It is the relationship between the number of events and the time. The distribution has a peak around 20\u00b4\u00b4, and a possible long tail beyond 100\u00b4\u00b4, with a maximum value of 310\u00b4\u00b4. The most likely time between goals is around 20\u00b4\u00b4. Below and above these values the probability drops rapidly, although the effect is much greater for short baskets times. For higher values of this peak, the probability decreases until attained a certain value, begin to be considered rare phenomena (low probability). For values above 100\u00b4\u00b4 is possible that it begins to exhibit similar behavior to a power law. The \u03bb value does not remain stable throughout the game. In the first quarter, it presents the lower values than the rest of the quarters. And the last one stands out for its end so marked. Moreover, the Coefficient of Variation tells us that the data behave uniformly (Poisson process) to a certain value, from which this tendency is weakened and becomes much more heterogeneous.<\/p>\n","protected":false},"excerpt":{"rendered":"