In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains egocentric ghosting models. The use of data-driven ghosting models is increasing as player tracking data becomes more widely available. The issue with the models that have been previously constructed is that their analysis of the players’ decisions is based solely on the…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains Convolutional Neural Networks. In deep learning, a convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyzing visual imagery. Convolutional Neural Networks use a variation of multilayer perceptrons designed to require minimal preprocessing. They are…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains conditional probability. Conditional Probability. Probability is defined as a measure of how often a particular event will take place if the experiment occurs repeatedly. Probability ranges from zero to one. Zero indicates that the event is impossible, and one indicates that the…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains boruta algorithms. Boruta is an all relevant feature selection wrapper algorithm, capable of working with any classification method that output variable importance measure (VIM); by default, Boruta uses Random Forest. The method performs a top-down search for relevant features by comparing original…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains Strategy and Match Modelling. This is a review of the team strategy model and match model research conducted by Leonardo Lamas, Junior Barrera, Guilherme Otranto and Carlos Ugrinowitsch. ———————- Find out how Sports Analytics Expert Victor Holman can give your team the…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains weighted Voronoi diagrams. In mathematics, a weighted Voronoi diagram in n dimensions is a special case of a Voronoi diagram. The Voronoi cells in a weighted Voronoi diagram are defined in terms of a distance function. The distance function may specify the…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains qualitative exploration. Probably the oldest of all scientific techniques, qualitative exploration is a method used extensively by scientists, researchers and analysts to study human behavior, opinions, themes and motivations. Different from quantitative data, which relates to quantities, amounts and measurements that can be…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains spatio-temporal trajectory clustering. Spatio-temporal trajectory clustering methods identify heterogeneous patterns and explore underlying mechanisms and are designed to consider both temporal and spatial information in trajectories. Applying the time-dependent shortest-path distance measurement and taking advantage of topological relations of a predefined network,…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains network-style models. The network model is a database model conceived as a flexible way of representing objects and their relationships. Its distinguishing feature is that the schema, viewed as a graph in which object types are nodes and relationship types are arcs,…
In this installment of Victor Holman’s Sports Analytics 3 Minute Drill, Victor explains measuring mean and median. Mean and median are two of the most useful methods used to summarize numerical data. The mean of a set of data is simply the average, which is found by adding up the total of all the observations…