Defining the Style of Play in the NHL: An Application of Cluster Analysis

Abstract

The placing of NHL players into categories such as stars, average journeyman forwards, defensive forward, etc., is a popular pastime of fans and professional sports commentators. This is usually done in an ad hoc manner, sometimes contentiously, and without reference to any detailed analysis of empirical data. This paper structures the debate and provides a statistical basis for such categorization. This is achieved through an application of cluster analysis to formally allocate players to these categories. Using variables from the earnings determination literature, we are able to define the style of play of both forwards and defensemen. In addition, there appears to be a link to external outcome measures.

Quantifying the Contribution of NHL Player Types to Team Performance

“…Next, we standardized each statistic by subtracting the average of that statistic over all players of that position and then dividing by the associated standard deviation. Therefore, we were able to create a similar scale for all the statistics and prevent one from dominating the clustering process (Vincent and Eastman 2009). For goalie statistics, we normalized wins and shutouts by GS and then standardized all four statistics.…”

“…For the forwards, we chose four clusters: Top Line F, Second Line F, Defensive F, and Physical F. Each category has a unique makeup and aligns with generally accepted player types. These categories are similar to those in Vincent and Eastman (2009); however, we have replaced their grinder category with the more granular Second Line and Defensive categories. Notice that the forward clusters provide an almost perfect ordinal classification scheme with respect to the standardized statistics.…”

“…Ingolfsson (2010) presents a comprehensive review of OR analyses in hockey. Vincent and Eastman (2009) inspired our research. They proposed a classification scheme for NHL hockey players by clustering them using their career statistics up to and including the 2002-2003 season.…”

“…In hockey, scientifically analyzing individual player performance is a challenging task because the game is so fluid. However, researchers have attempted to classify NHL players into different player types as a way to compare them and to analyze differences in salaries (Vincent and Eastman 2009;Jones et al 1997Jones et al , 1999. In this paper, we build on prior research by developing a novel classification scheme that is well suited to current NHL players.…”

Data Mining the Play-by-Play: Assessing and Applying NHL Performance Metrics Using Statistical Methods

“…By finding an appropriate measurement of player value, stakeholders could build around team strengths and weaknesses and help determine player compensation. 2 Vincent and Eastman [10] attempted to categorize NHL players through k-means clustering on player weight and per-game points, penalty minutes, and plus-minus. After mathematically justifying the number of categories of players, forwards were split into "grinders", "scorers", and "enforcers", while defensemen were categorized as "scorers" and "aggressors".…”