Sports Analytics: A Guide for Coaches, Managers, and Other Decision Makers by Benjamin C. Alamar

Author:Benjamin C. Alamar [Alamar, Benjamin C.]
Language: eng
Format: mobi, epub
Tags: Business &#38, Sports &#38, SPO068000, Recreation/Business Aspects, Economics/Information Management, BUS083000
Publisher: Columbia University Press
Published: 2013-07-01T14:00:00+00:00

SURVEY

The survey phase identifies and examines the state of both the relevant statistics currently in use and the availability of relevant data. Typically, whatever the goal of the new metric, there will have been previous attempts at filling the need. These previous attempts may not have managed to capture all of the important dimensions of the need or might have been scouting-based subjective grades or qualitative analysis. It is important to identify previous attempts to answer the same question in order to clarify the goal of the new metric. Identifying them will also inform the decision-making context to be used in the analysis. The result of the survey phase should be a clear and realistic concept for how to build a metric that will help inform the decision-making process.

The survey phase for the creation of PER began with identifying the tools currently used to gauge the effect players have on their team’s success. These included statistics such as points per game, rebounds per game, and field-goal percentage. The comparison of players was done typically by comparing this array of statistics, but Hollinger identified two key issues: the statistics were not comparable across players, and the statistics were not comparable to one another.

The existing basketball statistics did not take into account differences in playing time, which rendered them generally not comparable across players because a player’s opportunity to create points, rebounds, or turnovers is controlled by time on the court. A player who averages ten minutes a game has fewer opportunities to score than a player who plays thirty minutes a game. A starter may have a higher points-scored per game average than a better-shooting backup because he is on the court three times as much.

The statistics were not comparable to one another because they measured different outcomes; there is no clear manner to compare the value of a defensive rebound with two points scored or a personal foul. For example, during the 2011–12 NBA season, center DeJuan Blair of the San Antonio Spurs averaged 9.5 points, 5.5 rebounds, and 1.2 assists per game while shooting 53.4 percent from the field, and Spurs center Tiago Splitter averaged 9.3 points, 5.2 rebounds, and 1.1 assists per game while shooting 61.8 percent from the field (see table 5.1). Based on these statistics, Blair had a slight edge in points, rebounds, and assists, and Splitter converted a higher percentage of his shots. Are the small advantages in the first three categories enough to suggest that Blair is the better player, or is Splitter’s superior FG% the controlling factor? As Hollinger was trying to decide between two players, it became clear that a more systematic approach that allowed for the comparison across both players and statistics would create a clearer picture of the player’s overall contribution.

With these clarifications in mind, Hollinger could gather the relevant data. As the purpose of the metric was to combine all contributions into one metric, all of the measured court activities should enter into the calculation. Classifying the available data

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.