StarCraft 2 Replay Analysis

Peer reviewed: 
Yes, item is peer reviewed.
Scholarly level: 
Faculty/Staff
Final version published as: 

Video Game Telemetry as a Critical Tool in the Study of Complex Skill Learning

Date created: 
2013-09-18
Abstract: 

This data was used in Thompson et al. (2013).

We aggregated screen movements into screen-fixations using a Salvucci & Goldberg (2000) dispersion-threshold algorithm, and defined Perception Action Cycles (PACs) as fixations with at least one action. Time is recorded in terms of timestamps in the StarCraft 2 replay file. When the game is played on 'faster', 1 real-time second is equivalent to roughly 88.5 timestamps. A list of possible game actions is discussed in Thompson, Blair, Chen, & Henrey (2013). Skip league conditional inference forest classification was used (Bronze-Gold;Silver-Platinum;Gold-Diamond;Platinum-Masters;Diamond-Professional) to show changing patterns of variable importance with skill. Predicted attribute: League (Ordinal)

Description: 

NOTE:

The researchers have since discovered that a few individuals might have been able to submit multiple games to the study. In order to ensure independence of  observations, their consequent work drops the following games:

1172
2183
2652
4064
4075
5247
7011
7988
8236
8750

Doing so will drop the aforementioned players and all of the games they submitted.

Language: 
English
Document type: 
Dataset
Rights: 
You are free to copy, distribute and transmit this work under the following conditions: You must give attribution to the work (but not in any way that suggests that the author endorses you or your use of the work); You may not use this work for commercial purposes; You may not alter, transform, or build upon this work. Any further uses require the permission of the rights holder (or author if no rights holder is listed). These rights are based on the Creative Commons Attribution-NonCommercial-NoDerivatives License.
Statistics: