Gameplay analysis through state projection

Abstract

Analysis of gameplay data is crucial for evaluating design decisions and refining a game experience. However, identifying player strategies and finding areas of confusion is difficult because a designer may not know what queries to ask or what patterns to look for in the data. To make this task easier, we present Playtracer, a method for visually analyzing play traces that is independent of a specific game's structure. Playtracer applies multidimensional scaling to cluster players and game states, providing a detailed visual representation of the paths the players take through a game. We evaluate our method by analyzing an educational puzzle game and highlighting common hypotheses, pitfalls, confusing elements, and anomalies. Our results suggest that Playtracer can be an effective tool for game analysis and improvement.

Using extracted features to inform alignment-driven design ideas in an educational game

“…In these cases it may be more useful to generate features for student solutions based on the paths students take to arrive at the final solution rather than the solution itself. This would be similar to the work of Andersen et al who use Playtracer to map players' paths through a game space [3].…”

A Mixed Method Approach for Evaluating and Improving the Design of Learning in Puzzle Games

“…The method also fits seamlessly the type of small-scale user test frequently undertaken in early development phases, without requiring the large-scale data acquisition and advanced statistical or visualisation techniques that are involved in analytics approaches [e.g. 2,16,17,32]. The approach could be used throughout the development process -from being applied to evaluate learning difficulties around key game mechanics in low fidelity prototypes to evaluating the impact of specific design elements within more developed versions.…”

“…2,16,32] may prove useful in a commercial GUR context, those methods often require access to large data sets and are arguably less appropriate for earlier prototypes. While there are circumstances where such an analysis has been conducted on a smaller number of players e.g.…”

“…[2] who developed the Playtracer method, where multidimensional scaling is used to produce visualisations of how groups of players move through a game, and Owen et al [32] who present the ADAGE (Assessment Data Aggregator for Game Environments) protocol for tracking players interactions and the leveraged machine learning, data mining and statistical methods to assess learning within play.…”

An inclusive view of player modeling

“…To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Two simple examples of categorizing existing player modeling techniques using our taxonomy are the Playtracer system [1] and Yannakakis's preference modeling. Playtracer is a tool for analyzing recorded play traces, which produces Individual Induced Descriptive Action player models in our terms.…”