This course will cover basic concepts of probability and data analysis as they apply to the design and analysis of interactive media and games. Students will study appropriate use of probability distributions in the design of interactive experiences, and the use of data analysis methods to understand user behavior in games and other interactive experiences. Topics will include discrete and continuous probability distributions, programming techniques to produce samples from different distributions, descriptive statistics, exploratory data analysis and using existing tools to collect and analyze data from gameplay. This course counts toward the Quantitative Science component of the university-wide Mathematics and Science Requirement for IMGD majors only.

Recommended background: High school algebra

- Gain proficiency using modern tools for data acquisition and analysis.
- Understand basic probability and statistics as it applies to data analysis.
- Develop skills for presenting game data analysis both orally and in written form.

- Use a spreadsheet (e.g., Microsoft Excel) to analyze and visualize game data.
- Use a scripting language (e.g., Python) to extract and clean data recorded from a game.
- Apply summary statistics to game data.
- Compute probability distributions for game data.
- Write reports with graphs and tables illustrating detailed analysis of game data.
- Present part of game dataset report in a formal setting (e.g., class with a time constraint) using appropriate visual aids.

David M. Levine and David F. Stephan. "Even You Can Learn Statistics
and Analytics", Third Edition, *Pearson*, 2015. ISBN:
978-0133382662.

The book is unfortunately named since it suggests a degree of ineptitude in the student, but the content is well-presented with nice examples. The technical depth is sufficient to provide a decent foundation for game analytics.

List of topics covered in this course (not necessarily in order of appearance):

- Data analysis tools and pipeline
- Statistics
- Probability
- Visualizing and presenting data
- Hypothesis testing
- Regression

Final grades will be computed as follows:

- Projects:
*60%* - Presentation:
*10%* - Exams:
*30%*

**Projects:** There will be five projects done that have
students engage in data analysis for game development. This includes
hands-on work with appropriate data collection, statistics and
visualization tools for analysis. Work on each project will be
demonstrated through a written report.

Although content is a significant part of any project writeup, content means little if the reader cannot easily extract the information and main messages. Thus, written reports are to be graded on presentation clarity as well as content. Writing should suitably concise, follow directions, and provide the needed information precisely. While it is not the intent to play "English teacher" while grading, flagrant errors in grammar, organization, and/or style may affect the grade.

**Presentation:** Each student will present exactly one project
report orally to the class. The content will be drawn directly from
the student's own report, with appropriate visual aids (e.g., slides).
The presentation length will be constrained to a maximum of 8
minutes.

Presentations will take place the day after each project is due, where 5 people chosen at random will present. There will may be the opportunity for 1 voluntary "do over" if time permits.

The presentation grade will be based on speaking, organization and preparation, visual aids, clarity and overall professionalism.

**Exams:** There will be two in-class exams. The first is
roughly mid-way through the term and the second on the last day of the
term. There is a remote possibility of a pop quiz for which no
advance notice will be provided. Exams will be closed book and closed
notes, unless otherwise indicated. The majority of each exam will
cover basic ideas and objectives of the class with a few questions
testing additional understanding and insight.

The final grades earned will reflect the extent to which a student has demonstrated understanding of the material and completed the assigned projects. The base level grade will be a "B" which indicates that the basic objectives on projects and exams have been met. A grade of an "A" will indicate significant achievement beyond the basic objectives. A grade of a "C" will indicate not all basic objectives have been met, but work was satisfactory for credit. No incomplete grades will be assigned unless there are exceptional, extenuating circumstances. Similarly, no makeup projects or exams will be given unless there are exceptional, extenuating circumstances.

A letter "S" grade will be awarded to students with superlative performance, a recognition above that of an "A" grade. Only the top 10% of the class will receive such a grade. While the "S" grade will not show up on an official WPI transcript (it will show up as an "A"), students that receive an "S" are more than welcome to claim the grade on a resume, blog or any other forum wish - I'll stand behind it.

Exams are done, and due, in class. Projects are due online at
11:59pm on the due date, unless otherwise noted. Late projects will
be accepted with a 10% penalty of the total value for each day (24
hours) late, with the weekend counting as one day. Projects
will *not* be accepted more than one week beyond the due date.

This course is intended for serious students. Participants will be expected to adhere to all rules of professional behavior. It is to be emphasized that knowledge of material and professional behavior are tied together; failure in one of them negates any excellence in the other.

All work is expected to be done individually. As such, students are encouraged to discuss their work with each other, but are also expected to do the work by themselves. A guideline is that code, data, graphs and analysis can all be shared by students at the same place and time (e.g., by looking over each other's shoulder and explaining and discussing). But the line is drawn at digitally sharing such materials - e.g., there is to be no emailing, cutting-and-pasting or otherwise digitally copying code, data, graphs or tables from others.

You may use your own computer during the lectures, but do not just
follow along - research has shown that students who are "active" and
take notes during class retain more information, even if they do not
use those notes
later. **Take
notes!** Do this on paper or with the computer. So, laptop use
in class is acceptable as long as it is restricted to note taking, or
for a limited amount of information seeking. Any other activity
distracts you or the people around you, preventing thoughtful
participation in the class. Students will be asked to shut down their
computers in such a case.

Also, those who multi-task (e.g., with laptops or cell phones during class) learn less. In fact, those who believe that they are expert multi-taskers, often do the worst! Productivity can be reduced by as much as 40 percent when switching tasks. The "multi-tasking" state of mind has been labeled continuous partial attention. During class, try to limit multi-tasking with any device you have - stay with the task on hand of paying attention to the lecture. Maintaining focus on a difficult skill, but one that gets better with practice.

Any breach of professional ethics as evidenced, for example, by
copying exams or material (e.g., data and/or graphs) for the projects,
from other students or the Internet, or using outside help of any
kind, is considered adequate reason for an NR in the course and a
report to the Dean of Students. Refer to the
official WPI
statements on Academic Integrity for details. Remember this
warning - any breach of ethics will earn you an NR and an official
report. When in doubt, *ask*!

Slides from class lectures and other in-class materials are available shortly before or after they are presented. Both powerpoint and pdf are provided, along with an indication of the relevant chapter in the course textbook.

Topic |
Source |
Textbook |

Administrative | pdf, pptx | |

Introduction | pdf, pptx | |

Fundamentals of Statistics | pdf, pptx | Chapter 1 |

Presenting Data | pdf, pptx | Chapter 2 |

Descriptive Statistics | pdf, pptx | Chapter 3 |

Probability | pdf, pptx | Chapters 4 & 5 |

Inferential Statistics | pdf, pptx | Chapters 6 & 7 |

Presentations | pdf, pptx | |

Review | pptx |

Projects and due-dates will be placed here as they are defined. Here is what we have so far:

Project 1: League of Legends Player Analytics (

*mar 21*) (slides) (grading guide)Project 2: Mazetool Analytics (

*mar 29*) (slides) (grading guide)Project 3: TagPro Analytics (

*apr 08*) (slides) (grading guide)Project 4: League of Legends Game Analytics (

*apr 18*) (slides) (grading guide)Project 5: U-Pick Game Analytics (

*apr 24, apr 28*) (slides) (highlights)

This section has any samples discussed in class, exam preparation material, tutorials or any other demonstration-type class materials.

Exam materials:

- Mid-term exam topic outline
- Final exam topic outline
- See "Test Yourself" and "Problem" sub-sections at the end of relevant chapters

Step-by-step example of how to make a Normal Quantile Plot in Excel. Includes the resulting spreadsheet (.xls).

Presentation forms:

Riot Games API - Recent League of Legends games, ranked statistics, runes, masteries, and more available through a Web query (e.g., curl) and returned via JSON.

TagPro Analytics - Data and tools for analyzing TagPro matches, maps, players and the game in general.

A good example of descriptive text to accompany a graph made through data analysis.

Some data sets (in .xlsx format):

Mark Claypool (claypool at cs.wpi.edu)