Below you’ll find books that I’ve read in the context on game analytics and data science. I’ll only list books that I’ve read and somehow added something to my knowledge and day to day work which may include some less obvious choices like productivity and management books. I won’t add links as you might prefer to get them from your local store, online publisher or Amazon in all its international and local flavours.
An Introduction to Statistical Learning (Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani)
I often go back to technical books for a snippet of information but this is the only one I’ve read through more than once. I simply love this book! It is a great entry point to Statistical Learning, one that you can use for both data analysis and machine learning. It addresses a world of regression, classification and clustering models and many other things like resampling, accuracy, non-linearity, regularisation and many other things.
Two of the authors (Hastie and Tibshirani) are responsible for many wonderful models like the lasso and generalised additive models and they are the lecturers of Stanford’s Statistical Learning course that I reviewed in the Courses page.
Creativity Inc. (Ed Catmull)
If you work in game development you should know who Ed Catmull is. Texture mapping, anti-aliasing algorithms and z-buffering were invented by him. He is also the president of Pixar and it is in this context that he wrote this book.
Creativity Inc. is about problem solving and management. You won’t find any recipes here, only pure inspiration about what a leader is, even one that does not see himself as such.
It is simply my favourite non-fiction book and I’ll recommend it to anyone. However if you are a manager that deals with constant change on a highly creative environment this is possibly a life changing book.
Data Points (Nathan Yau)
Human beings perceive data visualizations in very specific ways. Data Points is my book of reference about this matter approaching everything from scale, area and colour to types of graphics and how to make the best of it.
If you do data visualisation, this is a must have.
How to lie with statistics (Darrell Huff)
This book is a classic. If you work with statistics you probably read it already. If you didn’t, shame on you! You’ll become a brutal ice cold skeptic… but in a good way. You’ll avoid many mistakes if you are a decent professional or learn every bad trick if you are someone I’d like to fire. This is one of those books that can be used for evil. But you are one of the good guys, aren’t you?
Modelling Techniques in Predictive Analytics (Thomas W. Miller)
I just finished this book so my opinion may change. It is quite a practical book with code for all the sections. Each section is about a specific problem, e.g., Association Rules. The code is in R which suits me nicely and it is pretty obvious in the subtitle Business Problems and Solutions with R that this is by design.
What I don’t like about this book is that it too intuitive and doesn’t go in depth in some areas. In this sense it is a bit academic where it shows us just enough for us to dig for the rest.
Predictably Irrational (Dan Ariely)
If you follow this blog it shouldn’t come as a surprise that I focus on user research quite heavily. One of the areas that explains most player behaviours is Behavioural Economics. The topics of this book that interested me the most were how a person sets its buying preferences and makes decisions regarding purchasing (or not) a product.
Highly recommended if you do any kind of user research.
The Data Science Handbook (Carl Shan, Henry Wang, William Chen and Max Song)
This book offers a set of interviews with very important people from the data science community being the most notorious DJ Patil, Obama’s Chief Data Scientist. It is not a technical book but very insightful seems it allows to avoid all the media noise about data science and have a direct link to the opinions of those who lead the data science community.
Recommended if you are interested in the insights.
The 7 habits of highly effective people (Stephen R. Covey)
One of the most (if not the most) influential books on productivity, the 7 habits is a fabulous piece, my favourite non-fiction book until I read Creativity Inc. The book addresses productivity from the perspective of dependence, interdependence and continuous improvement going into detail and explaining how these affect our productivity and the way we see the world.
A must read in my opinion.
The signal and the noise (Nate Silver)
Nate Silver became known by his methods to predict elections and baseball results. His blog FiveThirtyEight is a reference for all things statistics and this book a true master piece in how statistics are badly treated everywhere with case studies of unbelievable and yet very accessible detail.
A must read if you work with statistics.
Think like a freak (Stephen Dubner and Steven Levitt)
Take any advice I give about these two with a grain of salt for I am a fan boy. Since the day I read Freakonomics that I devour everything Stephen and Steven have to offer. I had a long debate with myself regarding the inclusion of Freakonomics and Super Freakonomics in this list for instance.
But in the end “Think like a freak” deserves a place here and the others don’t. The reason being that this book is probably the only one that simplifies the scientific method to the masses with great examples on how to and how not to do some things, like asking questions, gathering data, presenting results.
Visualize This (Nathan Yau)
A hands on book about visualisation more helpful to the data journalist than to the data scientist, Visualize This was my entry point to not use default visualizations. It is a good starting point about data story telling but much more useful to someone that uses data visualizations has the end result than as a tool.