Hello! Thank you for playing our little game. You’re helping some research
that we think is genuinely fascinating, and we can’t thank you enough!
Tom & William
🚨🚨🚨 NEW RESEARCH ALERT 🚨🚨🚨
We’ve released the first iteration of the RPGLite dataset! You can find our players, completed games, and application interactions in our public dataset by following this link.
We’ve also presented a paper at the GAME-ON 2020 conference, in beautiful Aviero, Portugal (via zoom). It’s on some of the things we learned developing the game itself, rather than research on this dataset. You can even read the paper we wrote! Go check it out here!
If you’re here for information about our game, you can find relevant documents
If you’re here to ask some questions, here’s an FAQ. If it doesn’t answer
your question already, feel free to email us! Our email is at the bottom of the
What do I do if my slots are all full?
- We decided not to increase the number of game slots beyond 5 to incentivise people to make their moves promptly. If your opponent has taken over 24 hours to make their move, then you can claim victory and it will be treated as though you won the game, and your opponent lost.
How do I unlock new characters?
- You unlock character 5 by finishing an online game with characters 1-4, then characters 6, 7 and 8, by finishing an online game with characters 5, 6 and 7 respectively.
How can I find new opponents?
What’s the deal with skill-points?
- Skill points increase by 30-50 when you win a game and decrease by 5-20 when you lose a game. They will not fall below a multiple of 100, unless you forfeit a game.
The game seems very simple, why?
- We cannot use intricate characters or abilities for research reasons (the methods we use to calculate what moves are optimal would not be able to handle them).
How do I report a bug?
- Send us an email at the addresses listed below.
What data are you collecting?
- We do not store any personal data from our users, no names, numbers or email addresses. We care only about how games are played and we do this by keeping a record of what moves are made in every game. For every user we store some information about how many games they have played / won and what characters they have used / won with to contextualise information from the way in which the game is played.
What kind of questions are you hoping to answer?
We hope that the dataset will allow to answer a series of fascinating questions, such as:
- Do players learn the optimal strategies for a simple game organically? And how long does the process take?
- How do more successful players interact with the app compared to less successful ones?
- What gamification elements do users interact with more?
- Can our users be grouped into distinct types?
- How realistic are our fabricated human-behavioural datasets?
- How do players respond to changes to game material / balance patches?
- How do users respond to seeing messages prompting them to explore various parts of the app?
As well as some more trivial questions, such as:
- Are any of the customisation options correlated with players who win more often?
- How do players react to exceptionally good / bad luck
- The effect of changes to a ranking system on player behaviour
Why have you built this at all and not just used player data from an already-released game?
- We have used modern model checking techniques to calculate all optimal and adversarial strategies for every combination of character in the RPGLite. The technique we use is one which we came up with ourselves and introduced in this journal paper.
Tom is available at email@example.com,
William is available at firstname.lastname@example.org.