Data-Driven Learning Tools for Esports

Competitive strategy games are the focus of a great deal of research in artificial intelligence and game-playing. They also are an important hobby to millions of people worldwide. Players and followers of esports very often have one common goal: learning more about their chosen game. With the release of Dota Plus, which is Valve's subscription service Dota 2, it is clear that both the industry and players/followers of esports are interested in data-driven tools to help players improve.

Multiplayer Online Battle Arena, or MOBA, games are typically 5v5 team games in which each player has a set of complex abilities, and each team must gain resources by killing the other team and the other team's army, with the final goal becoming strong enough to destroy the opposing team's base. This provides a lot of disparate problems to solve simultaneously and in real-time, such as micromanagement of your player-character, planning long and medium-term strategy as the game goes on, predicting the opposing team's strategy and much more. This provides a multitude of problems for a player to solve, which is why complex MOBAs are so good at player-retention: it's virtually impossible to solve the game and the learning curve never ends.

I aim to explore the creation of various tools to help existing players of competitive video games learn and improve. One important thing the tools must achieve is not being a crutch: if the tools simply tell you what to do throughout the game a player can follow the instructions blindly without learning anything. Instead the tool should encourage the player to make a good decision and reflect on why it was good.