World of Warcraft, is the most popular pay-to-play massively multiplayer online role-playing game in the world. This means that instead of being the only person playing in the world like in single player games, there are thousands of players in the same world as you, playing with you.

N.C . State’s Professor David L. Roberts and Ph.D . student, Brent Harrison, have co-authored a research paper based on a group of 14,000 World of Warcraft players and their playing habits.

According to the research paper, Roberts and Harrison are able to predict, with up to 80 percent certainty, how individual players will play World of Warcraft and what choices they will make in the game. 

Their research places players into groups that are based on the game’s achievement system.  Achievements are granted to players who perform specific tasks or challenges.  There are 749 achievements in total, ranging from fishing awards to beating dungeons.  But what’s intriguing is that Roberts and Harrison have found that achievements are often earned in groups.  They have labeled these groups “cliques.”

According to a press release, “if a clique consists of seven achievements, and a player has earned four of them, the researchers found that they will probably earn the other three.”   However, what’s strange is that these cliques are composed of largely unrelated achievements.  For example, achievements for unarmed combat are grouped with achievements for travelling the game’s world. 

The paper is focused on World of Warcraft but according to Roberts, “[the research] also applies to any setting where users are making a set of decisions.  That could be other gaming formats, or even online retailing.”

The researchers also hope to be able to help both game developers as well as the game players with their findings.  Predicting player behavior can easily be used to improve a game overall.  Because the paper uses a data-driven modeling approach, developers would have to do little work to find out what their players enjoy.

Developers will be able to implement the research by seeing what their players enjoy. Thus, they can gear future updates to their game towards the intended audience. 

Gamers themselves can also benefit from Robert and Harrison’s work. If the game developers can predict what players will want to do in their game, they can then program that into the game and help players find things that would be enjoyable to them.  Instead of players wasting their times on mundane tasks they couldn’t care less about, they will instead be guided to entertaining content that makes paying a monthly fee for the game worth it.

Kevin Dearing , a junior in psychology, was an avid World of Warcraft player until recently. His experience with the achievement system was not similar to what was found in Roberts and Harrison’s paper.  Dearing stated that he “didn’t usually group achievements together, except when they were in close proximity.”   Thus he did not follow the research paper’s model.

Some players however, chose to not even participate in achievement hunting. These users are outliers for the research paper. Their playing habits cannot be as easily predicted.

One such player is Will Jackson, a junior in business.  According to Jackson, the achievements were a “weak gimmick that took away from actual gameplay.”

Because he did not take part in the achievement system, his in-game behavior most likely would seem more erratic to the researchers.

Using the achievement system to predict player actions has another weak point as well.  Many of the achievements are earned simply by leveling up your character.  Therefore, the specific actions of the players would be very difficult to predict since there are a large number of options for leveling a character.

However, the system that Roberts and Harrison have developed has proven to be a useful for those players who strive for higher achievements.