Host: This week, researchers from North Carolina State University have developed a new method that can accurately predict the behavior of players in online role-playing games. The research team developed the data-driven predictive method by analyzing the behavior of 14,000 players in the massively multiplayer online role-playing game (MMORPG) World of Warcraft.
Melmoth: The method predicts that most MMO player behaviour will involve frothing with unbridled enthusiasm for a year or more before the game’s release, then playing for two weeks before unsubscribing. With the second week being optional.
Zoso: With a mere six months of algorithm development and three years of supercomputing processing power, the team determined that if a player kills nine rats there’s an 83.2379% chance that their next action will be to kill a rat.
Host: “We are able to predict what a player in a game will do based on his or her previous behavior, with up to 80 percent accuracy,” says Brent Harrison
Melmoth: “Our method shows that if a player is a player-killing smack-talking bigoted, terminal spacebar abuser, they will go on to lead a very successful Twighlight-based role-playing guild enacting heartbreaking eternally doomed love stories. Well, 20% of the time, at least.”
Host: The researchers are confident this work can help game designers. “This research can help researchers get it right, because if you have a good idea of what players like, you can make informed decisions about the kind of storylines and mechanics those players would like in the future.”
Melmoth: Having made use of this method, Blizzard reports that the next expansion for World of Warcraft will involve a big red button in the middle of the screen, which when pressed has the computer shout “OH MY GOD, YOU WIN, YOU’RE THE BEST!” in a sexy voice, and then showers the player’s character in epic armour and sparkly mounts.
Host: It’s not just games that stand to benefit. “This work could obviously be used for World of Warcraft or other MMORPGs,” says Roberts, “but it also applies to any setting where users are making a series of decisions. That could be other gaming formats, or even online retailing.”
Zoso: As a result of the study Amazon are trialling a shopping basket that forces you to buy ten pencils that you don’t need before you’re allowed to join four other customers in trying to buy the book you really want (so long as none of the others beats your “need” roll for it).
Host: The study could help players. “For example,” Roberts says, “you could develop a program to steer players to relevant content. Because it is a data-driven modeling approach, it could be done on a grand scale with minimum input from game designers.”
Melmoth: When asked whether their system simply put giant neon signs into the game at five yard intervals which read “KILL RATS THIS WAY”, the professor looked sheepish and refused to comment.
Host: The study is based on the order in which players earned achievements, and apparently an achievement dealing with a character’s prowess in unarmed combat is highly correlated to the achievement badge associated with world travel – even though there is no clear link between the two badges to the outside observer.
Zoso: Building on this work, a study of stamp collectors has revealed a high correlation between owning stamps with pictures of flowers on them and owning stamps with pictures of people on them – even though there is no clear link between flowers and people to the outside observer.
Host: According to the abstract, Our underlying assumption is that we can accurately predict what a player will do in a given situation if we examine enough data from former players that were in similar situations.
Melmoth: Let’s see what the KiaSA AI makes of it…
**** Reading previous player data ****
*** Data read ****
*** 14000 players killed rats for experience ****
*** 80% chance player will kill a rat for experience ***
*** 5% chance player will kill a wolf for experience ***
*** 5% chance player will kill a boar for experience ***
*** 5% chance player will kill a bear for experience ***
*** 3% chance player will kill a goat for experience ***
*** 1% chance player will realise the futility of attempting to win a skinner box based treadmill
system designed to keep them playing for the longest amount of time possible whilst using minimal
resources of the publishing company, quit, fund and recruit an elite team of soldiers of fortune
to hunt down the publisher and all its subsidiary companies and rid the world of the these games
once and for all, thus returning millions of man-hours into the world development resource pool,
thus enabling the betterment of mankind and harbouring the Fourth Age where the human race colonises
the stars. ***
*** 1% chance player will make a ham sandwich ***
Studio lights dim, theme tune plays.