Thursday, February 28, 2008
Chief Social Engineer
Sometimes people ask me whether I wanted to pursue a career in game development or journalism. No thanks! It is not just that I like to keep my hobbies apart from my job, but also some basic financial considerations. I have a good job with a six-figure salary (in US dollars), while the median income for a game developer is $73,000, and that is for working far longer hours than me. Sure, Richard Garriott is earning more than me, not quite sure about Tigole, but in general I'm better off in my current job than in most game development or journalism positions. As Darren, the Common Sense Gamer, recently noticed, there are a lot of kids applying for game development jobs wearing tattered blue jeans and a Half-Life head crab hat. And the industry reacts in a logical way to having lots of eager, technologically savy, but not wise in the ways of the world applicants: it exploits them by paying them less than they could earn if they worked in serious engineering or finance, and by having them work extremely long hours. Thus the EA widow and similar stories.
But even if I plan to stick with my current career, not believing in predictions that I will become the boss of Blizzard, I can dream about what job I would love to have at Blizzard, if I could set my own salary and job description: Chief Social Engineer. Because I believe that player behavior is very much influenced by game design, especially by rewards. And I am sick and tired of game developers stating that they wanted players to do one thing, and were surprised how most players instead did things the developers didn't want.
If a large number of players in your game does something you didn't want them to do, it is your fault, you designed the challenges and incentives badly.
Humans as individuals are unpredictable, able to perform acts ranging from saintly goodness to abominable evil. But get a large number of them together into the same environment, and their behavior becomes predictable. Whole sciences, like economics for example, are based on that. People react in predictable ways to obstacles and incentives, usually following the path of least resistance towards the biggest possible reward. If you have complete control of the environment, as a developer of a virtual world has, you can steer people in the right direction by simply setting up the obstacles and rewards in a clever way.
If you observed World of Warcraft over the past 3 years, it is actually a very good example how changing incentives changes people's behavior. If you had a graph that showed for every day since the start of the game how many people were busy doing PvP, solo PvE, group PvE, and raids, you would notice big movements linked to the big changes of how PvP works and is rewarded. There were times where you needed to play 15 hours a day of PvP for months to get an epic, and unsurprisingly not all that many people did so. When just before TBC the PvP reward system was changed to become cumulative instead of relative, a lot more players started doing PvP. When TBC came out, everybody was busy leveling to 70, and PvP declined a bit. But then every new arena season gave out better rewards than the previous one, also increasing the rewards you could get just for honor points, and nowadays the raiders are complaining that nobody wants to play with them any more, and everybody is in the battlegrounds and arena. The relative popularity of PvP changed significantly over time, and all because of how the incentives changed. And of course the popularity of PvP was also influenced by changes in obstacles. When Blizzard linked groups of servers together in battlegroups, it significantly cut the time people had to wait in queue to get into a battleground, and that made PvP more popular.
Now maybe Blizzard *wants* PvP to be the most popular activity in WoW. Having a strong PvP in WoW diminishes the competitive advantage of upcoming MMORPGs, which are mostly PvP-centric. And the earlier PvP weakness of WoW clashed somewhat with the lore of Warcraft as a series of RTS games. And maybe Blizzard *wants* WoW to be strong in solo PvE gameplay, because that certainly helped their sales. But as Chief Social Engineer I can't help but notice that there are problems with how WoW develops towards solo and competitive gameplay, and away from cooperative gameplay: it diminishes social contacts in the game, and that has a strong influence on churn rate and longevity. If all you do in the game is either solo or jump-in battlegrounds for which you don't need to form groups and guilds, then there are no social ties that keep you logging on after you ran out of content. There are a lot of games which you can play alone, and there are a lot of games where you can log on, frag a couple of random strangers, and log off again. It is social interaction and cooperative gameplay which make MMORPGs special, and ultimately justify paying a monthly fee.
Blizzard would be wise to hire if not me then somebody else as Chief Social Engineer, to have somebody to look at whether World of Warcraft's incentives are steering people in the right direction. Relatively simple changes, like increasing the group xp bonus, could already have a big influence on how much people play together and how much they play apart. It is a fallacy to think that "lots of people solo in WoW, this must be what they want". Some people want to solo, some people want to group, but many would do either, depending on what is more efficient. WoW's bad LFG system, level demographic development, and insufficient incentives for grouping makes soloing appear far more popular than it really is. If groups were easier to set up and more rewarding, more people would play together. The WoW devs correctly identified forced grouping as a weakness of games like Everquest, but then overcompensated and created a game which is already close to forced soloing. And if new players wander around Azeroth alone, can't find any new friends, and stop playing after a while because they feel lonely, Blizzard developers only have themselves to blame. If their previously most pampered class of customers, the raiders, starts posting "does Blizzard hate us?" articles, then maybe something went wrong with the design of incentives.