Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 22, 2014
arrowPress Releases
October 22, 2014
PR Newswire
View All

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

GDC 2011: An Epidemiologist's View Of  World Of Warcraft 's Corrupted Blood Plague
GDC 2011: An Epidemiologist's View Of World Of Warcraft's Corrupted Blood Plague
February 28, 2011 | By Kris Graft

February 28, 2011 | By Kris Graft
More: Console/PC, GDC

In 2005, there was a major epidemic that affected a vast population of 6.5 million people worldwide, but it went largely unreported by the mainstream media.

That's because it was an in-game epidemic in Blizzard Entertainment's World of Warcraft that is now known as the Corrupted Blood Plague, or the Hakkar Blood Plague.

Even though the epidemic was virtual, it still provided fascinating insight into real-life epidemics, said Dr. Nina Fefferman at GDC 2011's Serious Games Summit.

Fefferman, an assistant professor at the Department of Ecology, Evolution, and Natural Resources at Rutgers University, explained, "World of Warcraft had at the time a really unique demographic composition. It wasn't the stereotypical gamer."

There were mothers, deployed soldiers, doctors, politicians, scholars -- people from all walks of life. That mix of people made the game the perfect venue to study behavioral trends amid an epidemic.

The Hakkar Blood Plague was a completely accidental incident. Blizzard at the time had introduced a high-level dungeon called Zul'Gurub, whose end-boss was Hakkar. He had an attack called "Corrupted Blood," which would drain hit points and make affected players highly contagious, and able to spread the debuff spell among other players.

Corrupted Blood was intended to be confined to Zul'Gurub -- Blizzard assumed that players would either die at the hands of Hakkar, which would rid players of the disease when resurrected, or that they would defeat Hakkar, which would also end the spell's effects.

But some players just ran away from the boss, taking their infected pets and infecting anyone they came in contact with. Lower-level player populations in particular were devastated by the plague. Some of those infected would even maliciously infect other guilds. "People were running around exploding into blood, dropping dead and turning into skeletons," said Fefferman.

Blizzard tried to impose a quarantine in order to contain the disease, which failed. Blizzard eventually had to reset the servers after about four of five days.

The way that players behaved in a way that a real-life population might react to an epidemic, said Fefferman. "We saw some courage... we saw some fear, suspicion of a quarantine," she said.

Blizzard's quarantine attempt also reflected how U.S. Homeland Security might react to a smallpox outbreak in a big city. And the fact that it didn't work in World of Warcraft could show how a real-life imposed quarantine might fail as well, said Fefferman.

What Fefferman learned from the Corrupted Blood Plague is that as ever, human behavior amid an epidemic is the wild card that can throw a wrench in even the most accurate, complex mathematical models that she and other epidemiologists use to predict the spread of disease.

"People are actually very difficult to predict," she acknowledged. "Behavior... that part is really hard."

Related Jobs

University of Texas at Dallas
University of Texas at Dallas — Richardson, Texas, United States

Assistant/Associate Prof of Game Studies — Hunt Valley, Maryland, United States

Lead UI Engineer — Chicago, Illinois, United States

Lead UI Engineer — Hunt Valley, Maryland, United States

UI Software Engineer


Aaron Casillas
profile image
Great article...Since real life has no "reset server" function, what would this be analogous to?

Cate Ericsson
profile image
Spanish Flu in 1918? Just ran its course, there wasn't much anyone could or did do. The only thing they can really point to as the cause of the end of the pandemic, was that doctors started treating the secondary diseases that more commonly led to death (pneumonia mostly, but there are others).

Luckily for the paying customers, Blizz did have a reset button at their disposal. Definitely an interesting article.

Tetsuji Gotanda
profile image
I'd like to think that Blizzard had to teach their servers how to play tic-tac-toe, WOPR-style, before giving up and pulling the plug lest the plague spread to the real world.

On a more serious note, I'd love to see this topic turned into a TED talk at some point, if only to underscore the importance of drawing relevant corollaries between gaming life and real life. Why spend millions trying to interview several thousand human subjects for a large cross-section of population on a topic, when a similar population database is already available in the online world?

Jeff Beaudoin
profile image
I have seen a couple of articles on this, and it has sort of been misrepresented in every one of them. (edit: I don't mean to say the articles misrepresent the theory, only that the theory is misrepresenting the situation)

iirc, the blood plague would just kill you within 30 seconds or so, so you couldn't run around spreading it. What happened was hunters would get their pets infected, unsummon them, and then bring them out at the bank, infecting everyone there, killing them all almost immediately.

Seems more like a disease bomb than an epidemic. The only "fear, suspicion of a quarantine" was that people were somewhat afraid to hang out at the bank/auction house because you might be killed by this.

Also, "Blizzard eventually had to reset the servers after about four of five days." is misleading. They reset the servers to implement the fix to the corrupted blood sticking on pets after being unsummoned, not to "contain the disease", since the disease didn't need to be contained. It wasn't rampantly spreading across the server as is suggested by Fefferman.

Maybe epidemiologists should be looking at the undead plague that Blizzard used in Burning Crusade, which functioned much more like an epidemic than the corrupted blood ever did.

Kim Pittman
profile image
It killed you in 30 seconds or so if you weren't raid geared level 60. My level 20 hunter died in like 4 ticks, but I watched level 60s walking around without a problem because they had the health and ability to heal themselves enough to overcome it.

On Hellscream, the server I was on, play ground to a halt for 2-3 days. You couldn't go to any major cities. You had to be careful using flight points. There was no help for you if someone decided to grief and ran in to infect a newbie area. We were a high pop server, with dozens of raiding guilds on Hakkar, which might have made the difference between one server to another. But I can remember hiding in fear in less populated zones watching for *anyone* and moving away from them after having to die over 30 times just to get away from Ironforge.

They applied the hotfix that made the disease only able to exist inside the instance then forced restarts on all the servers to get it to implement. But I remember gathering with others in Ironforge to "say good bye" to the plague.

As for the Zombie Invasion it was quite different, and not as realistic as the corrupted blood, because it was predominantly spread by those who had died from it. CB spread from the living. And it was designed to speed up and become more aggressive over time, unlike CB which was static in it's damage and spread. Though it is easy to see where they got the idea for the Zombie invasion.

Jeff Beaudoin
profile image
Even level 60s couldn't survive it indefinitely, because it would continuously spread back and forth between them, so they would eventually run out of life and mana. At best, you could last a minute or two before dying to it.

And that is sort of my point. Yeah, it didn't kill you immediately, but saying that you could spread it around the whole of Azeroth is simply untrue. At best a single infection started by a hunter would last for an extremely limited amount of time and would only really affect the spot where the person decided to bring out their pet. Any epidemic seeming behavior of it spreading across the continent would have been caused by people manually going to different places and starting it up, with the specific intention of killing everyone in that area (which is what happened).

Claiming Corrupted Blood acted like an epidemic is like claiming that Doomlord Kazzak acted like an epidemic when someone would kite him to SW to kill all the lowbies. It just doesn't make sense.

This isn't a model for the behavior of an epidemic, it is a model for the behavior of griefing in an online game. Claiming otherwise just shows the researcher's lack of understanding of the game's mechanics and is honestly a waste of time.

Kris Graft
profile image
Hi Jeff. Fefferman's talk was just a relatively brief rundown that gave the general idea of what happened, which I found plenty sufficient to get her ideas about epidemics across. But your additional details are totally appreciated in any case.

Regarding the plague that Blizzard introduced, Fefferman actually touched on that. She said the problem with that is that players were expecting it, and it was planned by Blizzard, so players' behavior wouldn't really reflect what would happen in a real-life, spontaneous and unexpected epidemic.

Jeff Beaudoin
profile image
Hey Kris. I appreciate the article, despite my seemingly negative attitude. I don't mean it to sound like I disagree with your having posted it. I added an edit line to my original comment to make this more clear.

I just wanted people to be aware that the conclusions Dr. Fefferman draws seem to be based on a very tenuous grasp of the mechanics in play and their effect on the game world.

Thanks for the insight into why the undead plague was not considered by their team. It is funny to me though that she thinks player behavior EVER reflects what happens in real-life.

Jeremie Jasper
profile image
BS, I was one of the people doing it. My whole guild was. We had 20 plus level 60 healers spamming heals and kept it going for hours. Don't be a Narb and talk crap if you weren't one of the ones doing it.

Bruce Racey
profile image
I suppose the 'epidemic' was interesting, but it in no way parallels real life situations, or the actions of people in real life. The consequence of such a disease in real life would be death. Irreversible, unstoppable death. In WoW, you die, you resurrected, you go somewhere else. So, when you infect someone with the plague, yes you're responsible for their death, but that death means nothing overall. People would go around essentially griefing other players by giving them the disease, people would CONSCIOUSLY go to highly populated areas to infect others.

Would this happen in real life? Would people be so irresponsible as to deliberately infect large populations with an incurable disease? Maybe the few sociopaths, but not nearly on the scale seen in world of warcraft. The final conclusion 'behaviour is hard to predict' is true enough, but this analogy is a pretty piss-poor one.

Jesus Rambal Llano
profile image
I donīt think is a poor analogy. Even different ways to see situation can lead to different decisions. What you can see as a few sociopaths could be a bunch of people with another point of view that think they are doing the right thing.

Even here Jeff and Jeremy give two points of view of the situation. In real life, with higher stakes, this could mean a bigger number of bahavior paths. Just look at how populations reacts to actual epidemies.

Mark Laframboise
profile image
People didn't just run away from the boss as reported in the article (since it would wear off before people could get outside the instance). There was a bug that enabled hunters and warlocks to dismiss their pets while they still had the disease and they'd still have it when they re-summoned them in a major city, which caused it to spread to other players.

So like many other real-life epidemics, this one was also initially carried and transmitted via animals (and demons) :)

Kris Graft
profile image
Yes, added the pets aspect to the article. And good point about disease-transmitting animals.

Stephanie Boluk
profile image
Thanks for this article and the details provided by the commenters. I wasn't aware of Corrupted Blood at the time it happened, but it's fascinating to see how the players had to react to the unexpected event (as opposed to planned game mechanics).

I agree with the comment that few people would consciously seek to infect large populations. But oddly enough there is some historical precedent in plague writing, zombie narratives aside, for humans deliberately (if not perhaps strictly "consciously") seeking to infect others. One example I can think of offhand is from Daniel Defoe's Journal of the Plague Year (1722) where he writes how those infected with bubonic plague were "seized...with a kind of Rage, and a hatred against their own Kind, as if there was a malignity, not only in the Distemper to communicate it self, but in the very Nature of they say in the case of a mad Dog, who tho’ the gentlest creature...yet then will fly upon and bite any one that comes next him” That passage always sticks with me because it's so 28 Days Later. There are also other writers who have written about the relationship between the transmission plague and reciprocal violence (e.g. Rene Girard, Antonin Artaud). So perhaps the cruel behaviour of WoW players isn't totally off the mark? :-)

Jesus Rambal Llano
profile image
I always have liked this secondary effects on games. Fallout, Farcry 2, Morrowind, Oblivion. There are deaseses or addictions that changes something on the gameplay. But canīt really remember something like an epidemy on a game.

Is any case of an epidemic deasease on a game? One intended to really spread to the player and from the player?