Heroes of the Storm

From the Bullpen: A Tale of Two Shamans

From the Bullpen: A Tale of Two Shamans

There are two basic kinds of design flow, which many designers call “top-down” and “bottom-up” design. The terms mean roughly different things between designers at different studios, or even between designers on the same team! For the Heroes of the Storm design team, “top-down” design means a design that is derived from a fantasy we want to capture, and a “bottom-up” design is a design that comes from a particular mechanic we’d like to exercise. Think of the “top” as being the “high concept” (in a Platonic sense) or “feel”, and the “bottom” as being the practice of play. I think any game requires a measure of both top-down and bottom-up styles, and we certainly use both in Hero design. To actually explain what I mean, take the subjects of this blog: Rehgar and Thrall, two Heroes who, while similar in many ways conceptually, underwent radically different design processes.


Top-Down Design


In my previous blogs, I have tried to express how important the fantasy is for making Heroes. We’re drawing from concepts of larger-than-life characters with special powers and long histories: druids and paladins, space-marines and space-paladins, angels, demons, and terrifying aliens. From these concepts we build what usually begins as a “top-down” design, going from these high concepts to figuring out what sort of gameplay mechanics we need to have to deliver on that fantasy.

Thrall is a prime example of top-down design. Thrall’s design as a powerful melee fighter came from what we knew about him: he’s got a big hammer for smashing things, a history of Orcish shamanism, and of course, our experiences playing with Farseers in Warcraft III and Enhancement Shamans in World of Warcraft. It follows that his abilities should harken back to these fantasy elements. When it came time to bring Thrall back into the roster from his long hiatus since BlizzCon 2012, we found his old kit wasn’t really matching the fantasy that we had in mind.

It took a lot of iteration time to figure out which five abilities we could settle Thrall on, given how much Thrall could conceptually do. At this point, the mechanics – the “bottom” – really helped us in identifying what Thrall needed gameplay-wise in order to feel fun and effective. So even though we started with the high fantasy of Thrall to generate ideas, we tested and locked down those ideas through the mechanics – starting top, going down.


Bottom-Up Design


On the other hand, the concept we start with isn’t always part of a character’s class, traits, or appearance. Sometimes we have an idea about how we’d like the gameplay to go: going invisible and sneaking around, facing down colossal amounts of damage and emerging alive, resurrecting your allies, or picking off enemies from long distance, to name a few. When we roughly know what gameplay is cool, we try to find a character that the gameplay makes sense on.

For a long time, Rehgar Earthfury wasn’t on our radar at all – he’s not one of the more well-known characters from Warcraft lore, so there wasn’t a lot of demand to add him into the game. We had Thrall’s old kit lying around: Chain Heal, Earthbind Totem, Lightning Shield, Bloodlust – all very Shaman-themed abilities, but they fit together into more of a Support role that we felt Thrall wasn’t cut out for (again, we were concerned primarily with a “top-down” design approach for Thrall). Still, the gameplay of this kit was rock-solid – the spells felt impactful and were really fun to play with (especially neat interactions like putting a Lightning Shield on your Earthbind Totem). To preserve that gameplay, we simply had to find a character where that gameplay made more sense, and thus Rehgar was brought in to be our Restoration Shaman.

Once we found a Hero for the kit, we gained access to the “top” fantasy again, and we were able to complete the design with fantasy-driven concepts like the Ghost Wolf Trait. In other words, even though we started with the mechanics, we eventually came full circle to the fantasy of the character we were trying to create – starting at the bottom, and moving up.


These are kind of extreme examples we’ve encountered during the past years of Heroes of the Storm development. The reality is that the design process is usually a mixture of both top-down and bottom-up styles. They both have merits. Top-down tends to generate a lot of ideas we can sift through, and bottom-up tends to immediately point us in a solid direction with a mechanic we know is fun to play. Regardless of the direction, it’s very often the case that both approaches give similar results:

  • Top-down: Suppose I’m designing a ninja character (ninjas are cool). I know that ninjas are generally perceived in popular culture to have certain traits: they can sneak around, they’re generally quick and hard to catch, they’re efficient killers, and they’re probably martial arts experts. So I may explore mechanics like cloaking, melee weapons, high damage, and some kind of escape option.
  • Bottom-up: Suppose I really like the gameplay of sneaking around and waiting for the perfect time to strike, or I think the game would be better with a character that filled a role like this. In order to make my strike effective, I’m willing to trade survivability for damage, making for some edgy, tense play. What kinds of characters fit that description? A ninja? A Protoss ninja? Zeratul?

By exercising both approaches to design, we get a more and more “complete” version of the Hero: a moment of gestalt where conceptual fantasy and compelling mechanics come together to make the player. That’s when we know we’ve found the fun.


That’s all from the bullpen this week! Share your thoughts with us below, and be sure to keep up with John by following @BlizzJohnzee on Twitter. What’s more, our Heroes of the Storm Closed Beta phase is set to begin on January 13, so be sure to keep an eye on the official Heroes site to catch all of our latest announcements. See you in the Nexus!


Next Article

Featured News