Talking elderly Egyptian women and sexy Japanese men with Overwatch hero designer Geoff Goodman

Geoff Goodman has lots of notes and prototypes for heroes you'll never play.

With sniper hero Ana now established in team compositions, arguments still raging over Genji’s OP deflect and Hanzo’s generous hitbox, we thought it was time to sit down with Blizzard's Geoff Goodman and talk about the fast-paced, often nebulous process of hero design in Overwatch.

We also wanted to spend some time talking about accusations of cultural appropriation in game design, and to get Blizzard’s side of those stories. Here it is then; the prototypes that are just ideas floating around, how the ideal world of Overwatch is shaped not only by its heroes but the voice of the community, and how decisions seemingly set in stone can be reversed if it's what the game needs.

VG247: Geoff, could you start by giving us a brief overview of your role on Overwatch?

Geoff Goodman: I’m in charge of hero design and balance. Everything at the beginning of the character until after it’s launched, then balancing it going forward.

VG247: So, this is more about what the heroes do, rather than how they should look…

I’m in charge of the raw gameplay aspect, although it’s interesting that when you get going working on heroes that things get pretty close together. You know, maybe we should have this ability because it would be fun, but also because it makes sense for the style of the character or the lore of the character. It mixes more than I expected.

"We wanted you to play multiple heroes on both sides but just recently we reversed that. That was a contentious point that we were fighting for, but we were never saying we wouldn’t change it."

VG247: The team seems unusually responsive to community feedback about how the heroes look and behave. Was this a strategy decided for Overwatch before launch?

As a company we take all kinds of feedback from people all the time. You may see other companies say “We’re listening” but, you know, don’t necessarily post and say something. We’ve definitely made more of an effort to make sure that we’re responding directly. Myself and Jeff Kaplan are always talking about getting on the forums as much as we can. Sometimes it’s better to let a conversation take its normal course. It’s difficult to know when to jump in and say something, but we try to talk about our plans, what’s going on and what we’re thinking.

VG247: Is there anything that the team, or you individually, feel especially protective about? For example, the community can have its say, but at the end of the day McCree won’t lose his flashbang capability.

[laughs] There are definitely some things that we’ve drawn a line in the sand on. Something that happened even before the Beta, when people just saw the videos, was that people really wanted a radar or mini-map, like Call of Duty. That’s something we knew from the beginning that we don’t want to have. You don’t want to spend a long time looking at that, because it becomes a large part of the game. When you intend to build a game that way it makes a lot of sense but when we build a character like Widowmaker with an ability to detect enemies, we did that ‘in world’ rather than have a radar that pops up on your screen. The other one that’s pretty common for players coming from a Call of Duty or Battlefield type background is sprint. We wanted to build all these different heroes, and all these different abilities, certainly sprint could be one – and ended up on Soldier 76 – but part of the point is to make all the heroes and characters feel a lot different in the way they play.

We wanted you to play multiple heroes on both sides but just recently we reversed that, and when you go into Competitive Play you can only choose one of each hero on the same team. That was contentious point that we were fighting for, but we were never saying we wouldn’t change it. It was more like, well we want to see how it feels, but we thought maybe it was better if we change it in this case.

VG247: Is the hero limit now set in stone, or could we still see a revision for Season Two?

We’ll have to see. It’s still pretty new, what we have out there. So, we’ll have to see if it goes horribly awry or if it works out better. We found, both anecdotally and looking at stats, that having a free pick of heroes is a problem at the top end when you hear feedback from the pros really wanting the hero limit, but also the players at the lower end, the newer players, were having a problem with no hero limit.

It wasn’t so much “Oh man, two McCrees!” or “Two Tracers? this is over the top!” it was “There’s five Torbjörns! I don’t know what to do!” When you play the game enough and understand what’s possible with all the heroes you will eventually figure out what to do, but you do need that knowledge in order to do that.

It’s easier to employ a strategy like that than to understand how to counter it. So, it ends up being a problem more than we anticipated at that low end as well.

VG247: After six years in development, by the time the game ships, you’d imagine that a lot of this stuff would have been ironed out and you were happy with how it all balanced out. Does there ever come a point where you just want to (politely) say, “This is our show. This is our team and these are our plays, this is how it gets done. You need to trust us a little bit.”?

We don’t want to squash the conversation. Although sometimes ‘conversation’ is a nice way to put it [laughs] – sometimes it gets pretty heated. Those threads come and go, and I don’t want to jump in there and be a cop and say “No! And that’s the end of it” because even if I don’t agree with a lot of it and I feel that they don’t see everybody else’s perspective, even with all that going on, you can look at the thread and find interesting things, like “Ah, that is a good point to think about.”

Even if it’s not going to change my mind or anything, I still find interesting perspectives from people that I didn’t necessarily know. So, it’s still useful in its own way.

We just don’t want the forums to become this place that a lot of people don’t want to go anymore because it’s so bogged down with the negative side of things. We don’t want to push people away. That is kind of a big concern.

"A lot of times heroes can start, like Pharah did, as a bundle of gameplay mechanics and we have tools internally that I can make some really early prototype stuff. But there are other times where it comes out of the art, like Winston."

VG247: Is it a concern that, the more you play into the hands of the eSport and MLG guys, that you make Overwatch less accessible to the majority of people that just want to go in there and have fun?

Yes, definitely. When we take feedback, especially from a limited pool, we make sure that we understand its effect on people playing that don’t have visibility on that kind of stuff. So, in regards to the hero limit thing – that was definitely a concern. That’s why we kept it for Competitive Play but you can still have it in Quick Play if you want to.

We have a Custom Game feature right now but it’s not used very well, and a lot of that is because it’s just not very accessible to people playing random games. If I put together a Custom Game, how do I even advertise that? How do people know to join? If I want 11 friends to come join my game, I have to invite them manually. That’s not a fun approach. That’s a system, for example, that we want to look into and make a larger part of the game.

VG247: David Gibson spoke at GDC about how the heroes project their role through their appearance. Can we talk a little bit about how heroes are designed from that perspective? For example, Pharah started out as nothing more than a rocket launcher and a jetpack.

A lot of times heroes can start, like Pharah did, as a bundle of gameplay mechanics and we have tools internally that I can make some really early prototype stuff. This is just to get it into the game so that we can try it. A lot of characters come that way, and a lot of times we’re thinking about what’s needed from a gameplay standpoint.

But there are other times where it comes out of the art, like Winston for example. He came out of art, basically. I would never have considered, like “We really need this space gorilla!” [laughs]. But, they just drew some awesome stuff and it was like “Woah!” At the time we had mostly human soldiers, essentially. But that was so out there. I didn’t even think in that sort of space. We sort of all fell in love with it. It was like, “We’ve got to build a character around this!”

Working with characters, the gameplay is paramount to us of course but, for example with Reinhardt, he can do a lot of damage even though he is a tank and his role is not one where he supposed to be killing a lot of people. But if you look at his whole suite of abilities, he does a lot of damage if he can land his abilities. Even his Fire Strike, his ranged attack, it pierces and it does a lot of damage. It’s just that it has a cooldown on it and it has a wind up time before it fires off and it has a slow travel speed.

Everything that Reinhardt does is very deliberate, is very slow and does a lot of damage. It’s all reinforcing that he’s big guy, he’s really hard. He’s not going to go like Genji and pull a sword to slaughter everyone, but you don’t want to get hit by him. His charge is a mobility tool in some ways, but also basically just kills somebody he hits.

VG247: Was Reinhardt an answer to the need for a colossal guy in armour? Or was he originally a need for a hero that carried a huge shield?

The shield was our original concept for the character. Reinhardt was our first Tank character. But going into a game like this, what does a Tank mean? It’s not like we’re building World of Warcraft. The original concept was, like: “Well if the guy had a huge shield and he walked around with it, he’s pretty much going to block a lot of damage for the team.” That’s where the concept of those barriers come from. Your team can treat them like they don’t exist, but they block a lot of ults from enemies. They can eat a lot of damage if you put them up in the right spot. We ended up using these in a couple of places. It was kind of a Tank mechanic at this point, generally.

VG247: How did Ana evolve: was she always going to be Pharah’s mother; was this hero always going to be female, an older character?

She started as another gameplay prototype. Just before we shipped the game we started work on a post-launch hero. We had a full roster of 21 launch heroes at that point. People were playing the game and we were getting response about that roster. We were, like “What are people really missing now? What are people most excited about?”

We came pretty quickly to the conclusion that we needed another healer. A lot of teams were running two healer set-ups. People didn’t always want to run with Zenyatta because he was less powerful. We saw Lucio and Mercy in every game, basically. We definitely needed more variety there.

One of the things we hadn’t really done was to aim your heals. Lucio is an aura, Mercy is a beam. Zenyatta is a little bit fire and forget. What would it mean if you actually had to aim?

"Ana started as a prototype called the Alchemist. She would drink this potion that would make her jump really high, and really fast. But it was only for one or two seconds, a huge burst so that you could hit it and super-jump somewhere really far. It was sort of interesting, but it had a lot of problems."

She actually started as a prototype called the Alchemist. The syringe gun that she has now was always there. One of the early abilities of the Alchemist was that he or she or it would drink this potion that would make them jump really high, and really fast. But it was only for like one or two seconds, a huge burst so that you could hit it and super-jump somewhere really far. It was sort of interesting, but it had a lot of problems.

There was also an idea where that potion would provide a really long heal over time. You would drink it then walk into battle and survive because you were healing the whole time.

At the time that we’re thinking about a prototype, trying different gameplay mechanics, we’re also thinking about what is the actual character. Is it a brand new concept we’re working? Is it an Omnic, or something? Ana was a character that we always wanted to do. The problem was that her backstory is that she is this incredible sniper. We already had Widowmaker, so it was kind of hard from a gameplay standpoint to put another sniper in. But we figured out that, actually, this alchemist character would work pretty well with Ana.

It basically went like a rocket from there and never stopped.

VG247: How long was the Alchemist character around as a concept?

A couple of weeks, I would say. We’re working on so many different things at once. It’s kind of like spinning a lot of plates. Whenever I can get a prototype at a decent stage, try to get a playtest going, have everyone jump in and get their feedback. You can do that around the team, I’ll send a playtest out to everybody to get feedback across the board.

VG247: How many new heroes are currently underway, at a stage where the whole team has an understanding of what they might be working with, that could become something?

Man, it’s hard to give a number. We have a great bunch of tools to work with, to make different prototypes. I probably have three or four prototypes that are at various stages, and a lot of them probably won’t go anywhere.

I have way more just sat on paper, it’s all like a one off. “This could be a cool gun!” and I’ll write that down and it just sits there. “This would be a cool ability somewhere!” Sometimes when I’m looking at a new character I’ll remember that, “Oh yeah that gun. That was a cool idea.” It runs the gamut from raw notes and gameplay snippets to some early prototypes.

We find it really useful to make a lot of stuff like that, to have a lot of different prototypes. We can do these playtests, and people can get really excited, and generally there will be a lot of excitement around one prototype. We’ll say, “This seems like it’s going in a good direction”, “Let’s take this out.” There might be one cool idea from a prototype, even if the prototype itself didn’t really work. Maybe that character didn’t really work, but that one ability is pretty cool, it could be used somewhere else. It’s very fluid.

VG247: How much are views on sexualisation and cultural appropriation a consideration for the team when designing heroes? Recently there was a very compelling Tumblr post from an Eyak Indian mentioning how they loved the Pharah Tunderbird skin, for example.

What it really comes down to is, you know, this is Blizzard’s first game where it is set on Earth. It was really exciting for us. We’re not making Azeroth here, we can do all these places around the Earth. There’s a lot of interesting culture, we can take a lot from all these places. That’s always been our outlook. We’re trying to highlight very positive things.

I read that post that you’re talking about. It was interesting from their perspective that “This is cool – we don’t get to see our art around anywhere.” We hear a lot about Ana too, where people are saying that we don’t normally get to see elderly females in games, let alone Egyptian elderly females. That’s not a thing in games.

"We’re not trying to offend anybody. We’ve shown that if we’ve overstepped our bounds somewhere that we’re willing to correct things. It’s a line we’re trying to be careful of."

We’re not trying to offend anybody, of course. We’ve shown that if we’ve overstepped our bounds somewhere that we’re willing to correct things. It’s a line we’re trying to be careful of, obviously. We definitely want to highlight the awesomeness of all the different cultures everywhere where we can. It’s part of what makes Overwatch, Overwatch.

When we start to do something that’s from a region that people are not intimately familiar with, initially, it’s a lot of research. We want to have accurate portrayal, to get everyone excited about what they’re seeing. There’s a lot of research on the art side, and on the cultural side of things – what people are about.

VG247: Specifically regarding the female heroes, most of them are very slender, they’re quite sexy and healthy looking. Whereas, on the male side, you have Roadhog and Junkrat.

We are planning to release a lot more heroes. I don’t want to give anything away, but there’s definitely more to come in all aspects, so...

VG247: A sexy male hero?

[laughs] We have Hanzo, he’s sexy.

VG247: Finally, which of the 22 heroes have been the most fun to create so far? Who are your favourites from both a personal and professional standpoint?

Oh that’s hard, it’s like they’re all my kids. Genji was really interesting for us. He was a very difficult challenge because he started out as a character with a sword as his primary attack. It was incredibly difficult to pull off. He actually started with invisibility too, so he could run up to you and backstab you. It was pretty awful! [laughs] Given all his history internally, how much we wanted this ninja thing and kind of where he is now… granted some people have issue with his power level, but he really plays that vibe really well. He doesn’t have stealth and he does pull his sword occasionally, but he has that agile, jumping off the walls thing really down and it’s come together really well. He is perfect in that role.

And in the same way I really like Reinhardt, he feels that way too where – you know Genji is very is fast, he throws three stars really quickly, whereas Reinhardt is slow, and hits hard. These characters came together really well for what they’re representing.

Overwatch by Blizzard Entertainment is available now for PC, PlayStation 4 and Xbox One.