Two weeks ago at 2017’s Game Developers Conference in San Francisco, Riot Games design director Greg Street took the stage for a panel titled “Balancing League of Legends for Every Player, from Bronze to Bengi.” You can watch that full panel here.
Before Street’s presentation, though, we had a chance to sit down and speak with him about the current state of League, what’s coming next and what Riot has learned from recent struggles to balance the game.
Note: This interview was conducted shortly before information on the upcoming Galio character rework was revealed. You can find further details on what’s new with Galio from Rift Herald’s coverage.
Rift Herald: What will you be speaking about at your GDC panel?
Greg Street: I’m talking about game balance, and specifically strategies for balancing a game with such a diverse player skill level. The title of the talk is “Bronze to Bengi,” because that is the entire gamut of competitive League play. I was actually invited by my old boss Rob Pardo, who sits on the board at GDC. He was like, “Hey, we’d love to hear a talk on how you guys balance League.”
RH: It’s one of those topics that — looking at League and looking at the number of characters you have and the combinations of items and everything, balance has got to be one of the most daunting elements of working on a game like League.
GS: And players are never happy with the state of balance.
RH: Of course! One thing I find interesting about Riot’s approach to balance is that you’re doing it constantly. It seems like there are differing schools of thought. Some developers release maybe one patch every three to six months that’s like a really big patch where they change a lot.
GS: Yeah, like Dota 2 has one patch, and then nothing for a while.
RH: Exactly, or even other types of games like Counter-Strike or Hearthstone.
GS: Or the Smash Bros. community has pretty much accepted that it is static, and it’s up to the players to evolve things. The developers aren’t going to step in.
RH: What is it that led to Riot’s strategy of patching constantly and at least changing little things very frequently?
GS: It goes way back to the founding of the studio. Mark and Brandon played a ton of games together, and they got frustrated at — I don’t even know what game it would have been at the time. It was a real-time strategy game, I think.
GS: Maybe, although StarCraft was updated fairly regularly. But they were playing games where the meta was solved. Everyone knew the right way to play. And Brandon and Mark felt like there was so much more there if the developers would just patch it.
So they decided with League, they wanted something that was — the word we use a lot is “curated.” The developers are always going to be paying attention, and we’re going to try to react to things very quickly when they happen. It’s just a different style. You can certainly create the type of game where you interfere as little as possible.
RH: In the current season of League, just based off what’s going on right now, what people are talking about on Reddit, what you guys are seeing yourselves internally, what do you think is the biggest balance challenge that you are facing at the moment?
GS: What the players would say is that it’s ADCs 2017. I don’t know if that’s actually, holistically the real biggest challenge or not. I think we are in a situation, for a number of reasons — like the strength of junglers, strength of tanks, changes to masteries, changes to items — we’re hearing from a lot of players that ADCs, particularly in the laning phase, don’t have enough agency. That’s something we’ve been looking at for a while.
I worry that it even goes back to auto-filled supports. If you’re getting an auto-filled support, then as an ADC, you probably don’t have someone who really is experienced trying to keep you alive in the early game. Plus the first tower bonus really encourages four-player attacks on bottom lane, because you can get two kills while you kill a tower. That probably all factors into it. It’s kind of a complicated thing. That’s what we’re hearing about the most from players right now.
Holistically, I would say it’s also things like: Are gold income levels where we want them to be? Are items doing the right things in the game? Those are kind of more systemic changes we’d like to look at long-term, less about patch-to-patch balance.
RH: Let’s talk about some of your bigger updates. Do you view mid-season and off-season class updates as an ongoing cycle of something you’re going to be doing indefinitely going forward? Or are you hoping to address these outlying issues and champions, and then at some point, you won’t have to do class updates anymore?
GS: The initial intent behind the roster project was that we would go through and hit every class, and then stop and be done. The mentality was that we could go in and change one juggernaut, but if we decide to add weaknesses to this one juggernaut, all the sudden the others would still be overpowered and people just wouldn’t play that one. The only way to fix it is to go in and do all of them at once. That was the original promise.
I would say we haven’t exactly delivered on that promise as we’ve gone on. The updates have been more and more small character updates. We’ve gotten a lot more into abilities and player satisfaction than we were originally going to.
So we’re going to do at least a couple of more class updates with tanks and divers. Beyond that, I’m not sure we’ll keep that strategy. We’ll have to see how these next couple go. That will take us through another year, at least.
RH: Camille is League’s most recent champion. It seems like she was introduced with the idea of having a character who’s relatively easy to come in and understand.
GS: Compared to some of the recent champions, sure. Like Aurelion Sol and Taliyah, who ended up having a really high skill curve.
RH: Exactly. Are there any lessons that you learned from Camille about designing champions like that? Especially with how she came out and was considered pretty overpowered by a lot of people?
GS: I mean, she was. We missed the mark.
RH: Well, do you think the nerfs have been enough? Did you learn any lessons about creating that type of character?
GS: I don’t know that it’s something special about Camille. With Ivern, for example, we knew that this was going to take a while for players to get used to. It wasn’t a surprise at all that it would take a lot of Ivern games to get him. In your first game as Ivern, you’re probably going to lose horribly, because you have no idea what’s going on. So we didn’t think that she was going to be that steep, but I don’t think it’s anything special with the way Camille moves around the map or shields or anything.
When we did Bard, he came out undertuned. It took a while for him to recover. It took a while for players to consider playing as him. He’s in pretty good shape now, but since then we’ve been really careful not to undertone. With Camille, we definitely missed the mark, and as a result we’re going to change the way we do things internally a little bit.
It used to be that the champion team would own a champion for three patches after launch and then hand it over to live team. We’re now going to do that hand-off immediately, so that the live team can get up to speed with the areas we’re concerned about and be able to jump on something really quickly. Hopefully we won’t have a target that’s off that bad again. If we do, we can respond very quickly. Because even internally, once we were like, “Yeah, she’s overpowered,” it still took a while to get changes out there.
RH: One question I have that’s not about balance but more generally your approach to characters, and particularly story: League has always had this, but especially recently with new characters like Camille and Aurelion Sol and Taliyah and even going back to Jhin, you’re seeing more and more of a community build up around being really passionate about these characters and wanting to know their backstories. You’re also seeing a lot more people pushing for representation of a broader spectrum of characters, which I think Riot has been doing a great job with in characters like Taliyah.
GS: Or like Illaoi, who has a different silhouette than most of our female characters.
RH: But the one thing that I have seen players talk about that they haven’t seen in League of Legends is LGBT representation. Obviously that’s tough to some degree in a game where you’re not presenting a lot of story. There’s not room to have a lot of romance, necessarily. But I’ve especially seen people in the League community talking about it more since Overwatch did the Tracer reveal. Is that something that Riot is thinking about or wants to do?
GS: Yeah, we definitely want to do it. We owe it to players and, I think, to the world to do something like that. What I don’t want to do is be like, “Okay, team, next character, whatever you do, has to be lesbian.” I don’t think we’ll end up with something good there. The character has to be conceived from the very beginning — which, I don’t know if Tracer was or not. I assume she was. But from the beginning, it has to be that character’s identity. I’m sure we’ll do it at some point. I don’t know which character or when it will happen.
RH: And I’m sure it’s something where you have to figure out the right method to get it across. That was part of it with Tracer — there wasn’t anywhere that they could naturally get it across until that comic. It’s not like they added a line in the game where she shouts, “By the way, I’m not straight!”
GS: You know, both League and Overwatch are global games. There are countries whose laws around things that we consider pretty normal at this point in the U.S. are not the same way. One way you can get around stuff like that is by having some of the storytelling outside the game. You can be like, “Look, within the game itself, we don’t talk about it. Outside the game, it’s something else.”
There are also times when it’s worth having that battle. We just have to be careful, because it’s not necessarily just about a game company taking on the government of some other country, which may be very exciting for players. It could end up with players in that region not being able to play a game, which may be striking a blow to those players. We don’t want to indirectly hurt players.
RH: Moving over to esports, the 10-ban system has obviously been really warmly received in competitive play. It’s been awesome for me as a viewer, because I feel like there’s such a wider variety of characters being used.
GS: Less so in bot lane, but yeah, mid, top and jungle have all been awesome to watch. And I hope it keeps up throughout the season.
RH: Yeah, it’s been great. With that in mind, when do you expect that system to move over to ranked? Do you have any idea what form it will take?
GS: Yeah, we’re working on it right now. It’ll be pretty soon, one of the next few patches. As with many things we’re working on right now, there’s this chicken-and-egg problem of, do we make it for just the new client? Or do we build it in both clients? I think for the pick-ban stuff, we decided let’s just build it in both clients, or we’re going to be waiting too long.
The goals of the 10-ban system are a little different in ranked. A lot of the changes to pro play were about the viewer experience. It’s fun for viewers to see a diversity of champions. If anything, it may hurt pros a little bit, because they have to play a wider variety. We think for regular players, even in ranked, that pick-ban is much more about their lane. “I don’t care, I just don’t want to play against Yasuo. I’m going to ban Yasuo.” We think 10 bans just works well, since every player gets to ban something, probably something in their lane.
Because we don’t anticipate the kind of strategic level discussions of how you’re building your comp and what does the enemy comp look like — which pros certainly engage in — in solo queue, we think it’s fine to just have the bans be simultaneous. There’s going to be less discussion. Anything that feels like organized play, like an actual tournament or five-person ranked, we would probably make it work like the pro system of banning one at a time and discussing your strategy.
I also want to add that it is definitely a goal to not make champ select take any longer. If anything, we want to speed it up a little bit, which simultaneous bans will help.
RH: One thing that’s near and dear to my heart: Galio is my favorite character, and I know you are doing the rework soon. Is there anything that you can tell us about the rework at this point? Any hints about what your goals are for the Galio rework?
GS: I equate Galio as very similar to Poppy. There was nugget of something cool there that was never delivered on. Poppy went from being like a D-minus to an A-minus where she is now. I think the Galio is the same way. He’s a champion that a lot of people have forgotten about. He’s just a gargoyle dude.
The new conception of him, even thematically, is really cool, and I think players will get excited about it. And then the gameplay is nice as well. Sometimes we do an update where we’re not sure if players will like it or not. Like Yorick, we said, “We want him to be a little more aspirational than a gravedigger,” and some players were like, “But that’s why I like him! I want to be a gravedigger!” We were nervous about that one. I’m not at all nervous about Galio.
RH: As long as he’s still a gargoyle, I’m good.
GS: He is a gargoyle! He looks great.
RH: Are there any champions in the current meta, the current patch, the current season, that you think are really strong but players haven’t caught on yet? Something that you’re waiting for someone to crack it and figure out that this character’s really great?
GS: The answer I use in my talk is that Anivia’s winrate is really really high, and I don’t think players know that yet. She’s a bit of an acquired taste. I don’t know that she’ll become super mainstream. I think she’s fun to watch. I like control mages like that, because they do neat things to the geography of the map and aren’t just shooting each other. We’ll see.
Hopefully when Zyra and Malzahar are under a little more control, we’ll see some interesting things happen with support. We’re seeing a lot of weird picks in support, even in ranked, and again, I’m wondering how much of that is auto-fill. You get auto-filled support, and you’re a mage player, and you decide, “Well, I’ll just play my mage.” It’s not necessarily a bad thing, it’s just different.