- Hi, I'm Greg Street. I'm the lead designer of League of Legends at Riot Games, and I'm here to help solve some unsolved mysteries.

Q: How many times are you gonna have to rework Ryze before he's balanced and fun to play?

- Honestly, the answer is as many as it takes. We're not proud of the fact that we had to rework Ryze so many times, so you know, that's on us. We have not been able to find the right way to keep him as the, like, super powerful mage that can do crazy spells that really change the course of the game and still have him be accessible to mere mortal players without having him dominate pro play. We don't have any immediate plans to change Ryze. His latest iteration seemed like it's in pretty good shape, but if it's not, we'll have to take another swing at it. And I will apologize when we do so.

Q: As the game's evolved, Riot has done a lot more champion reworks. Which one has worked particularly well, which one has not?

- Overall, I think our champion reworks have been pretty successful. As a player, I really enjoy the Fiora update where we made her feel a little more like she had to actually-- You know, she's a fencer, she's supposed to actually outduel the champions she runs into.

- [Game Announcer] Shutdown!

- I thought the Poppy update was pretty good. Now she has some pretty consistent stuff around, using her shield to hit people, or hitting them with her hammer. The Talon update where he kind of delivered on the fantasy of a parkour assassin guy who can run along walls has been pretty nice. So I would say the ones that haven't worked out as well: Skarner, this scorpion dude. We updated Skarner, but I feel like along the way we lost a lot of the Skarner mains. So, not particularly thrilled with the way that one worked out. We know that we can't make everyone happy. Even changing Urgot from this weird spider dude to this half-mechanical cyborg with these cool legs. Even then, I know that we lost some people, what they loved about this champion. We'll never make it to where 100% of mains come over to the update. We just want to keep enough of them, and we want players to say like, "Yeah, I still recognize that as the champion that I've spent so many games with."

Q: One big mystery is that the people want ARURF a lot more often. Why do you only break that out once or twice a year?

- URF is a game mode we made many years ago. It stands for Ultra Rapid Fire, and the idea behind URF is that everything is kind of League of Legends on steroids. All the cooldowns are down, mana regen and things like that are very fast. The idea is that you can go into lane and just blow all of your abilities really quickly. These characters weren't designed to be played in this manner. It's also a lot of fun. If we have it on all the time, people quit playing League of Legends all the time, and that would be not great for players or for Riot. So we're trying to have it both ways by turning on URF, including ARURF, but trying to see how many times we can offer it in a year. Like, what's the right frequency that players feel like they're getting their URF fix, but they also stick around to the game more.

Q: The most recent support champion to be added, Pyke, doesn't play like a traditional support. Are more hybrid heroes part of the plan as the game continues to evolve?

- Traditionally, supports have not always had a ton of agency in the game. They couldn't earn gold the way other players were doing because they're not getting a lot of champion kills. Pyke is an experiment to say like, "Look, much like Bard, who we made a few years ago, can we have a support champion that plays in really different way and doesn't just stay in lane, kind of babysitting the marksman, but can get out and roam a little bit?" And Pyke can go roam and kill enemy champions. And we hope that that will bring new players to support who didn't like the role before but were just kind of missing the right way to play support.

Q: Why does the summoner spell Ignite kill Brand?

- So, the joke here is Brand is a mage who's always on fire. Ignite is a fire spell. The real answer here is that the gameplay has to win out over the thematics. And whenever the thematics of the gameplay collide, we have to do what's healthy for the game. We thought it made sense that Brand was immune to Ignite. We could go that route, but I think they'll probably end up being more confusing to have Ignite work on some champions and not others.

Q: The question is how do we balance making a game that's exciting for LCS viewers, with ever-increasing complexity, but also accessible by someone trying to play for the first time?

- League of Legends exists as a game that millions of players play, but it's also a professional eSport. We make life hard for ourselves because we use the same set of rules for the eSports part of the game and the normal player part of the game. But League has this really virtuous cycle where you watch a really cool match, and you wanna go play the game, and when you wanna get better at the game, some players go look at the pros to learn how to do that, so we wanna make sure they're playing the same game. Philosophically, the champions that work the best are the ones that are kind of easy to learn but really, really hard to master. We call that skill floors and skill ceiling. Even if you've played 500 games with a champion, you still learn things you can do differently. And you can see that when the pros play. They can pull off some crazy move that you never thought was possible. They somehow managed to eke that out of these characters. We have characters like Master Yi, Garen, Yasuo, who are really hard for beginning players to counter. So making sure that both new players and very, very experienced players have answers to these characters, I think, is one way to tackle that.

Q: During your time at Riot, it seems that champion mobility's increased dramatically. Have we reached peak mobility, or are you going to continue to push this?

- I mean, I really hope we reached peak mobility. The question's kind of getting at the fact that we have a lot movement abilities now. We have a lot of ways for champions to zip around the map, close distance with enemies, or quickly back away from them. Now you can jump over walls and teleport and do all sorts of crazy stuff. I don't think that's necessarily bad. I mean, part of the reason we add mobility is because it's fun. It gives you a lot of agency. It gives you a lot of ways to control the narrative of the game. If you get into trouble, you can run away. It gets to be a problem when you feel like you can never get a kill because everyone's always running away. Or when you can't plan anything out in advance because champions can be right up in your face in a way that you could never predict or understand what's happening. I think as long as it doesn't get like that, and as long as the older champions don't just immediately fall off because they can't compete, there's nothing wrong with mobility overall.

Q: Which champion has your favorite backstory?

- The answer I'm gonna do here is one that might surprise you, which is Miss Fortune. She has a really cool motivation. She really wants to, kind of, take over Bilgewater and be, kind of, the queen pirate of this area. Which will probably end in her doom. So, we have this really interesting tension between what she wants, which is to be in control, and what she needs, which is probably to just get the hell out of Bilgewater and get her life back.

Q: What is the ideal number of total champions in League?

- I think it is a fixed number. I'm not gonna say that the number's infinite. And I don't think in 10 years we're gonna have like 500 champions in League. I think that would be insane, and I don't know if I'd want to play that game. On the other hand, I think it's more than the 140-ish we have today. At the rate we add new champions, we're only adding, like, five or six new champions a year right now, so it'll be a while before we're even talking about 200 champions in League. We can continue to grow the pool as long as grows kind of slowly and over time.

Q: Talk to us about how character design informs a champion's abilities. How do you signal to players what a given champion is all about?

- When the narrative and the thematic line up, it's just magic, and those characters feel really, really good to players. They're some of the champions I think players really like to play the most. Gnar is a great example of a character whose abilities and thematic don't necessarily line up. When you see a screenshot of Gnar, he's just like cute little Yordle. He looks like Pikachu, and the designer who made Gnar says that he looks a little like Pikachu. His gameplay has really got a lot going on. And even if the abilities don't look super complicated, the way you play Gnar, you don't have a lot of control over when he kind of Hulks out and turns into big Gnar. It requires a lot of experience to get good at this character. So I'd say that's an example where the thematic of like, "Oh, yeah, can't wait to play that character," and the reality of, "Oh, I'm gonna play a lot of games before I even decide if I wanna stick with this character as a main," really don't line up. Jinx, who I think her abilities Jinx, who I think her abilities and her design line up super well. She's crazy, she's a loose cannon, she's kind of unpredictable and takes a ton of risks. The designer tried to make this happen at first by doing things like, "Oh, well, if she's crazy, that probably means she can't really control her abilities." But it wasn't fun. It was a really frustrating ability because you don't know what's going to happen, and it may not actually hit the targets you want. I think that's a really good example of where the thematic and the gameplay line up perfectly.

Q: Some players were surprised by how severely Baron's damage output was nerfed in a recent patch. How do you decide what the proper balance is internally, and how do you know when you've gotten it wrong?

- League is a PVP game, and Baron is a PVE element. Meaning, he's kind of like a raid boss. You go and you get your team together, and you have to kill him. That's not super interesting. His abilities aren't that complicated. We don't really want them to be more complicated. What we want out of Baron is interesting teamfight opportunities. So, when you see that another team is about to go get Baron, you have to make the decision about, "Can we try to stop them? Could we even try to make the steal?" Those are where the really epic moments come. So, overall, what we were trying to do is make the encounter a little more interesting. Not necessarily make it harder, easier to deal with. We just didn't want, you know, we weren't trying to build it up to be a PVE encounter, and what we thought here is making him a little sturdier rather than doing so much damage probably was gonna deliver on the experience we wanted, which was having a chance to go there and interact while another team was fighting him.

Q: Do you have a favorite map?

- Summoner's Rift is League of Legends. That is the map for League. We are not interested in League having like tons and tons of maps, and you decide on which one you wanna play on in any given game. We do have alternate game modes. In particularly, Howling Abyss. I think we made an argument that, aesthetically, it's got a really good vibe. Rather than the mystical forest that is Summoner's Rift, it's this really cold, uninviting bridge, chunks that are falling off. I think it has a really good aesthetic, but in terms of gameplay, it's all about Summoner's Rift.

Q: How long will you typically play test a change to the game before it goes live?

- For something really dramatic, like Runes, or like the Clash feature, or like the upcoming Ranked changes, we play tested a lot because it's the kind of thing you don't want to get wrong. These could mess up the game if we're kind of cavalier. We were doing the mythical plus or minus five AD. We definitely will still play test it to try to make sure the change does what we want. But we know that that testing is gonna be nothing compared to millions of games that get played as soon as the change goes live. So we don't just give the entire burden to players to solve, but we do understand the reality that the sample size we're gonna get when the change goes live is much, much larger than anything we can recreate internally. If we do something like we did a few years ago, where we tried to change the way queuing works we did with dynamic queue, and it doesn't land well. That's much harder to fix. That's not the kind of thing we can just push a day to change and have every player pick it up right away. So hopefully we have solved some of your unsolved mysteries, and we look forward to you unsolving some more mysteries for us to then solve and have the cycle continue. Thanks a lot.