Game designers Fearless, Ghostcrawler and all held Q&As on the official forums. Here are all theÂ questions they answered:
WHO AM I?
Hey everyone, Iâ€™m Jo Graylock. Before I came to Riot I was an RPG designer for most of a decade, as well as Jiu Jitsu instructor and metal radio show host. The vast majority of my developer experience comes from my time at Obsidian Entertainment. Lots of awesome people making awesome games there, but I couldnâ€™t pass up a chance to work on my favorite game.
I try to play a ton of everything, but favorites are LoL, Fallout 1 and 2, Xenoblade Chronicles, Tactics Ogre, Dino Run, FTL, and Risk of Rain, with healthy amounts of D&D and Warmachine at the table. When not gaming, Iâ€™m usually doing lots of tabletop painting, playing volleyball, or spending time with family.
I currently own: 2 cats.
WHAT AM I DOING AT RIOT? WHAT CAN YOU EXPECT NEXT YEAR?
I started as an individual contributor at Riot, working on the Sona update before moving into systems work, first on the Jungle Update for 2015 before following up with Cinderhulk. After that, I moved to lead the Lead Design spot for Gameplay Systems. Iâ€™ve been lead for the Juggernaut items and the 2016 Season work so far. Iâ€™m really looking forward to keeping this season exciting throughout the entire year.
WHAT AM I GOING TO TALK ABOUT?
Iâ€™m going to be here to talk about all the systems stuff in LoL. This means items, the map, the jungle, and all the other rules level stuff in the game. A lot of our changes can be pretty easy to miss, but also a little mysterious or obscure. Iâ€™m really excited to bring this stuff into the light, as well as the reasons behind why we change the game. Iâ€™m also really happy to talk about how we work as designers, both in what our day to day looks like, and some stories about how we think about problems and overcome them. Riot designers were super inspiring to me before I got here, and Iâ€™d love to give back a bit of my time to pay that forward.
WHAT WOULD YOU LIKE TO HEAR FROM ME?
Letâ€™s talk about stuff, for realzies. The openness and honesty of Riot was a big reason to work here, and Iâ€™d like to keep that tradition going. 🙂
EDIT: Going to meetings for a bit, will be back around dinner time.
EDIT2: Alright, heading out for some dinner. Thanks for talking everyone! Look forward to a lot more!
We are looking to slowly reduce power, but the base goal is one we want to keep. Just need to pull levers to get it into balance.
Current thought is that max CDR shouldn’t be something that mages just get by accident, and should be something that actually has meaningful trade-offs. ADC builds are still very much in flux and we’re actively still looking into tweaks to make sure that going into Reaver is powerful, but has a few more trade-offs as well. 5.24 has a small change to delay the rate at which ADC’s get the full CDR, which should help with how free that stat is currently. Totally possible we’ll need to do more there still.
We have some more communication on this coming next year. TLDR is that these items are all answering the question we wanted them to answer, but some could still get some small changes to adjust how well they answer those questions.
We’ve talked little bits about these guys, but the crowd is pretty small. Given that we have some pretty large blocks of champs that are in very difficult spots, my guess is that the work on this group might be a ways off. That or they might get some love from the projects targeted towards other groups. That tends to happen a lot when it comes to systems projects, just because everything is so interconnected.
Cull is on watch, and probably needs a small buff. Staring items are extremely sensitive to small changes, so have to tread carefully.
At Riot we don’t split up game designers much besides Designers and Tech Designers, with the main difference there just being the ability to check in code. In terms of things like systems design or mechanic design, these are competencies that each design has some ability to do, but will be stronger or weaker depending on who you talk to. At Obsidian, I moved between being a Cinematic Designer, a Combat Designer, a Systems Designer and even a Technical Designer for one project. Each company has their own ways of understanding their teams, and I think that’s smart for each company to set out who they want and how specialized they want them to be.
And for the last question, there is a place at almost every studio for designers that don’t have tech educations, definitely including Riot. I have a film background before I got into gaming, and I’ve worked with Designers with every sort of education, from Art, Philosophy, Architecture and Literature. I’ve also worked with plenty of incredibly talented designers who did not complete a degree. The most important thing for a game designer is that they are great at understanding and making fun games. Some people learn things that help them do this in school. Some people mod and some people figure this out along some very strange paths.
When I think about the work that I did, really everything has been some form of system, apart from Sona. For example, even though I was the Cinematic Designer for Alpha Protocol, most of my work there was building out the game experience of the story. AP wanted the cinematic sequences to still very much be an interactive part of the game that gave players the space to express themselves, experiment, and feel like they could “win” the story by getting to be the character they wanted to be. A lot of the decisions we built into how players picked their dialogue options and how those payed off and showed progress all went back into making sure the cinematics could be that gameplay experience.
Basically, being a systems designer was simply a matter of taking on a big project, and then consistently thinking about the entire impact and interactions of the project. It also really helps if you’re comfortable working on the nuts and bolts level stuff, knowing that when you do your job well, players might not even notice a lot of what you’ve done. Great systems design is often like great film editing, in that many of the best examples are amazing because of the subtle things that they do that most people won’t (and shouldn’t need to) notice.
I’m really not an expert, or even particularly familiar with the other game modes. Before I started working at Riot, I was actually pretty into Dominion, and would play a few games every week, to test out champ match-ups and builds in the faster setting. Once I got to Riot and found myself on a team that is entirely focused around SR, I pretty much only play SR to make sure that I stay as current as possible and am testing and experimenting as much as possible in the space that lets me build awesome stuff for players.
Sorry I don’t have more deep thoughts to throw in here.
I’ll be completely honest: we have neglected them, especially Dominion and TT. We view the heart of League to be SR, and that’s where we spend most of our development effort. We aren’t happy with that answer however, and we have some ideas for how we can give Dominion and TT a little love. We’ll talk more about this soon. We are unlikely to add new permanent maps in the future. Often, they lack the depth of SR, which means they are popular at first and then slowly lose players over time. In some regions where we have fewer players, it can also cause long queue times by sub-dividing the bucket so much.
I worked on Neverwinter Nights II, Alpha Protocol, Dungeons Siege III, and South Park: The Stick of Truth, along with some projects that never quite saw the light of day.
Map geometry is an element we generally keep fairly fixed, while letting champs determine how they want to interact with it, given that it’s going to very likely be stable. A lot of systems work is about making sure we have a solid rules base for champions to interact with and navigate, and the map is a very base layer set of “rules.”
I didn’t understand where the decimal point for move speed values were supposed to go, so I had a Sona that pretty much just blinked where you clicked for as long as her E was up. Caught before it got into a playtest, but very fun for a few minutes.
The marksman items had a lot of revisions, some of which had some extreme power, and of course some of these lined up with champ iterations that were very strong. Combine that with the fact that we have some high skill marksman specialists in the main playtest group I’m in, and we had a few playtests that felt like an alternate game mode where the goal was just to run as fast as you could away from the marksman. Not good for the game, but very entertaining to see the roles flipped so hard.
I think our bounty changes, coupled with the game pacing changes to towers and minions are very exciting, personally. I care a lot about making sure the game is still fun, even when you’re losing, and these changes make it much more clear what you should be doing when a team is down, and cut down on some of the worst game states to experience as a losing team.
WHO AM I?
My name is Greg Street, Riot Ghostcrawler. Before working at Riot, I was a lead designer on World of Warcraft for 6 years. Before that I was a lead designer on Age of Empires for 10 years. Before that, I was an oceanographer, which is where the crustacean avatar came from. I was an army brat who lived lots of places, but I identify most as a Texan.
My favorite game of all time is Dungeons and Dragons. I have played for 30 years, in all 5 editions. For video games, Iâ€™ll just list broad families of games that I love and that have influenced me: XCOM, Final Fantasy, World of Warcraft, Chrono Trigger, Diablo, Doom, Civilization, Legend of Zelda, Dark Souls, Baldurâ€™s Gate, Planescape: Torment, Mass Effect, Metroid, Grand Theft Auto, and Okami. Outside of games, I enjoy travel, reading about history and science, and being outdoors. My last research project as a scientist involved collecting live, radioactive alligators. I have been attacked by animals as diverse as: Portuguese Man-o-war, Roseate Spoonbill, Southern Stingray, Rock Rattlesnake, Coatimundi. My favorite animal is the Mantis Shrimp.
I currently own: 4 dogs and 2 cats.
WHAT AM I DOING AT RIOT? WHAT CAN YOU EXPECT NEXT YEAR?
I joined Riot in 2013 and I am now the game design director for League. I describe my job as a combination of three things: I’m ultimately accountable for the design quality of League, I work with the other leads to determine long-term strategy and our big bets for the next year, and I manage the design department. I work a lot on projects such as our Season 2016 plan. I help answer things such as what kinds of champions should we be making, what do we want the pace of the game to be, and what new features does League need.
WHAT AM I GOING TO TALK ABOUT?
The best kinds of questions to hit me with are about our design philosophy or core design values. The kinds of questions I can best answer are “Would Riot ever consider this Big New Feature?” â€œWhere do you see League in 5 years?â€ “How does Riot balance for such a wide range of skills?” “How much counterplay is acceptable?” “What were Riotâ€™s goals with preseason?” “Is League too snowbally?” “Does Riot make too much change to the game?” “How many champions can League support?” I don’t directly balance champions or design new items. I’m fine if you blame me for bad balance, but if you want to know our thoughts on a particular change or the state of a champion, you’re better off asking the Rioters who are working directly on them.
WHAT WOULD YOU LIKE TO HEAR FROM ME?
If you know anything about me, you hopefully know that I strive to be really transparent with players and that I have very thick skin. AMA.
We feel overall that the ramp up in complexity on champs (the extra passives and double passives and so on) isn’t a great trend to stay on. It’s a challenge when you’re in the trenches, because you want players to get excited about each new champ that’s released, so we keep trying to outdo ourselves. If we released [next champ] with Vlad’s Q, I think we’d all be disappointed. But that’s the short term view, and in the long term view, we don’t want to end up at a point where champs have 4 passives and 20 abilities and so on.
I think by around 300 we’d need some kind of seasonal roster or something. At the current rate we make champions, we won’t have to worry about that for some time. The game can handle a lot more than it has today IMO.
We agree with just about everything you said. “What is the problem with Vlad’s Q?” used to be a question on our design test for potential new hires, but we removed it because it was too easy. Thematically, he’s fine. He needs better abilities. We’ll get there, potentially with the mage roster update this year.
We strive for evolution, not change for the sake of change. That doesn’t mean we always nail it, and it’s an easy trap for developers to fall into, especially folks who haven’t been making games for very long (which is almost every Rioter). It’s a compass heading.
Morde was not change for the sake of change. Morde was not working as champion and we didn’t think it was possible to deliver on his current direction in a healthy way. He really needed a change the same way Poppy and Sion needed a change. Were there other changes we could have made that might have worked out better? That’s entirely possible. Was the community right and we should have listened to them? In 20/20 hindsight, that happens all the time. I’m trying to illustrate there is a difference between a bad change and change for the sake of change.
Change for the sake of change comes about when you see devs say things like this:
- I’m bored. I think I’ll try something crazy.
- I never envisioned this champ as X. I’m going to try Y because that fits my personal impression.
- Z hasn’t gotten any changes for awhile. It’s his turn.
- I don’t know what will work, so I’ll try this.
Change to add depth to the game come about when you see devs saying things like:
- This pattern lacks counterplay so we need to add an opportunity for the enemy to do something.
- We can’t increase this champion’s winrate without their pick rate skyrocketing and players seeing them every game. We are going to have to change some of their abilities.
- This system is something solved that players don’t really think about anymore. We need to get interesting decisions back in.
- Our values or goals are X, but this champion isn’t meeting those goals. We need to make changes to meet them.
In general, the changes we try to make are strategic, try to deliver on goals or values, try to open up options or decisions for players, and try to fix real problems.
Big games are very expensive to make. This makes studios (and more importantly, their publishers who put their money up front) reluctant to take big risks. There have also been some spectacular failures of big games that were super high concept that just didn’t resonate with players and lost millions.
If you look at indie development, you will see a lot more risk-taking and even spot some of the next new trends. You’ll also spot some coming out of other regions, especially China. (I realize asking you to download a Chinese mobile game is a non-trivial ask.)
Conflicted purpose. If you have an ability that is say both a shield and a heal, it can be an interesting decision about when to use the ability. You may need the shield now but you may need the heal even more a few seconds later. It is possible to screw this up, which was his point, but it’s possible to make it work out well as well.
He is my boss, by the way, so I’d love to tear down his arguments in a way that could publicly humiliate him, but I actually agree with his list for the most part.
Pros: Great for “feels” feedback. Good at catching bugs. Good as a point of reference for players to discuss on boards and Reddit.
Cons: NA only. Players don’t know how to participate. Bad for balance data (sample size is so small that matchmaking suffers, and players aren’t generally in tryhard mode anyway).
We are fine with hybrid champions or rule-breakers, but we need to establish that is our intent and not let it just sort of happen accidentally because the kit is overloaded. We didn’t really have a good framework for this before say 12 months ago when we started considering rosters.
We think her kit is fine. Not every champion needs to have a ton of mechanical complexity so long as they still have strategic complexity. For Sona, positioning (when to hug and when to not) are supposed to matter enormously.
I haven’t socialized this with the team, so they might disagree, but back in the day, League made the conscious choice not to go with the Strength, Agility, Intelligence model that Wacraft 3 originally used. The idea was that items were a bit more agnostic and any champion could use them. Was that the right call? I’m not sure. While there are still a lot of champ agnostic items, we also are jumping through hoops now to make sure e.g. Marksmen items are attractive to Marskmen but not necessarily to mages.
So interestingly, our data suggests that games are no more snowbally than before the preseason changes or all of last year. We realize there is a common perception among players that it is more snowbally, so we are trying to unpack that and figure out why the discrepancy exists. Games are definitely shorter, but a shorter game that isn’t more of a shut out would actually be a pretty good place for the game to be.
We also would like to see 10 different ADCs get played. Balance and strategic diversity play into that for sure. But it’s not the only factor. Players (of any game, not just League) really hate to be wrong, especially in front of strangers. That can lead to a situation where everyone wants to know “Who should I pick?” by which they mean “Which champ should I pick that is currently considered the best so that nobody will get mad or make fun of me?” In other words, the perception can sometimes drive the reality.
The perception that there is a best and second-best ADC can often cause players to choose those champs, even if their winrate would be better with a champ they’ve played a lot or really enjoy. Seeing what the pros pick can also have a huge trickle down effect on ranked queue champion choices. The difference between a “strong” and “weak” champion can often mean the difference between a 55% and 45% win rate. Think about that. That means you’d win approximately 10% more games if you picked the right champion. If you played 10 games, you might win 1 more because of your champion pick. Your individual experience with that champion can have a far bigger impact. Overall, League is a pretty well balanced game. That doesn’t mean we can’t do better, and we will strive to do so. But we’re not talking about champions who are twice as likely to win a game as another choice.
Yeah, it’s pretty much just the gems.
Seriously though, a lot can go into the determination of which champion update we work on next, including whether we have potentially good ideas for how to fix the champion (thematically, visually and mechanically). We do have someone working on Warwick, because he desperately needs it, but it’s too soon to tell how or when that may pan out.
Generally speaking, we do two kinds of changes — things that have a chance of a long-term fix and things that are band-aids just to tide things over until we can get to a long-term fix. That said, Mundo is somewhat simple, but I don’t know that he needs a major update.
Yasuo’s winrate generally isn’t all that high because his skill cap is insanely high, one of the top 2-3 in the game. That is the challenge for why he is tricky to balance moreso than the double passive, IMO. We’d have to ask the champion designer who worked on him the origin of the double passive though.
It’s less about free stats and more about whether a champion is nothing but a bag of stats. If your win rate is high because your stats are higher and not because you make good decisions, that’s the kind of thing we want to fix. We have fewer of those champions this year than we did last year.
I tend to play in bursts like lots of players. I am playing fewer games now because I have been spending so much time on Fallout 4 and the Bloodborne DLC. Before I came to Riot, I’d say I played mostly Support (I played a priest in WoW) and ADC. I have tried to double down on improving my weaknesses rather than my strengths since I have been here. Jungle is by far my weakest role. I need to spend more time there.
We want the early items to be a little safer because you aren’t locking yourself out of potential later items. You may not have enough information early in the game to know which items (or even in which order) you’ll want at the end. This means some amount of similarity among the builds themselves. There probably are some opportunities for those mid-tier items to branch out a little more.
I think there is a tendency to want to design a champion that gets a lot of YouTube videos made because of flashy plays. Similarly, in a game with 130 champions, you can’t just offer a kit of skill shot + shield + steroid. I’ve said before that I think the concern that too many of our champs lately have been on the more complicated end is very fair and something we want to combat. You can have champions that make strategically interesting decisions (or even big plays) without them having double passives and lot of gigantic tooltips.
Easily ten years. It’s not going anywhere. I think the biggest gains we have to make are outside of the core game loop. We can’t ever take our eye off that ball, but it’s pretty fun already and we don’t want to bury it under tons of complex systems. On the other hand, some of our systems outside of the game are pretty weak. For example we have focused a lot on fixing player behavior problems but not as much on helping you play with your friends.
It was a good proof of concept for the roster project. We’re happy with Garen and Darius, less so with Skarner and Morde. Overall, hitting 4 champs is probably not enough to feel like a roster was really updated. However, compared to Marksmen, Juggernauts barely even had a function in the game, so in that sense, we feel like it was successful.
We don’t want to be in a perpetual arm’s race. We have some fear that if we did 10 today it would be 12 next year and so on. There is some evidence that giving each player one ban in solo queue might be a better experience. We have some changes with the new team builder (with still just 6 bans) and we will see how that plays out first and then evaluate.
This is a great question but I don’t have a great answer. Maybe another Rioter who has been here longer will share a good story with me that I can then relate to you guys. Going into a new region is a fairly expensive commitment in that we have to localize a whole lot of text and VO and then either set up servers and matchmaking ourselves or find a regional partner who can do so. We have to be convinced that we can develop a sizable player base in a new region for it to be worth it. That is usually the biggest barrier to going into a new region. There are some interesting quirks though — we can’t play on Korean servers because Korea requires you to have the Korean equivalent of a social security number to play. (We were able to play when we visited Korean PC bangs though).
I agree we could have more, and we’ll probably add more over time, but not toooo many more. Imagine a game with 10 “Skarners” and 10 “Skarner crystals” all over the map.
We don’t make the changes just for those mains for a few reasons. Maybe they main that champion for reasons we don’t want to encourage. Maybe there are a bunch of potential mains out there who aren’t playing the champ but might with a good rework. Ideally we don’t alienate mains, and we do always target our survey questions towards both mains and non-mains.
It’s a cool space but it increases the amount of information you need to play the game by a lot. Consider the Poppy rework — you have to know to step on her shield. Maybe you’ll stumble upon that at some point but maybe you won’t.
We look at “play as” first. If nobody wants to play the champ, it doesn’t matter if they are fun to play against. We definitely look at how frustrating they are to play against once we really get into development.
I’d love to see a more casual yet still competitive mode around some of those less-played maps.
It depends on your target. If you’re looking to change our development culture or something, that’s pretty hard. If you want to give feedback on a champion change, there are a lot of ways to do that. Posting in the threads (often the PBE threads) where the developers are asking for feedback or discussing upcoming changes is a good way to focus your efforts. Just don’t assume that there is some kind of magical lever that you can pull to guarantee your suggestion will get incorporated. That’s just really hard to do when you have millions of players with their own suggestions. 🙁
It’s fine if they are in the same space but have different play styles. It’s probably not possible for us to come up with 130+ different strategies.
I think I mentioned this earlier, or maybe it was Twitter or ask.fm.
I think it sucks when we do an update that is okay but not awesome. The reason is because it generally gets a champion into a healthier play space (so they are no longer at the top of the “on fire” list) but we haven’t made them awesome either (and because they are no longer at the top of the “on fire” list, they may not get touched for awhile).
We’d like to hit Cass again when we look at mages in 2016. I can’t promise because it’s impossible to predict how the projects will work out. I really wanted to update Volibear for the juggernaut update, but it just wasn’t coming together.
WHO AM I?
Iâ€™m Mark, but you can call me RiotScruffy here. Iâ€™m a totally obsessed video gamer (in the best possible way), and games like DOTA, WoW, LoL, and pretty much anything with the Star Wars brand have occupied the best years of my life.
Iâ€™ve been a few places around the game industry before Riot, most notably I worked at Treyarch on Black Ops 2 right before I joined. They taught me some great stuff about how to make everything feel OP, and how to really sell a fantasy.
I currently own: 1 cats.
WHAT AM I DOING AT RIOT?
Thereâ€™s kinda two sections here for me, the past and the future:
The past (and present): Iâ€™m the Lead Designer of the Champion Update team, which has worked on many of your favorites (and unfavorites Iâ€™m sure) like Sion, Poppy, Gangplank, Fiora, Juggernauts, Marksmen updates and many more. I have a TON of history with this team, having helped build it up from the ground when I started at Riot almost 3 years ago until now.
The future: Iâ€™m going to be moving over to be Lead Designer of the Champion team. A new set of challenges and a really talented new group of people there is really exciting. Building up the next generation of new champions and hopefully surprising you all with some great new innovations is my goal.
WHAT AM I GOING TO TALK ABOUT?
Champ-Up has grown and changed so so much over these past years, and I can talk about so many things. â€œWhy did we start doing group updates like Juggernaut/Marksmen?â€ â€œHow do we decide how much to preserve or redo on a champ-up like Poppy?â€
Or even less specific stuff that is more related to our champion/game design philosophy like: â€œHow have we evolved our strategy for making interactive gameplay without overusing skillshots?â€ â€œHow do we build up a new champion from concept to completion?â€ â€œHow did we go about categorizing our champions into archetypes so that we could better understand them (see juggernauts)?â€
Iâ€™d also like to bridge the gap – when I can – of how our general game design practice can carry over into good character mechanic design and vice versa. It might be cool to do some topics that explore how to apply our champ design principle to other systems or even games altogether.
WHAT WOULD YOU LIKE TO HEAR FROM ME?
Anything about Champ design or Champion updates is definitely on the table for me. Get creative, Iâ€™m open to whatever cool topics you all would like to talk about.
- Yorick – Both super unhealthy for the game and incredibly unsatisfying/lame
- Taric – So much promise with so little delivery. He has a ton of upside in making him more interesting, cohesive and higher quality
- Urgot – incredibly unclear fantasy and unhealthy gameplay
- Talon/Akali/diana – the trio of assassins with the potential to be great, but lacking so much in gameplay depth and interaction that if they ever are strong its the most frustrating experience
- Xin Zhao/Aatrox/Pantheon/Irelia – The grouping of fighter types with overly one directional gameplay patterns. This results in a “stat check” where they press their buttons in the right order and the rest is up to numbers not skill. (fiora used to be in this camp)
I don’t see why not. It’s a cool way to create new feel and gameplay – see Vel’koz geometry Q
We want to solidify his role as a tanky protector type support with armor affinity. The gameplay is pretty nearly 100% and we are working hard on the new visuals. There are some really great surprises in his new kit/visuals but I will not spoil those.
No comment on Ao Shin sorry, this was just a way premature reveal and there are a lot things in the pipe on the new champion team right now.
A lot of the work I’ve started on the Update team will be finishing in 2016, so I will still have a loose hand in it, but my main focus will be running and creating awesome improvements and innovations for our new champion team.
I think there are a bunch of new fantasies that I’d be excited to explore. I’m especially interested in pushing for really really innovative ideas that will completely break the mold of what we currently consider champs can be.
My pet project – The “branching future character:” A good warrior comes in contact with a corrupted weapon, his/her choices throughout the game will either keep them pure or cause the corruption to take over. By the end of the game you have transformed into one of multiple possible kits.
This is a good question. Statikk and Feralpony have been posting about morde and he is in their team’s hands. They will be the better people to have the long term discussion with.
My personal opinions:
-Solo morde needs love, he may not be the “main” role but he has a lot of room to become more playable
-W in solo has had some improvements, but is still a pretty lackluster ability without an ally, this could be a good area for us to look into again
I think the original thinking was that morde was oriented towards a solo lane so clearly before, that it would take a lot to make him even remotely viable in the duo lane. In retrospect we clearly overestimated how much we would have needed to do there.
Whenever possible we get the original champ designer to consult heavily on the rework project, but it’s usually a good idea to get someone fresh to actually own the project. (attachment too strong can be a problem)
I think that point and click spells have a bright and consistent future in LoL, they provide a clear, satisfying, and reliable output. The key is making sure that the champ kit outside of the point and click offers enough places for interaction and counterplay. Take GP Q for example, which we kept point and click because the new E is so interactive.
If you have any questions, feel free to ask me atÂ @NoL_ChefoÂ or e-mail me at email@example.com.