On Champion Design: An Open Letter to Riot

When I started playing League of Legends, the meta was constantly changing. We would see both teams with junglers and teams without junglers. We would see many champions being played in innovative ways that the original designers probably never thought of. Believe it or not, triple lanes were a thing, once upon a time.

This is only because the game was not being seriously studied at the time.

If you take the current pro players, and give them the game from season 1, you'll find they will tunnel on a very narrow meta.

I'm a bit late to the party, but I'll throw in my 2 cents as well. I'm sure most people have said something similar already, so read at your own peril.

TL;DR: Meta is reinforced to provide better experience for new players and casual spectators. Most changes are small to prevent community backlash and to preserve integrity of competitive circuit. Fixed meta and stable competitive circuit make it easier to cement League as a legitimate sport/esport.

Long answer:

• Riot has a vested interest in trying to standardize key gameplay elements in League. The fact that the meta exists makes it easier for spectators who might not be active or skilled players to understand what is going on in a game. The fact that they don't understand why the meta exists isn't relevant for the ability to enjoy watching a game. This is important because Riot actively seeks to legitimize League as an esport (or sport, depending on who you ask), instead of just presenting it as a competition. They are seeking to establish a paradigm that rarely changes and if it does, then only in ways that are simple to understand such as swapping top and bot lanes against a stronger 2v2 matchup - and even this was discouraged by changes to the towers themselves. By seeking to remove creative strategies like the old version of duo jungling (instead of jungle-buddy system where one solo laner abandons lane vs a lane swap), roaming and non-standard lane compositions, League's early game is effectively standardized. One team will usually come out ahead from the early game, and it's easier for spectators to follow how and why the mid game will focus around the winning team pressing their advantage with the real strengths their champions have. Most games end around the mid game or early late game, so late game being more swingy is rarely an issue in professional league games. Key point: Standardized gameplay is easier for spectators to follow, and easier for newer players to learn. They don't need to understand why the game is played that way to enjoy it as long as the game is actively designed to work best that way.

• League approaches patching differently than Dota does, preferring to maintain the integrity of the weekly LCS games. DotA often has few huge balance patches where nearly every hero receives changes, and more often than not those changes are buffs. Only the truly overpowered are nerfed in DotA, and even then they usually get some buffs to compensate (dark seer being largely the exception with his vacuum cd nerfs). League has far fewer changes at a time, but changes are made more often. The big element that holds back dota - style massive changes is that League doesn't have a "stable/newest" patch split, and Patches are rolled onto the competitive circuits as quickly as possible - keeping in mind parity between teams having played the same number of games before the new patch hits. This also preserves parity between the games the pros play and the games the rest of us play. DotA only keeps up parity in Captains Mode, where certain heroes are disabled for months at a time until they've been tested enough and Mr. FrozenAmphibian is happy with their balance. Key point: League prioritizes parity between pro and casual games. Pro games shouldn't drastically change quickly, because it can disrupt integrity of competitive circuit. Thus, casual games don't change drastically either.

• League is and has been trying to distance itself from Dota since around season 2 or so when the meta was beginning to cement itself. There are many reasons why Riot would want this to happen, and I won't speculate on which ones are correct. That being said, League started to create its own identity, trying to shed the idea that a team would only have to rely on one late game ADC to close out the game. Instead, many roles were changed so that they couldn't solo the entire enemy team no matter how much ahead they were. Teams essentially have to work together to some degree to close out a game unless there is a huge gap between player skill on the two teams. In DotA, many carries are designed in a way to allow them to 1v5 the enemy team or take out the entire enemy base quickly if they are ahead (Anti-mage, Void, Medusa, Terrorblade, Tiny, Gyro, Naga, Lycan, CK, etc). This means that the design space for League champions is inherently smaller than it is for Dota heroes. League champions are designed according to a "late game fantasy" of what the champion should be able to do, and balanced according to that ideal. This further limits the potential of the secondary roles of champion, because of the fear that there could be significant power creep if a champion is able to fulfill their two intended roles at the same time through some funky itemization. Many players would see this as a strength, because alternative build options can often be fun, but it's also a nightmare for the balance team to work with. The prime example of this are the champions who were mainly considered supports when the changes to supports hit. In general, support champions lost damage and gained scaling utility (for instance, Janna could no longer clear a creep wave at lvl 6 with one tornado, but her shield got a scaling AD buff). This was because supports started gaining a lot more gold than previously, and could build items instead of just buying wards. The fear was that supports would be far too strong later on if they could retain their damage ratios, and wouldn't really be supports but rather secondary mages or bruisers. This had the adverse effect of effectively nerfing all the existing supports, pushing many mid lane staples to the support role. The most prominent of these would be Annie, who was not considered a support before the support changes around season 3. Key point: The removal of many secondary build paths was done to promote/force the game toward late game teamwork instead of trying to establish a strong carry or semi-carry who could single handedly close out the game. Individual champions matter less, but grouped champions matter more.

• By designing champions for specific roles, the design teams have a clear frame to work with, making the process easier on them. The individual champions are what attracts people to play these games in the first place. In general the communities around these types of games are very off-putting, so the main thing keeping the players playing are the character they can play as. Riot puts a big emphasis on visual design largely for this reason. The gameplay designers work with an existing image of what the character should be able to do, so they disregard themes that don't make sense based on the visual design of the champion. That's why you don't see Kalista with AP ratios on her abilities. That's also partly why Yi's and Rengar's crazy AP builds were nerfed. Visually they don't fit the image of mages using arcane artifacts to enhance their abilities (the other reason being balance issues - AP rango was pretty crazy). While these kinds of alternative builds can be fun for more experienced players, they challenge the key identity of the champions in question. When the identity of a champion is challenged, then people who liked that champion for the visuals might be put off from playing that champion. Imagine if Ahri was changed so that she could be played in the ADC role for instance. Ahri's key identity is being a beguiling mage/assassin. If she suddenly starts to tear buildings apart with tiny projectiles while her huge abilities do almost no damage, then many players or spectators can be confused about what's going on. Same for AP Yi. Why is a wuju samurai doing almost no damage when he swings his sword, but suddenly bursts for a huge amount when he disappears? Why does he just stand in the middle of a teamfight while being nigh unkillable due to his meditate self-heal? Visually the champion stops making sense when you build him as AP. When the AP build became the dominant one on him, he was changed to preserve his identity. Key point: Champion design focuses on establishing the visual identity of champions and their ideal late game "fantasy". If a secondary build is stronger than the primary one, then this identity can be challenged. If a champion looks like a mage but is a warrior, then players who don't know that will be disappointed when they purchase that champion, leading to a worse customer experience. Visual design is important, because it sells champions and makes the $$$ in a F2P game designed around purchasable characters.

That's all the points I wanted to make. They might give you perspective on why things were changed. That being said, I personally agree with many of the points you wanted to make. I also enjoyed old league (season 2 - season 3, and a bit of season 4) more than new league. The rest of this post is going to be a rant about the things I don't like. Go ahead and skip if you don't want to see the tantrum.

• I think that champions who are designed primarily for one role are bad design in general.
• I also think that reinforcing the meta by changing stable elements on the map (like top lane turrets being different from bot lane turrets) is silly.
• I would like to see actual roaming supports and trilanes being viable.
• I'd like to see the return of running 3 solo laners and duo junglers being viable again.
• I would like to see the support champions in particular having powerful alternative roles they could be played in.
• I would like to see more variety and strategy in picks and bans, and more flex-picks who could be used to trick an opponent into thinking about the wrong lane matchups.
• I would like to see real melee carries and not just assassins.
• I would like to see revive being reintroduced as a useful summoner spell.
• I would like to see flash being removed from the game completely.
• I would like to see the decline of the mobility creep.
• I would like to see more powerful point-and-click CC or instant CC to counter mobility.
• I think resourceless champions in general are bad design, and would like to see those champions designed around a resource, but given more power to compensate.

I heard that murica was the country of freedom, good thing that meta enforcing doesn't exist oh wait.

Remember when we could take mele carry instead of ranged carry? me neither

You forget the fact that Riot has never been able to balance both playstyles of the AD/AP champions. AP Trist/Yi/Sion were cancerous, awful champions that neglected half of their kits. AP Ezreal does not see buffs because that's not the most effective way to play him, and giving AP Ezreal buffs would send AD Ezreal over the top.

Also

League of Legends was dynamic and full of surprises once upon a time.

That was because nobody knew what they were doing back then.

It's a lot easier to balance a game when you create a champion for a certain purpose and meta. If Riot were to create champions that weren't designated to a particular role than this game would be a hell of a lot worse in the balance department.

DISCLAIMER: So I've got some kinda heavy stuff to say about this topic, and if anyone bothers to look me up, they're gonna think I'm just not qualified to talk about it because my stats make me look like an abysmal casual LoL player. I don't play ranked, and I probably never will. I have 34 wins in normals. I mostly play bot games. The reason for this is that I am a perfectionist. I don't believe in doing things "for real" until I'm sure I'm doing them "the best I absolutely can" and I take this approach to normals, which is why I play so few of them. I know I'm not great so I don't try to take my game to the next level until I'm sure I deserve to be there. That said, I have been playing since season 3, and almost every one of my 34 wins came with an off-meta pick that I got flamed for constantly, kicked out of team builder for picking, and sometimes absolutely dominated with; Support Veigar.

Now I believe this makes what I have to say even more salient in this post because as we are now aware, it took Riot and the community at large 2 years longer than it took me to figure out why support Veigar was good. Nowadays you don't have to look far to find a thread where someone just casually says "well Veigar deserved nerfs because support Veigar could shut people down too easily." What's completely ironic about this is that I'd bet any amount of money those same people saying that would be the exact people telling me that Veigar is useless as a support and kicking me out of their games before they saw some Koreans an/or LemonNation prove to the whole world something that I figured out 2 years ago. My biggest problem is, no matter how good support Veigar is, I personally just suck at League of Legends so there was no way I could possibly prove that to anyone =p

Anyway a lot of what you say is pretty spot on and I agree with it probably 90%. But I'm afraid I have to explain to you why Riot does what they do and why it's never going to go back to the way it was.

Here goes.

Having champions with tons of options for their builds is arguably a very healthy and in fact a sound tactical option in a game like LoL, but the more complex the game gets, the less possible it is to balance unless you start defining things in particular ways. Remember, it's pretty obvious that absolutely no one wants to play "League of Not Even Close-to-Balanced."

You mentioned several times in this thread about how "older champions" (Morgana specifically) are the ones who still have variable options, and that the newer champs are shoehorned into a pre-defined meta (which you don't like.) The issue is, and weirdly I think Riot is only just now starting to figure this out, Riot does not get to decide what the meta is. In fact they never have. The meta is, was, and always will be determined by the players.

It developed because the millions of players who play this game every day (which is billions of games a year, maybe more) did math, way more math than Riot could ever do in the same time-frame, and they came up with the simplest solution that was also the most rewarding. And I'm not even claiming most of them did it explicitly or on purpose. Just the collective experience of playing billions upon billions of hours of League of Legends led people to a collective understanding over time, of what "works best." Finding the path of least resistance is just something human beings do, and that's true no matter what you're talking about.

If you have a problem that needs to be solved, a goal that needs to be reached, a game that needs to be won, if people work on it, especially if TONS of people work on it, and share that experience, a prevailing method will emerge, and that method will emerge because it's the best one. If and when (I can't stress the importance of that phrase in this context) that method is improved upon, it will it change again.

Now there are people out there, myself included, and you as well obviously, who just want to question everything. They ask; "Can it be done better?" And surprisingly, sometimes the answer is yes. The problem is, once people learn that someone found a better way, once that enters the collective consciousness of the masses, it replaces the old way and becomes the stagnant thing (in this case, the meta.) The larger issue isn't that Riot stops things from changing so that they can keep things the way they are forever, it's that they prevent the new adaptations that they never foresaw from becoming the prevailing thing and therefore too stagnant (and it's important to remember that only unbalanced things will become "stagnant" at this point.) Now, I'll grant you that this is a slippery slope and we don't always know what's "too powerful" and what isn't hence the subjectivity and argumentativeness on the topic.

You made a really good analogy to this by bringing up how sports have set rules that haven't changed in a really long time (stagnation, or enforced meta.) But I don't see Soccer and Football getting less popular. So in the case of an eSport, when the developer can introduce new elements (unlike in real sports,) they need to have a design in mind for those elements. When characters don't fit that design you get a soccer player who's allowed to use their hands, or a running back who's allowed to make a forward pass whenever they want. This obviously can't happen in real sports so when it happens in LoL, Riot is within their right to put a stop to it. If anything they should've seen it coming first, but it's not ok to say "well it's in the game now so it can't change" which is effectively what they did in a large number of cases at least in the beginning. They used to only nerf things because "this was clearly just too powerful, our bad." (And sometimes they still nerf things for that reason because what was "too powerful" the whole time just went unnoticed aka. Tristana.) Now they nerf things because "the community is doing things with this that are outside the rules for the "position" that character plays, and we need to put a stop to that so we can preserve the integrity of the game we're expected to present." Now it might sound like fun, and sometimes it is, but "anything goes" is no way to run a successful business, or a successful sport.

Right now there are a few off-meta things that happen which you mentioned, lane swaps, the support roaming with the jungler early, but are we ever going to get away from the 1-1-2 + jungler meta. Probably not. It's just too mathematically efficient. (Interestingly though, and this is a point for another post, but the next character to come out, Bard, is supposed to roam, and not just sit in bot lane, and I'm personally pretty excited about that. But there's a large chance this could fail and Bard will be next to useless, though I personally don't think he will, but I digress.)

If you want proof of everything I've said may I offer up Team Builder? Team Builder was released during the time when Riot used to constantly pepper their own forums with "we don't enforce the meta." They constantly said they didn't care how you played and fun was the name of the game, and that they didn't want to allow people to be able to "call their roles" before matchmaking because they thought that would "enforce the meta." But the community was so toxic toward meta-breakers that Riot was forced to find a solution. That solution was Team Builder which allows you to not only call your role but also your starting position and hailed it as the thing that would solve every problem with the game. "Tired of getting flamed for wanting to break the meta? Wanna try duo-top? Think all-mid is a good idea? Just queue up for that in Team Builder so you'll find people who agree with you and never have a toxic game again." And what happened? If you didn't queue for the 1-1-2+ jungler meta, you'd sit in matchmaking for hours, and even when you did it could take 15 minutes to find a support. Not because Riot was enforcing the meta, but because everyone else was. Now, the only reason Team Builder is better than Blind Pick is so you can guarantee the roll you want and avoid people trolling, and thank you Rito for that!

Eventually Riot learned the valuable lesson from this experience. It's nice to include everyone, but if you aren't catering to the overwhelming majority of your customers, you're not doing your job right. It simply wasn't cost effective for Riot to release champions without telling you what they're supposed to do and without having a clear design for which well established roll they fit into. The ability to do something else with them still exists, and there are people like you and me who will still try to do it, but League has reached a point in it's lifespan where nurturing the "anything is possible" idea is just no longer wise. Some people still believe it, and often times these people do find new things that work, even when they're told "this champion only does X." But it's now gotten to the point where it's in Riot's best interest to limit these possibilities in order to preserve the thing that's gotten them where they are today; people who want to do everything along the path of least resistance, which means no thinking, no testing, no failure, just look up a rune page and item build that someone else figured out and climb freelo all day.

Now I wholeheartedly agree with you, if Riot becomes too rigid and inflexible on issues like this very one, they will lose the creative and innovative players, which could ultimately doom them, but that will take a long time. I also just think they expect us to take their approach with a grain of salt because they know that we think on that next level already and the message they're sending was never designed for us anyway.

Riot has said that they like the current meta and don't want to see it changed (more or less). With that in mind, I don't see a problem with them making champs for the meta.

Braum is a strong and resilient melee support[...]

I have nothing else to add.

What do you suggest we play Braum as? He doesn't have enough damage for top or midlane, and he can't clear fast enough in the jungle. He carries a giant shield. What did you expect?

I completely agree with the Alistar one, that was just stupid and Riot shouldn't do that.

But with Sion, they were talking about the old Sion, so they changed him to something more versatile. That's a step in the right direction.

So yes, I see your point, but not all of your examples are viable.

Some points I agree with, others not so much.

The main thing I agree with you on is how they're hypocritical in saying that Bard should be a roaming support, but they will not stand for Alistar doing the same thing (and yes I play the cow, so sprinkle salt on me if you will).

The main thing I disagree with is how most champs should be played with AD or AP kits. The problem with AP in kits is that some old champions became downright broken with it.

AP Yi was probably the biggest offender; he'd click Q on a minion wave, obliterate the entire thing, and if you were in the way, you'd get chunked without being able to fight back since he'd be untargetable. If you retalitated afterwards, he'd Meditate and not only reduce damage, but he'd heal for a ludicrous amount; your retaliation would be completely absorbed. Once he got blue buff, how would you go about fighting this playstyle? Seriously, how? I didn't even play against many AP Yi's during his 15 minutes of fame, but holy hell, the idea of it is not fun at all.

Old AP Sion was not want Riot wanted out of him. You read his lore (which they attempted to preserve in the relaunch, which I think they accomplished) and he's supposed to be this frontline juggernaut who shrugged off arrows and blows like nothing, all while being this undead monstrosity. They ....tried to create that. The reason it wasn't intended was because of his ult, in my opinion. You want to be slashing at people, healing up a ton while being in everyone's face....but if you build AP, you have this cheesy 2-skill burst combo which had really high AP scaling for no reason. You're telling me a dude would run into battle with this bigass axe and NOT bother to use it?

The new Sion is leagues better than old, and with his new kit, you technically still CAN build him AD or AP, but in my honest opinion, he's just better off building very tanky, so that he can run into a teamfight and soak up so much damage...then die, and STILL be a threat for a few seconds. It matches what Riot's lore tells you about him, and he also looks straight-up amazing now, compared to Ahnald Pizzafeet Sion of old.

Sorry for my rambling, but I wanted to throw in my 2 (4? 5?) cents. I still hit 'Yes' to your poll.