/dev: On Champ Diversity

League of Legends has, of this writing, 133 champions, with more on the way. But at times, you might feel like not all of those champions are available to you. Maybe the champion you want to play is legitimately underpowered during the patch you want to play her. Maybe there’s a different champ that you enjoy playing, but even though his numbers are tuned to the right place, he just doesn’t quite bring enough to a team comp to really handle current strategies — a situation often referred to as “being out of meta.”

This situation is particularly true of esports. About 40% of League champs made an appearance in Worlds this year, and of those several only got a game or two.

With esports high on our minds (and probably yours as well) after the fantastic final showdown between SKT and SSG and all the pre-season roster changes/rumors, it seems like an opportune time to share some thoughts on champ diversity. Do we think diversity is important to players? Is it important to Riot? Why? How does it stack rank with the rest of our goals?

Why Designers Value Diversity

From a game design perspective, having a variety of healthy champions that are viable picks does a few nice things for the game. For example, the ability to swap champs around from game to game can help prevent player burnout. Many of you may very well have mains, and that’s fine, but almost everyone can play at least a few champs. If your matches start to feel like they’re playing out the same way every time, swapping to another champion can help keep the game fresh. 

Adding a new champion is one of the most exciting moments for League from a development perspective, and updating old champions has virtually the same impact. For these releases to feel relevant, you have to be convinced that these new or updated champions will find a niche in the current state of the game, and you have to be convinced that you can eventually rely on decent win rates with those champions once you’ve got some experience under your belt.

This gets at a third benefit to having a lot of available champions: The never-ending path to mastery is one of our core design values. Having so many champions to master helps make sure League’s mastery curve stays truly endless. Seeing yourself get better at something is satisfying, and when learning a new champion, the mastery slope tends to be the least steep early on. You’ll see performance improvements almost immediately in your first few games. Even if you hit a point of diminishing returns on Yasuo, maybe you can see rapid improvement by trying to get better at a somewhat similar champ like Riven or Ekko.

What Diversity Provides to Players

From the point of view of players, diversity is important for a few different reasons (that are also important). As I mentioned, many of you do play mains, and it can really suck when your main doesn’t feel like a legit option. All that time you spent getting better at a champion can feel wasted if all of a sudden you feel like your only choice is to abandon your main and pick a new one. As I mentioned above, swapping your focus to another champ can be fun, but it’s better when you’re opting into that choice, not feeling forced into it.

If we’re going to claim that League has 133 champions, then the expectation is that those are actually playable champions. As gamers, I think it offends all of us at a visceral level when we see a system not living up to its potential. It feels like visiting an amusement park when half the rides are shut down. Maybe you can still make the most of it, but there’s this nagging suspicion that you’re not getting the promised experience, and that risks tainting how you experience other parts of the game. I think this is the same reason why very experienced League players still get bummed about our tutorial experience, even though it doesn’t directly affect them. It just feels crummy.

And finally, getting back to esports, it’s exciting when you see one of your favorite champions picked by pros, and disappointing if you realize that, for example, only Jhin, Cait, Ashe and Ez are going to see a lot of pro ADC play.

Barriers to Diversity

So if having a lot of champions being played is important to us and important to you, then what’s stopping us? Why isn’t diversity near 100% representation all the time? Well, the TL;DR is that it’s a big thorny challenge.

For starters, we do have a lot of champions. It requires a lot of bandwidth to focus on all of them, all the time. If you think about balance as a complicated mathematical equation, then consider that we are constantly adding more and more variables to that equation as we add champions, update champions, and change the game systems that surround them.

It’s also worth stating that a lot of Rioters are focused on creating those new champions, updating the old ones, and trying to improve the underlying rules of the game, in addition to working on things like matchmaking, rotating game modes, new skins, better clubs, the new client, and a whole host of other things. In other words, while champ diversity is important, a lot of other stuff is important too, so we aren’t throwing anywhere remotely close to a majority of development resources at the diversity problem alone.

Our balance philosophy for League is that we essentially use one rule set (high-skill players and those who aspire to be) when tuning champion balance from new players all the way to pro. This is different from say basketball, which has different three-point lines, or golf, which uses different tee boxes for pros. This philosophy can create convoluted problems when a designer wants to buff a champ for normal play without creating a 100% pick/ban monster for pro, or to nerf a champ for pro without pushing them out of normal play completely. We still think this philosophy is the right one: It lets League players follow League esports because we know the abilities and items are all going to work the same way, and inspires us to try and get better in our own games by seeing how the pros play. But it does make balance even more challenging to achieve (as an aside, I’m not convinced that being able to tune champions separately for different skill tiers makes the job any easier — now you don’t have to worry about affecting pro Gnar when you tune Silver Gnar, but you still have to tune all of those different Gnars).

And following on to the topic of pro League, the pro experience is pretty different from the way most of us play League. Pros often pick champs with strong team synergy, while many of us need to rely on a champ with some amount of individual carry potential since we might not be able to rely on a cohesive team in a solo queue experience (though I will insert my soft pitch here to try the new Flex queue and see if it can provide you with a competitive experience that also includes team-building as a strategy). This difference between pro and non-pro can cause specific champions (say Elise, Orianna and Rumble) to frequently show up in pro even if their play rate for non-pro Ranked is much lower. Pros also have busy schedules. Investing a ton of time in getting good at an unusual champion may be lost time if the conclusion turns out to be that no, that champ isn’t going to really work. If they find a “good enough” champion, it’s probably more efficient just to practice and scrim with that champion rather than play the field. Then again, when crazy picks like Support MF show up, it’s pretty exciting.

Most importantly, we’d rather have a fun, healthy game than a diverse game. Buffing very frustrating champions just to increase their play rates is irresponsible for developers who are supposed to care about the player experience. It has an icky “meeting quotas” feeling. On the other hand, the right solution is to address the source of their frustration, not just to leave them weak. So there will definitely be cases where we have to take a step backwards in diversity in order to achieve other goals. As an example, we made some turret changes pretty late in the pro season to discourage lane swapping as a strategy. We knew that change would hurt diversity and we knew we wouldn’t have any patches left to try and get some champs back into pro play. But we still felt like that change was worth it so that Worlds would be a good viewing experience (and I for one, feel like it really was).

Finally, it’s important to keep in mind that players, even pros, make their decisions about who to play for a lot of reasons. Individual champ strength is certainly a factor, but so is whether their mechanics are fun, whether the champ is on free rotation, their visuals, their backstory, and changes to other champs (talking about things like counters, team synergies, or even item or mastery changes) and so on. 

WHAT DOES GOOD DIVERSITY LOOK LIKE?

At the end of the day, we aren’t really looking for 100% champ representation at Worlds. Even though we were fortunate to have Finals itself go to five games, you’re still only talking about two ADC players, Ruler and Bang. How many Marksmen, realistically, can show up even in a best-of-five that goes all the way? Even when you expand that to all the ADC pro players, it’s still not likely that every Marksman will show up. Pros, like the rest of us, are always going to have favorites, and that’s not only fine, I’ll argue it’s good for esports.

Equal play rates aren’t a goal. Players love Jinx and Ahri for reasons other than their win rates. Even though he desperately needs an update, Urgot will probably always be a niche champion, and we’re fine with that, so long as some players out there love them some Urgot.

I don’t know if there is a magic number. Maybe it’s 60% instead of 40% representation at Worlds. Maybe it’s 80%. Some of that is going to take more extensive reworks on our part. We are going to have to lower skill ceilings and raise skill floors on a lot of champions before we see Garen played in the upper tiers of ranked, or see mere mortals playing someone like Azir or Kalista to their maximum potential. One thing that we do think will help, at least for esports and other forms of organized play, is expanding the number of bans to 5 per team, and having these bans made in two rounds. We think ban changes will make it easier for players to clip off the ends of the bell curve. The S-tier champions will dominate less and hopefully more of the A and B champions will get more regular play. While we don’t have a time frame yet, more bans is something we are actively working on.

I get that topics like this are largely philosophical. I understand that for some of you it might be a no-brainer to balance pro differently than we balance Bronze, or to balance pro even at the expense of Bronze. We could have stronger champ counterpicks during champ select, or alternatively, we could move our more niche champions towards being more generalist picks so at least they’d get more play time. I don’t mind if you disagree with us on our goals – as I said, this stuff is highly philosophical – but at least now maybe you’ll understand them a little better, and help us to deliver on them.