Data and Champion Balance – Part 2

Click here to check out our first Dev Blog on data and League Balance!

Hey guys, Jag here from the Live Gameplay team. We’re back with another dev blog, this one addressing our balance strategy a little more broadly. While the last blog covered some of the metrics we use in assessing champions performance, in this article we’ll talk about the more subjective, value-based decisions we make with this data: these are balance changes we make to change the state of League of Legends in ways that are more consistent with our design values.

Before we get to it, we’d like to clarify that this article is meant to give an understanding of how we tend to view champion power in regards to how we change them. To emphasize, these are tendencies; not hard rules. Some of the trends we discuss here will not always apply to every champion or in every scenario.

Let’s start with an easy concept: the underlying counterplay of a character. Generally, if a character has low counterplay, then they also tend to be more binary on the champion strength scale (either oppressively strong with no answers, or incredibly weak with no agency), whereas the opposite is true of champions with lots of counterplay. The best example here is Blitzcrank: if a Blitz is particularly successful, he’s highly impactful to the point of virtually winning games outright by turning them into 4v5s. That’s a champion with a lot of power. That said, most of the play around dodging Blitz’s hook seems pretty fair as an opposing player. If Blitz’s hook was a targeted spell instead of skill shot, he would feel pretty oppressive while consistently winning games, so ensuring champions have good counterplay makes it fairer for them to succeed. This is one of the many reasons why we pursue having counterplay options against every champion; even when they’re strong, they don’t automatically tilt over into oppressive territory.

One of the more interesting balance challenges we face is how to properly calibrate champions that buff or empower allies. Internally, we have started to use the term "Force-Multipliers" to describe those (traditionally) support champs who express most of their power through their teammates (i.e.: Lulu). Since these champions invest in conditions that are more teamplay-centric, we’re inclined to give them a significant amount of power when they put themselves in the correct situations. Perhaps the best example of this is Janna; while she does have a few ways of making that “big play” that both teams can appreciate, most of her actions enhance her teammates' potential through low visibility influence, without a huge amount of power that she can express directly. So while a good Janna can flash ultimate a high value target into her team, a great Janna may also keep an eye on her friendly Tristana so she can shield her right when she hits Rapid Fire, leading to Trist landing a Pentakill (and getting all of the credit...). Ultimately, given that League is a team-focused game (and teamplay is one of our core design tenets), we tend to be more in favor of abilities that require good coordination or teamplay to fully ‘unlock.’

A consequence of empowering Force Multipliers is that when they become too strong, their power isn't appreciated in obvious ways. Another example might be pre-update Sona, who was giving her teammates a ton of hidden power that few were attributing to her. There would be many games where she was instrumental in her team's victory, but due to the nature of how she expressed strength, her teammates would be credited as the reason behind the victory (maybe she should speak up!). We may never truly ‘solve’ Sona’s invisible power problem because of her identity as an aura-focused support, but our recent updates have pushed her toward having more appreciable power. Ultimately, taking action in cases where the perception of a champion’s strength is significantly misaligned with their actual power level (one way or the other) is a large risk for us, as players can sometimes not understand why a change is necessary or even warranted.

Balancing champions who function in multiple lanes is another ongoing challenge we face. We have data that tells us how often a champion will win a game depending on if they’re in the top, mid, the jungle, or in a duo bot lane. Sometimes when champions become strong, we emphasize certain roles more than others if the character’s play pattern is healthier there. An interesting recent example is Sion following the introduction of Cinderhulk. While he was a very strong pick in the top lane where he was favored by the majority of the player base, most thought he was close to a balanced state and only needed a few small power reductions. On the other hand, we also saw that Sion’s strength as a jungler was absolutely monstrous (close to a 60% win rate), but this was not clearly visible to players since they tended to calibrate his strength as a top laner. Consequently we made changes that reduced his overall win rate but were targeted at his jungle power. While some players understood our intent, others were quick to register their confusion at nerfing a ‘balanced’ top laner. In the case of these types of changes, we need to be very proactive about providing the right context in places like the patch notes or patch rundown. Lastly, while we’re also trying to improve our design craft as to how targeted our changes are, there will always be some bleed with a change targeting one position to another.

A final area I’d like to touch on is champions with distinct, exploitable weaknesses. In general, it is far easier to allow for high strengths in champions with clear weaknesses versus generically powerful characters with no exploitable counterplay. Certain strengths are very exploitable (range and mobility in particular), so when a champion can be easily attacked or their opponent has multiple methods to counter, we’re satisfied with that champion also having very distinct and powerful advantages. Take Sivir as an example; she has an extremely low basic attack range for a markswoman and has no teleport or dash moves, so she can be dealt with in many ways. You can pick Draven and beat her in a matchup where Spell Shield doesn’t give much help in; you can bring mobile dive champions like Irelia or Jarvan IV; or you can force engagements when she tries to siege with her low range. Because of this, we can allow Sivir to have very powerful strengths like top-tier waveclear and one of the most powerful ultimates any marksperson has in League of Legends.

Hopefully that gives some insight into how the Live Gameplay team treats balance in League of Legends. With any luck, if there is anything these last two blogs have showed you, it’s that the power of a champion is a concept that needs a lot of context and investigation. Thanks for reading, Summoner.