/dev: Balance Framework Update

To recap, the goal of the balance team is to make Wild Rift the most balanced mobile MOBA game, and we really want to make sure all of our champs are viable and to make sure no champion or strategy dominates the game. What we decide to change is dependent on the specifics.

A few months back, we developed a balance framework based on the data and feedback we have gathered from the Regional Alpha, Closed Beta, and Open Beta. The framework acts as a guideline to help us determine what champs we should tune, but keep in mind that the framework is not a hard set of rules that we must follow, as there are a ton of edge cases.

The main statistic we look at for champion balance is champion winrates. Once we identify imbalances, we then use various other data to then help us conduct further analysis, such as average damage dealt per game. Currently, our framework examines how champions perform in different MMRs as well as specifically in the 5 meta positions.

Balancing around winrates by skill

Essentially, if a champion is performing too well at any MMR band, then the champion is probably overpowered, and if a champion is underperforming at all MMR bands, then the champion is probably underpowered. Additionally, if any champion is above 56% winrate across the board, we will consider doing an emergency hotfix.

Let’s look at some examples:

• If Dr. Mundo is at 53% at Elite MMR but has a pick rate of merely 1%, we probably will not nerf him.
• If Master Yi is at 56% at Average MMR but lower than 50% at Skilled and Elite MMR, we will probably nerf him.
• If Lee Sin is at 47% at all MMR and has a 3% pick rate in Elite MMR, we probably will not buff him. High skill cap champions like Lee can be very punishing when they are strong, so we are comfortable leaving them under 50%.
• If Xayah is at 46% at all MMR, we will probably buff her.

Balancing around winrates by position

For specific positions, we incorporated pick rates as another key statistic, mainly as a way to weed out champions that are not supposed to be played in that position. These framework numbers are subject to change as metas shift all the time. Because some positions are more impactful than others, the acceptable winrate bands differ from position to position.

Some examples of how we use the framework are:

If Evelynn mid is at 54% winrate and has a 5% pick rate, we will probably nerf her mid lane power.

If Jarvan IV support is at 55% winrate but has a 0.01% pick rate, we will probably not nerf him.

Just to re-emphasize, these numbers are just guidelines. Champions like Gragas, who can viably be played in up to 4 different positions, will be harder to apply these frameworks to since they are so versatile. The goal is to make sure that a single champion or a small pool of champions does not dominate a position and that each position has a wide variety of viable picks.

Evolution of the framework in the future

The framework will get a rework in the future, and we will give you guys an updated version once we incorporate 2 new aspects – pro play and champion bans. Furthermore, as we release more champions, the pick rate benchmarks we use in the framework will also change.