Methodology

We analyze high-level Ranked matches: those with an average rank of Masters 1 or higher, or where most players reached Masters this season or the last one (everyone drops in rank when a season starts, and this way we don’t lose those matches).

The system checks for new matches every few hours and recalculates the meta for the current patch. The meta of every previous patch is kept as history.

Win rate alone is misleading. A brawler that is almost always picked to answer another one wins a lot because of that matchup, not because it is strong in general; and a very popular one is played by players of every level, which lowers its win rate even if it is very good.

That’s why we use a model that looks at each full match and separates what each thing weighs: the brawler, the map, the matchups against the enemy brawlers and the skill of the player. What remains is the strength of the brawler itself.

Each brawler’s score combines its strength (70 %) with how often high-level players pick it (30 %). In the overall meta, doing well on many maps counts too, not just on one.

Tiers are assigned by position: the top 10 % is S, the next 20 % A, then 30 % B, 25 % C and the rest D. To reach S or A a brawler has to be picked often: one that wins a lot but almost nobody picks can reach B at most. In S, it also can’t lose clearly more than it wins.

So the tier list doesn’t jump with every update, a brawler right on the border between two tiers keeps the one it had.

Faded portraits still have few games, usually because they are banned a lot or rarely played. Their spot is less precise, they can’t go above A and they will adjust with more data.

On each map we measure each brawler’s strength there. If it still has few games on a map, it takes what it lacks from its mode and from the overall meta, so the result is reliable.

Since November 2026 we know what each player bans. We show how often each brawler is banned on each map and on its page.

We don’t use bans to build the tier list yet. A heavily banned brawler is picked less because it is almost never available, and that can lower it a little; we are measuring how to account for it.

We do use them in the ban suggestions: a brawler that is banned a lot on a map moves up the suggested bans.

Brawlers that didn’t change start with the strength they had in the previous patch; the ones that changed start from scratch and move with the new games. This way the tier list makes sense from the first hours.

In the first days of the patch we show arrows for who goes up and down a tier; the meta map shows how each brawler moved since the previous patch, and the patch cards how the changed brawlers are doing.

Suggestions take into account the mode, the map, each brawler’s strength there, who it beats and who it struggles against, the combinations that work together and keeping your team balanced.

Matchups come from the same model: it is what changes who wins when those two brawlers face each other, without counting the rest of the teams or who plays them.

We also think one turn ahead: if the enemy can still pick something that punishes your pick, we take it into account, especially on the picks where you won’t be able to answer.

If you have first pick, we suggest safe brawlers to open with and what to ban to protect them. If not, we prioritize banning the strongest threats on the map.

Picks follow the Ranked order (1-2-2-1): on your turn we highlight the best options and on the enemy’s turn, what they will probably pick. When the draft is done, we estimate how likely you are to win with it.

As a base we recommend the Shield and Damage gears, the most versatile in Ranked. For some brawlers we recommend gadget, star power and gears by mode or map.

Each team’s strength comes from its matches against the other World Finals teams since our first records: beating a strong team counts more than beating a weak one, and recent matches weigh more. Only team games count (scrims and tournaments); what each player plays in Ranked on their own is shown separately.

With that strength we compute the chance of winning each game, each set and each match, and from it the World Finals predictions. They are an estimate, not a certainty: on stage the map, the draft and nerves matter too.

Suggestions are a guide, not a rule: communication with your team and how well you know each brawler matter too.

The data comes from high-level matches. At lower ranks, different brawlers sometimes do better.

We don’t know in which order each brawler was picked in a match or which build each player used: the data source doesn’t provide that.

After a balance change, the meta takes a few days to fully reflect it.