Griff Counters

Best counters to Griff

Ordered by the low end of each brawler's 95% range against Griff, so a lucky small sample never tops the list.

Brawler	Win rate vs Griff	95% range	vs expected	Games
Gus	57.2%	54–60%	+1.1	944
Amber	57.7%	54–61%	+0.3	702
Shade	56.4%	53–60%	-0.7	908
Wendy	58.1%	53–63%	-1.1	363
Bo	55.5%	52–59%	+1.9	663
Starr Nova	57.6%	52–63%	-2.0	264
Surge	54.8%	51–58%	+7.8	834
Mandy	57.8%	51–64%	+7.4	211
Piper	55.7%	50–61%	+6.4	307
Pearl	57.1%	50–64%	+3.1	177

"vs expected" is the difference, in percentage points, from what the two brawlers' usual win rates alone would predict. Coloured values are outside the 95% range of chance; grey ones could be noise.

Specific counters: better against Griff than usual

These brawlers win against Griff clearly more often than their own usual win rate predicts — the matchup itself favours them, beyond being strong overall.

Brawler	Win rate vs Griff	95% range	vs expected	Games
Surge	54.8%	51–58%	+7.8	834
Mandy	57.8%	51–64%	+7.4	211
Piper	55.7%	50–61%	+6.4	307
Shelly	33.3%	28–40%	+6.4	228
El Primo	53.2%	49–57%	+4.7	577
Rico	51.9%	49–55%	+4.6	1,155

How Griff's counters change by mode

The strongest counters within each game mode. A brawler that answers Griff in one mode can be a poor pick in another.

What BrawlPick's draft engine picks into Griff

The picker's own answer when the enemy's only known pick is Griff, on current Ranked maps — before any bans, your team's picks or your collection are taken into account. It weighs map and mode fit, its counter rules, synergy and the observed map statistics, so it can differ from the measured counters above. It is the picker's model, not a measured result.

Maps: the current Ranked maps in the modes where Griff is played most in BrawlPick's data.

Map and mode statistics behind these answers last updated September 19, 2026.

Brawlers Griff beats

The other side of the coin: opponents against whom Griff's team wins most reliably — risky picks once the enemy has shown Griff.

Brawler	Griff's win rate	95% range	vs expected	Games
Shelly	66.7%	60–72%	-6.4	228
Nita	63%	57–69%	+3.8	230
Edgar	58.1%	55–61%	-1.7	1,037
Crow	59.3%	54–64%	-0.3	344
Colt	56.9%	53–60%	+1.2	794

About this data

Source:

battle logs from the official Brawl Stars API, collected by BrawlPick. Not copied from another site.

Players:

seeded from BrawlPick users, public leaderboards and top clubs, then widened to the players they met. Likely skewed toward more active, higher-skill players; not a random sample.

Battles:

every 3v3 battle with a win or loss (Ranked and regular matches), each counted once even if seen through several players. These numbers come from battles played Sep 18 – Sep 28, 2026.

Win rate:

share of a brawler's appearances where its team won. Both teams are counted, so the average is about 50%. "Games" is the number of appearances.

Pick rate:

how often a brawler was picked per 100 battles in the same scope. Every battle has six brawlers, so the pick rates of all brawlers add up to about 600%; a brawler on both teams of one battle counts twice.

Battle counts:

unique battles, recovered from the stored pick rates (which share the battle count as denominator): exact only when the rounded rates allow no other number, "≈" when they leave a small range, left out otherwise. "Brawler appearances" counts each of the six brawlers in a battle separately.

Matchups and teammates:

counted from battles in which all six brawlers were identified: a matchup is a game where the two brawlers were on opposite teams, a teammate pair a game on the same team. The "expected" figure combines both brawlers' usual win rates in those battles, so a large difference points to the pairing itself rather than to one brawler being strong overall.

Small samples:

win rates are only shown from 30 games (100 for matchups and teammates); the 95% range shows the remaining uncertainty, and lists are ordered by its low end. That reduces, but does not remove, the chance of a small sample ranking high.

What it cannot tell you:

the range only covers sampling noise. It does not correct for who played a brawler, the drafts it was picked into, the same players appearing in many battles, or balance changes during the collection period.

Game version:

not recorded per battle. If a balance update landed during the collection period, the numbers mix battles from before and after it.

Full methodology: how BrawlPick works.

Related

• Griff stats, best maps and teammates
• Gus stats
• Amber stats
• Shade stats
• How counter-picking works
• All brawlers