Rogue

Three lenses on the same data, all filtered to enemies that actually beat Rogue (matchup win rate above 50%). The threat ranking uses raw edge over 50% so a real-impact 53% threat isn't softened away; the win-rate number you see on the row is the same sample-confidence-adjusted value shown on the hardest-matchups card, and we drop matchups below 200 samples entirely so tiny-sample noise never shows up here.

• Combined — pickrate multiplied by win-rate edge above 50%. The default. A hero who beats you 53% of the time but is meta-frequent ranks higher than one who beats you 60% but is rarely picked.
• Pickrate— how often the hero shows up in this slice. Useful when you want to ban someone you'll definitely face.
• Counter strength— matchup win rate against you, ignoring frequency. Useful when you want to remove your worst matchup regardless of how often they're picked.

Per-map ban data isn't available yet — matchups in our dataset aren't broken down by map.

The confidence dots on each row reflect the matchup sample backing that specific recommendation. A niche hero with thin data lands lower on the dots even when their threat score is high, so you can tell which bans are battle-tested and which are still forming. Widen your rank or game type to thicken the samples.