Everyone ranks cards by win rate. I think it's misleading, and I'm pretty sure there's a better metric. The problem with win rate: it's attached to the card, not the context. A card can sit at 48% WR because it shows up in a bunch of bad decks and trash experiments — while actually being a core piece of several top decks. And a card can hit 55% WR just by riding along in one very popular meta shell, then crater the moment the meta shifts. So WR tells you "does this card tend to win when someone plays it" — which sounds right but quietly measures the player pool and the deck pool as much as the card itself. What I did instead: filter first for decks that actually work — 50%+ WR, 500+ battles per deck, Path of Legends only — and then count how often each card shows up in that pool.