Wild Swing

Removal that gambles with itself: you choose exactly three nonenchantment permanents and the card destroys one of them, picked at random rather than by you. The variance is the entire pricing mechanism. A four-mana sorcery that simply destroyed a nonenchantment permanent of your choice would be a fine, unremarkable card; bolting a one-in-three lottery onto it is the trade that lets the destruction reach any single nonenchantment permanent type (creature, artifact, land, planeswalker) without the usual narrowing clause. The skill the design invites is to make the randomness irrelevant by aiming everything at things you want gone: point all three at the opponent's permanents and one of them dies for certain, no matter how the die lands. The trap is the opposite instinct: name a single enemy threat alongside two of your own permanents and you have just cut your odds of hitting that threat to a third. The only honest reason to feed it your own board is when you actively want a death (a sacrifice payoff, a recursion enabler) or when you simply lack three enemy targets. That tension sits at odds with how players read removal, which is as a guarantee, and it is why a card built on biased destruction tends to get filed under novelty rather than tool. The mana value is steep and the variance never fully resolves, but the underlying lever (destruction you weight rather than dictate) is a genuinely different shape than the point-and-kill template.