So, what's the deal with eventual consistency versus strong consistency?

hey everyone im wrestling with consistency models lately in a new project were building a distributed system and need to decide between strong consistency and eventual consistency for our data ik strong consistency is simpler to reason about but it often comes with performance penalties and availability issues eventual consistency can be more performant and available but its trickier to implement and debug whats your go to scenarios for picking one over the other any war stories or best practices you can share