Stop Bad RNG | Build a Pity System

This video walks through a Unity loot system from the ground up: first a weighted random table, then a Monte Carlo pass to measure what players really see over many sessions. That matters because “fair” RNG often looks fine in code but produces ugly streaks in practice, especially in progression-heavy games.

The second half adds soft pity and hard pity rules to cap bad luck while preserving uncertainty. For teams shipping loot, gacha, drops, or reward systems, the useful bit is the methodology as much as the implementation: simulate the distribution, inspect the tails, then tune the experience instead of guessing from a few test runs.