When making decisions about complex, messy systems, it is impossible to reach 100% confidence with precision of ±1% because too many interacting factors are too complicated to fully account for, making attempts at that level of certainty either produce overconfidence (ignoring relevant factors) or paralysis (endless analysis).