How large the treatment effect actually is — separate from whether it’s statistically significant.A large enough trial can make a trivially small effect statistically significant (p