Estimates vary across samples. A confidence interval comes from a procedure with a stated long-run coverage rate under its assumptions; it does not resolve design bias.
Hypothetical intervals in outcome units, illustrating a nominal 95% procedure; eight intervals need not achieve exactly 95% realized coverage. The 95% describes long-run coverage under the model, not a 95% probability that the fixed mean lies in a particular realized interval. Sampling intervals do not remove design bias.