Confidence interval
Measurement
An interval calculated from sampled data to express uncertainty in an estimated population quantity.
For a 95% confidence procedure, intervals would cover the fixed true value in about 95% of repeated samples under its assumptions. This is different from a model's confidence in one answer. Sample size, sampling method and dependence among examples affect what the interval can support.
In practice
An alignment estimate based on 20 similar clips provides less evidence than a well sampled evaluation across many tasks.