When you calculate control and test CTR confidence interval and probability of statistical difference, you probably expect the results to be positive numbers. Ask yourself, what is a negative CTR on the lower end of the confidence interval? That does not make sense but that is what you can see if you use the standard formulas based on a normal distribution.

Some calculators just chop off the negative part but that does not really represent the full interval. We use a formula that is based on a distribution that has no values less than zero. That is what you really see in the real world, is it not?