Hi @flaky-noon-11399 There is a bug with the relative lift calculation (% Change in the screenshot). When the metric value in the Control group is significantly lower than the value in the Variation group the % Change, Chance to win, and confidence intervals are calculated incorrectly. This issue seems to occur only when the Control metric is substantially lower than the Variation metric. In the opposite scenario everything is calculated correctly. Can send screenshots in DM.

Hi @freezing-kangaroo-80111 good morning and Happy Friday! Please can you share the screenshots via DM so we can further investigate?

Are you using informative priors? Can you click in to one of the affected metrics and scroll down to the Overview of the metric settings and share if you have a prior set? You can also go to the Debug panel when you click in to the drilldown to see the impact with and without priors.