A percentage can make a Zi Wei Dou Shu pattern sound rare. Before using it, ask what was counted. A table built from every possible birth hour in a chosen period counts chart inputs, not the number of people actually born in each hour. It also says nothing by itself about careers, income, or relationships.
Check the date range, time convention, transformation table, and level of detail. Two charts may match by major stars and differ once supporting stars or transformations are added. Finer categories naturally produce smaller groups; that does not rank people. To link a pattern to an outcome would require separate, reliable population and outcome data. Use frequency to understand a classification, then return to the palace structure and the person's real question.