A popular statistical advice is to meet N people (30..100), determine the level of “bestness” (ignoring extraordinary anomalies) and then to continue until you

A popular statistical advice is to meet N people (30..100), determine the level of “bestness” (ignoring extraordinary anomalies) and then to continue until you find something like that again. Calculating chances of being that last person is left as an exercise to the reader.