Relying on anecdotes is intuitively a bad idea, but why? When we rely on a very small sample of outcomes, the results may not represent the population as a whole. This is the basis of gambling, alternative medicine, and it's something that science based medicine has very carefully removed from the process. We trust statistics because they are more descriptive and predictive than anecdotes. J Clin Epidemiol. 2008 Dec;61(12):1197-204. Epub 2008 Oct 1. The conclusions on the effectiveness of homeopathy highly depend on the set of analyzed trials. http://www.ncbi.nlm.nih.gov/pubmed/18834714 This was a comment on: Lancet. 2005 Aug 27-Sep 2;366(9487):726-32. Are the clinical effects of homoeopathy placebo effects? Comparative study of placebo-controlled trials of homoeopathy and allopathy. http://www.ncbi.nlm.nih.gov/pubmed/16125589 QualiaSoup's elegant video on anecdotes (a personal favorite): http://www.youtube.com/watch?v=NPqerbz8KDc Dr. Wikipedia: http://en.wikipedia.org/wiki/Anecdotal_evidence