There is a scene in chapter one of the Sign of the Four that shines a light on the absurdity of how Sherlock Holmes solves crimes. The first step of Holmes' method is to have ludicrous amounts of detailed information about every topic at instantaneous recall. Second, you have to accept that definitive conclusions can be drawn based on casual observations and an encyclopedic knowledge of everything. Watson enters their apartment and Holmes observes that Watson sent a telegraph at the post office on Wigmore Street. Holmes points out that he recognizes the dirt on Watson's shoe and it can only be found in a particular London neighborhood and since there is construction outside the post office Watson must have gone there. If Holmes had suggested that Watson had been to Belgium because he recognized the dirt on his shoe I might have believed it, but I refuse to believe that the dirt in various London streets is so unique that it can be identified without a microscope. Secondly Holmes claims to have deduced that Watson sent a telegram based on his observations. Deduction means you have removed all other options and there is no other explanation remaining. What Holmes is doing is induction not deduction. He has proposed a plausible explanation but there are other possible explanations. Everything definitive thing Holmes claims to deduce is actually just a possible explanation based on casual observations and inductive reasoning.
Doyle is a very talented mystery writer and his stories are always entertaining, although Study in Scarlet has a massive digression in which half the novel turns into a Western set in Utah with a bunch of evil Mormons. I enjoy Doyle's writing, but it bugs me sometimes that people aspire to Holmes' level of brilliant deduction when Doyle had to know it was all smoke and mirrors.