Sherlock Holmes as Data Scientist

 I never guess. It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. 

-Sherlock Holmes A Scandal in Bohemia.

Data! Data! Data! I can’t make bricks without clay! 

-Sherlock Holmes, The Adventure of the Copper Beeches

Sherlock Holmes is a good model for the effective investment analyst. He employs inductive logic although Sir Arthur Conan Doyle, the inventive author, sometimes uses the term deductive to describe Holmes. The police try to solve cases using their deductive skills, and Holmes as the consulting detective relies on his inductive skills. Deduction is not wrong; however, in many cases withholding hypothesis and focusing on observation may be more appropriate.

Deductive logic starts with a hypothesis and moves to observations and confirmation or rejection. It is a top-down approach. The police will have a theory that they present to Holmes, but Holmes inverts the process through starting without a hypothesis. Observations are made and from this information there is drawn a hypothesis or theory. Inductive logic extrapolates from observation. This is a bottom-up strategy. The solution is not developed first but is derived from a careful review of the facts. Induction starts with asking why about the information observed in an attempt to understand facts. There is not a hypothesis that tries to employ facts to support a view. 

The quant investor who focuses on data employs inductive logic. There may be a hypothesis tested which is the basis for deductive decisions, but the focus is on asking the question - what does the data say?