In financial credit modeling it is common to transform variables from their raw state into weight of evidence (WOE) transformations for binary problems such as probability modeling. One advantage of using WOE is that it handles missing value, outliers, and categorical variables. It also reduces noise and provides an intuitive way of comparing the variables' predictive power as well as producing directional relationships. In this paper the examples and explanations will be from a finance credit modeling perspective. Instead of the common terms of "event" and "nonevent" for modeling a binary outcome, the term "bad" will be used for the event of a loan default and the term "good" will be used for non-defaulted loan. WOE is used in many industries however this paper is using finance to create a more tangible example.