We sampled about 1,000 queries, mostly from 2008. The data was very literal about the strings "chance of" or "probability of", e.g. not including "chances of". We used only queries where the user is apparently seeking to discover the chance of something. The most common exclusions were references to the movie Cloudy with a Chance of Meatballs and the TV show Real Chance of Love; also to song titles/lyrics and the Kid's chance of ...... organization. Also excluded were apparent homework exercises e.g.