Adjusting the models

Because almost all the variables in analyst models represent unknowns, there's a lot of room for fudging. But at the end of the day, the numbers derived from them have to bear some intuitive relationship to reality. As Hecht put it, "I think my models should always reflect my qualitative feelings."

Thus when the assumptions lead to numbers that feel wrong, analysts typically will adjust one or more of their initial parameters to come up with a more "realistic" conclusion. For example, if Stone likes a company and the numbers don