Valuations are like leprechaun gold: they look really shiny in the pot at the end of the rainbow, but get too close and the money could disappear. The more mat

Valuations are like leprechaun gold: they look really shiny in the pot at the end of the rainbow, but get too close and the money could disappear. The more mathematical description is that valuations are a linear extrapolation of two data points. The first data point is (shares = 0, price/share = 0) and the second data point is (shares = [some small number], price/share = [relatively large number]). Valuations then linearly extrapolate to arbitrarily large numbers of shares. Linear extrapolation will only provide good results when used to extend the graph of an approximately linear function or not too far beyond the known data. http://en.wikipedia.org/wiki/Extrapolation#Linear_extrapolation http://en.wikipedia.org/wiki/Extrapolation#Linear_extrapolat...