This is very interesting but, are you sure that you can apply this metric universally while disregarding all other factors? Amount of users holding a currency may not be the exact correlation between price. For example, it could just be a side product of another variable such as, the amount of merchants that accept a currency as a means of payment which would naturally equate to a wider network but would not mean that the price is correlated with the network size but rather, the range of use-cases currently available.