If the reasoning behind the number of TRC needed to set up a masternode is based on the $ amount needed (as explained by clockuniverse, predicting future prices of TRC) then maybe it should be better to fix the price of a masternode in $
That originally was my idea, but the masternode is locked to an exact amount in a transactions so an ever changing amount won't work.
My predictions are total guesses. Optimistic total guesses.