That observation is true and not recognised by many. There is a definite relationship between power and thrust. It only gets upset by prop disk load and scale factors, which allow larger props to operate at higher disk loads.
My first calculations did not account for the disk load yet, and I used 6 lbs specipic thrust for each hp. These absorbed hp needed of course different prop rpm with different prop makes. You might want to check that figure against your data.
My [link=http://www.mvvs.nl/prop-power-calculator.xls]current calculations [/link] show the specific thrust per unit of power obtained with the calculated disk loads.