I'm not following the logic. Well, I understand how variations in prop efficiency at different RPMs affect the scenario. The problem that I still have is the variation between engines that my very small sample gives.
The IO-360 runs pretty close to your 1"/100 RPM rule of thumb, and that is based on a 300 RPM change. The O-320 runs about 1/2 that figure for a 100 change. If prop efficiency were a factor wouldn't those ratios be reversed?
Or, lets assume for the sake of argument that Prop A when put on the IO-360 and Prop B on the O-320 end up giving the results that you claim. What happens when you switch the propellers and put Prop A on the O-320 and B on the IO-360? Or switch from a two-bladed prop to a three-bladed? Are we still going to see the same 1" corresponding to 1000 RPM?
What I suspect is happening is that the lack of accuracy in the resultant measurement equipment (ASI, VSI) that you've used for your tests overshadow the power differences so that casual observation would fit your hypothesis. The change in knots or FPM may be too small to be noticeable maybe? Anyway, if that's true I still think it might be disingenuous to actually take those casual rule-of-thumb observations and turn them into statements of fact like "23/2300 = 24/2200 = 25/2100 = 26/2000."
I don't know, I'm just looking at a small amount of data and not coming up with the same conclusions that you are. I haven't flown a piston aircraft in years and I don't have an easy way to do field tests so I have to use what little logic and knowledge I have available to me to try to follow along.
I'd be interested if taylor or fish have any insight into this question as well.