Mystery solved. I am guessing this is close since the 8' mark 1. is at the top of your door and windows. I would almost bet the top of your door is more like 6', if that is the case subtract 2' from each increment. Using the mast to do a rough calibration of 10' I was able to use photo shop to plot some rough measurements. I think your statement "the base of the 8' antenna is at 52' AGL" is way off. That would mean your single story ranch home has a roof peak of 42'. I live in a 2 story condo with 10' ceilings downstairs,10' ceilings upstairs with 10' of attic space above it. Measuring from the peak of the roof to the ground with a tape measure I got 32'. I would guess if you re run your calculations using a 25' AGL to be realistic; you will see it is much closer to the results you are seeing in the field. It is easy to over estimate height from the ground. The below image is a 35' tower next to a typical single story house to compare. My coverage results using 25' AGL, 6dB gain, 30W of power with an 70% reliability factor. I used 2dB of TX-RX loss to factor in the coax and duplexer. for the other end I used 6' antenna height and 2dB of gain to simulate the average car. When using 5' and 0dB of gain (HT) I got less then 3 miles. Interactive Coverage Map I have 2 commercial sites and have used this program for FCC coordination and coverage, the results have always been real close to reality as long as you input the best possible data. The difference in 25' to 35' is huge so you have to be as close as possible.