At RTD, we had a tool called "the egg" which was an egg-shaped piece of cardboard. The center of the egg was placed over a prospective park-n-Ride location, with the small end pointing at the CBD, or university, or medical complex, etc. Based on license plate studies, it was remarkably accurate at showing where a rule of thumb 80% of the demand would come from. IF I remember correctly, the long end of the egg went out about 10 miles, keeping in mind that this was for bus facilities with few priority measures.
Of course, there may be a pick-up with Wyoming plates sitting in the lot, but that's why the rule of thumb only predicts for 80%.
For the Oregon state employee commuter buses started up in the 1974 Energy Crisis we called the plotted home addresses of our members the "comet" map. The head of the comet was at the parking lot, with lines trailing out to home and almost no one backtracking. The lack of backtracking was probably due to the Interstate being wide open at their location and a shorter drive to Salem. There were a few oddities, such as cyclists, connections from Tri-Met transit buses, or people who were dropped off in the evening at different locations than home. Recollecting all this, the 80% rule and the longest tail being 10 miles probably applied to the comet map, too.
These methods were especially handy when they would reveal overlaps where infill locations were being promoted by interested parties.
