Been thinking about this for a while actually. There's an old spur line that runs through the valley near where I used to farm, and I've heard three separate accounts from locals over maybe six or seven years all within about a quarter mile of that corridor. Never saw anything myself but the consistency bothered me enough to start mapping it out.
My working theory is it might not be the railroad itself but what the railroad follows - these lines were often surveyed along natural terrain features, low ground, water drainage, that sort of thing. Which overlaps pretty significantly with ley line pathways in some cases and also just basic wildlife movement corridors.
Has anyone actually cross-referenced dogman reports with old USGS rail survey data? I'd be curious whether the correlation holds up statistically or if its just confirmation bias doing its thing. Oregon has a fair amount of abandoned track and I could probably pull together some local data if there's interest.