by John Andrews (revised 4 April 2006)
Each year we suffer 500 rail corridor trespass fatalities in these United States. According to Pamela Caldwell Foggin, Federal Railroad Administration, these fatalities do not include fatalities caused by vehicles passing rail gates, nor do these 500 fatalities include suicides.
According to Betsy Goodrich, New England Office of Rail-to-Trails Conservancy, there are 142,000 miles of active rail corridor in the United States. This means 500 fatalities per year per 142,000 miles. Or one trespass fatality for each 284 mile-years.
Using data provided by Mia Birk, senior author of Rail-with-Trails, Lessons Learned, we learn that we have 4,400 mile-years of rail-with-trail (RWT) experience in the United States.
If trails do not increase the danger, then we can expect one rail trespass fatality for every 284 miles per year, then 4,400 mile-year should result in 15 RWT trespass fatalities since the first RWT was opened. If trails increase the risk of pedestrian fatalities, then we might expect many more fatalities. Maybe a ten-fold increase or 150 fatalities in 40 years?
But, have we experienced 150 RWT fatalities? Have we experiences even 15 fatalities? No. In the entire United States, there has never been one RWT fatality.
My probability professor would not give me a failing grade, if I claimed adding trails to rail corridors reduced pedestrian fatalities by 25-to-one.
When I first looked at my math, I felt my math must be wrong. Einstein once told an acquaintance, “If your math does not match your common sense, check your math.” I've asked many people to check my math. Many have rejected my conclusion, but no one has questioned my math.
Assuming the conclusion it valid, or even close to reality, how do we explain it? This troubled me for months after I first ran the numbers and looked at the result. It now makes sense, at least to me.
Whenever people trespass in rail corridors, the walking is generally awful but usually easiest on the rails or rail ties. So kids walk on the rail ties or rails. College students leaving a pub may choose to walk an unlit rail corridor to return to their dorm. Hunters seeking game walk the rails. But, if a well-engineered trail existed beside the tracks, most would apparently choose the easier path. Therefore, it does make sense to me that adding a trail to a rail corridor could reduce fatalities by as much as 25-to-1.
Common wisdom in the rail industry has been that anytime a person enters a rail corridor the risk of a fatality increases. It's obvious to experienced railroad employees that adding a trail and inviting public access should increase the fatalities. Claiming that it will save lives? Crazy! Reading Steven D. Levitt's Freakonomics inspired me to look at existing rail fatality data. Why? Because in his book he demonstrates that common wisdom is often wrong.
Craig Della Penna, well known RWT advocate, asks the rhetorical question: “Could a railroad be successfully sued if it had refused to allow a RWT where subsequently a trespass fatality occurred?” A NYC attorney with railroad litigation experience is interested in the question.