How to compare apples and oranges ?Would you expect the Amtrak fatalities statistics to include statistics from Norfolk Southern, CSX, Union Pacific, Illinois Central RR, Kansas City Southern Lines, Burlington Northern Santa Fe, Canadian Pacific RR and Canadian National RR just to mention the Class I railroads operating on US territory ? Go to the website of the Federal Railroad Administration to find proper statistics for comparison with the HSE figures. The total number includes trespassers, but the numbers are broken down in groups so that e.g. comparison directly between passenger fatalities is possible. The numbers on the FRA website will change the first impression of safety of US rail travel versus UK rail travel, but it will not resolve the problem of whether to justify the figures on basis of - total number of rail miles travelled (only passengers miles travelled or all) - density of rail network (only passenger network or whole network and how to account for the scarcity of rail service in the midwest US and the busy northeast corridor between Boston and Washington DC) - frequency of trains - total number of rail trips - long-distance trips versus short-distance/commuter trips - average speed of passenger trains (?)so maybe the matter is better left as it is...

We should all quit spending time on this board and go trade and get rich enough to have drivers drive us in limos while we read and make phone calls in the back. I wanna wet bar with an expresso machine in mine. <img src="i/expressions/face-icon-small-wink.gif" border="0"> >>My dream has always been a private railroad car. When you want to go somewhere, you notify the railroad and go to your car. They pick you up and drop you off. It may be fast, it may be slow, but who cares if you're comfortable?When I was 8 the family took a train from Seattle to New York. About an hour after leaving my mother tried to turn down the steam heat and the knob came off and blasted steam throughout one of our two tiny compartments (one adult, four kids). In apology the railroad moved us to our own car (today, they'd charge extra for the steam bath). It was three delightful days exploring the luxurous accomodation, much nicer than our house.

Fabulous! My favourite private train belonged to the villain in the Bond film Goldeneye.

(I got a request for clarification of my position on relative auto safety.)Automobiles are the safest, because if you want to be safe, you can be.Automobile-fatality statistics largely reflect the deaths of people who made no effort to be safe!Meaning, there is nothing about getting into an automobile and driving, that automatically puts you in the same statistical group of people who died, when they got into their cars.I guess, you could just as easily say that the suicide rates of guns are higher than the suicide rates of frisbees. So, if a happy person wants to be "safe" from committing suicide, he should buy a frisbee!Nonsense! It is people who are unsafe, and cars are simply the mechanism which, unlike air travel, empowers these unsafe people to manifest their habits.You might, say "people are unsafe" - in that people engage in risky, self-destructive behavior - but that would be a stupid, roundabout way to determine if you, yourself - by being a person - are unsafe.Finally, I stated that you are comparing apples and oranges, since cars and airplanes aren't used for the same trips, and substituting one for the other would be much more likely to cost you your life in ways other than an accident. Meaning, if I insist on driving instead of flying on business trips - but I also choose to drive safe by doing all my driving in the right late between dawn and 10:00 AM on Sunday going the speed limit in a large SUV in the right lane - I will lose my job, and won't be able to pay my heating bill, and will end up working as a bridge painter, and riding a bicycle to work.But, in summary, auto and air fatality statistics are disaggregated as to cause or bayesian pools at different levels. For the most part when we get on an airplane we are all equal, and nothing particular to the individual, but only the riskiness of the form of transportation, is reflected. MP

Several years ago [actually over a decade if memory serves], there was a wonderful advertisementin Scientific American -- one in a continuing series on the scientists working in General Motors ResearchDivision [this was the 80s -- they likely got booted in the downsizing]Anyways, when analyzing auto safety, mandatory airbags, seatbelts etc these fellows were chargedwith developing a cost/benefit measure and rankings of various options for auto safety.Their key insight was one that many people forget; no one can save someone's life; they can onlyprolong it. Therefore, it is a question of how many $$$ spent to increase potential lifespan. This maysound cold and calculating, but it is something that we tend to forget.

MP,I must disagree with you on this point. The factors that are under your control on an auto trip are only a subset of the total factors that influence your likelihood of having a fatal accident on that trip. You may be the safest driver in the world, but you can't do anything to stop the drunk, blind guy in the buick from crashing into you.Also, you can only control the location and time of your driving to a certain extent before it becomes unrealistic. I think you alluded to this fact in your comment about ending up riding a bicycle to work (more people should.) Claiming that a person can choose to drive only in the right line during a few hours on Sunday morning is equivalent to claiming that a person can hire a private jet and helicopter for every trip out of the house. Technically it may be true, but it's not meaningful.

Yeah, but the blind, drunk guy in the Buick can hit you whether you're driving, walking or sitting the in the park. I agree with MobPsycho. I'd say about 90 percent of traffic fatalities are due to factors under the driver's control. Looking at it the other way, the safest drivers have about 10 percent of the risk of dying per mile driven than the average. Of course, as MobPsycho points out, some of those factors are things like when and where to drive, which most of us cannot practically choose. But mainly it's a question of driving defensively.On a vaguely related point, I believe that more than half of single vehicle fatalities are suicides, which skews the statistics.I disagree with MobPsycho on the apples and oranges point. I think a careful statistical study can determine whether driving or flying is safer. Of course the answer will not be one extreme or the other, but will depend on factors like driving skill, length of trip and others. It can be answered both in a public health sense (which one causes more deaths) and from the viewpoint of an individual.