The recent improvement in car accident rates is pretty impressive.
The first graph show we have made driving far less dangerous per mile.
The second graph shows the likelihood that any given person will die due to traffic.
This is akin to announcing a 'war on cancer fatalities' but only caring about improving cancer treatments, not removing carcinogens on the environment.
Obviously, it is a bad idea to maximize the wrong metric. In the case of cancer, just measuring survival rates is a bad metric, but so is measuring cancer deaths per capita. A better metric is to measure the per-capita deaths due to cancer within specific age groups.
I'm not missing your point, and cbr's reply makes it obvious that he isn't either. The prevalence of driving is a separate variable from how dangerous it is. Driving is far less dangerous and far more prevalent.
> Here's a gedankenexperiment for you: I instantaneously and permanently disable every car and car manufactory, thereby reducing all traffic fatalities to zero. Did I make driving safer?
You certainly would have done an effective job of executing a 'war on car fatalities' which is what the harryh was claiming was demonstrated by the chart he posted.
While I agree that driving is safer than it used to be, the chart harryh posted doesn't actually show that. it just shows that driving is safer per mile.
Analogous to the point that cbr made: Without changing any of the risks of driving, I could make driving safer per mile by hiring a million drivers to drive 10MPH on a closed track year round.
While the above point is rather tongue-in-cheek, I do suspect that a significant part of the drop indicated in harryh's chart is due to changing driving patterns, particularly the increase in freeway driving.
I don't think eliminating all driving would be a good way to reduce traffic deaths (because it all of the other consequences that would have). However, if we had spent more effort in trying to reduce unnecessary and high-risk driving we would see a larger reduction in traffic deaths. This includes solutions like walkable cities, adequate public transit, and higher penalties for drivers engaged in risky behavior that cause fatalities. Focusing purely on deaths per mile excludes those types of interventions. Thus you can't use that measure to claim a 'war on car fatalities'.
You then 'charitably' missed my point twice, and argued against a strawman. You followed that up by 'charitably' accused me of only arguing semantics rather than responding to any of my actual points.
The original graphic shows that driving is less dangerous now than ever before. However, it does not take into account the prevalence of driving, which is also higher. Therefore, the original graphic is "misleading", because "traffic fatalities" is the combination of the danger of driving and its prevalence.
Given this current understanding of your point, my only argument now is that I believe that is an uncharitable view of the original post. I argue this because the view is entirely based on that singular interpretation of "war on traffic fatalities". A charitable approach is entirely based on attempting to view an argument with suspension of your own beliefs and interpretations in order to understand the other party's. Through such, common understanding can be worked towards.
I'm also sorry that you feel that I have been uncharitable to you. I honestly do believe your argument is entirely centered around the semantics of "war on traffic fatalities". The information you provided in your graphic is very good, and the point you are making is not wrong. I didn't have any issues at all until your response to cbr's original post, which is where I feel you began arguing semantics over content. I feel this way because cbr was also not wrong, and if anything was an equally good summary as what I stated above. And yet you still seemed to take an issue with it.
> Terrorism has never been a problem. The same number of people died in car accidents during September 2001 as they did in terrorist attacks. Where is our war on cars?
My main point, and reason for making the distinction between metrics is explained here:
"However, if we had spent more effort in trying to reduce unnecessary and high-risk driving we would see a larger reduction in traffic deaths. This includes solutions like walkable cities, adequate public transit, and higher penalties for drivers engaged in risky behavior that cause fatalities. Focusing purely on deaths per mile excludes those types of interventions. Thus you can't use that measure to claim a 'war on car fatalities'."
Are you against hospitals too? Deaths in hospitals have skyrocketed since they were invented.
You are being a pedantic jerk. Read this:
http://paulgraham.com/disagree.html
Here are some more reasonable discussions of the issue:
https://en.m.wikipedia.org/wiki/Transportation_safety_in_the...
http://www.theatlantic.com/international/archive/2014/02/a-m...
"DH0. Name-calling. This is the lowest form of disagreement"
> Are you against hospitals too? Deaths in hospitals have skyrocketed since they were invented.
"DH1. Ad Hominem."
Good job, you've hit the bottom of the disagreement heirarchy. In my mind my responses are all on the level of DH4 and DH5. If I have failed to do that, please provide specific examples because I strive to disagree well and wish to improve.
> Here are some more reasonable discussions of the issue:
Neither of those are discussions of the issue at hand, which is the number of deaths due to cars. Nowhere, except in replies that attack a strawman argument, was the safety of driving a point of debate. I have even stated elsewhere that I agree that driving has gotten safer (though I do dispute that fatalities per distance is an ideal metric for measuring that since it is also affected by usage patterns in addition to safety).
I have made solid, non-pedantic points about the effect of using different metrics and what types of interventions will show up in those metrics.
If you are talking about traffic safety (say vs air travel or between countries), fatalities per distance is one of several applicable metrics (fatalities per trip being another).
If you are talking about traffic deaths, as a overall source of risk (say vs terrorism) then fatalities per capita is a more correct metric.
This is not a pedantic point, this is a very important point that is key to the discussion that you seem to be avoiding.