Comment by blogimus | Hacker News Reader

Comment by blogimus | Hacker News Reader

This analysis is a joke. Have you looked at the data on this page? It is inconclusive. The context is insufficient. What about overall traffic, weather, non-fatal accidents? other factors such as visiting travelers or data by county. Nowhere do I see data demonstrating any speed safety paradox. Comparing fatalities in only two years? and look at the green and red boxes between 1998 and 1999 for averages.

If you want to do a real study, start walking through some data covering a wider range of years, like the 2003 reports here (which covers fatalities by county going back decades):

http://www.doj.mt.gov/enforcement/highwaypatrol/annualreport...

Then you also have to consider improvements in auto safety over the years and how many of what kind of car and driver demographics were in the fatalities.

Edit: More traffic and fatality data:

http://doj.mt.gov/enforcement/highwaypatrol/forms/annualrepo...

http://doj.mt.gov/enforcement/highwaypatrol/forms/annualrepo...

http://doj.mt.gov/enforcement/highwaypatrol/forms/annualrepo...

http://doj.mt.gov/enforcement/highwaypatrol/forms/annualrepo...

http://doj.mt.gov/enforcement/highwaypatrol/forms/annualrepo...

The article alleges that matters of fact were also lied about, such as the nature of the effect on highway funding.

If there were a study that suggested similar conclusions, what would the result be? More lying?

It seems to me that the proponents of the law should have the burden of conclusive proof, not the opponents.

Speed limits are a major imposition on drivers. They would have to have a significant effect on fatalities to be worthwhile. What about spending the same effort on other life saving approaches, such as mandating safer vehicles? Is that more efficient?

> It seems to me that the proponents of the law should have the burden of conclusive proof, not the opponents.

Why? It seems like the norm is speed limits, I thought that the burden of proof rested with those contesting the norm.

Depends on what you mean. If Montana currently had no speed limits, and someone was proposing that Montana impose speed limits, should such a motion then place the burden on the opponents to defend the current practices, or should it be on the proponents that are proposing new practice? Trying to say, "well everyone else is doing it, so it is 'the norm' and we should just follow their lead," is rather disingenuous. You can't propose a new idea counter to an existing idea, then try to push the onus of proof onto the supporters of the existing idea while skirting any proof of the new idea.

Similarly if the tables were turned, the proponents of removing the speed limits should have the burden of proof placed on them as well.

Eh, maybe. As far as I can tell, the speed limit was repealed in Montana without much to back that course of action either. I don't see how you can put the burden on the speed limit group when that burden was not borne initially by those who supported its repeal. Seems to me that under that regime whoever can get the law passed most sneakily gets to put a heavy burden of truth on the other team, and then say that they're in the right.

Anyway, this generalization

> The lower–than–US fatality rates on the German Autobahn (where flow management is the primary safety strategy), and now Montana's experience, would indicate that using speed limits and speed enforcement as the cornerstone of US highway safety policy is a major mistake.

is totally out of line in an article that doesn't even address the technical question of statistical significance or deal with potential biases Montana's or Germany's automotive demographics put on the data.

I agree, the fact that the german Autobahn is compared directly to US highways is laughable. It's comparing apples and pears: not relevant.

Why can't you compare apples and pears? They're both fruit that grows in trees and they both have similar use culinary purposes. I can make cider out of both, and the results are actually quite similar.

Reply on news.ycombinator.com